Biomimetics doi: 10.3390/biomimetics11090662

Authors: İsmail Çalıkuşu

Electrode-level compressibility may not imply transferable hardware redundancy in biomimetic myoelectric interfaces. This study tested whether sensor-count sufficiency discovered by trial-level analysis survives participant-grouped evaluation. Dataset A comprised 398 archived Myo Armband trials from eight gestures. Dataset B contained 864 one-second trials from 36 participants and six gestures. A timestamp audit identified extensive repeated channel values; Dataset B was therefore analyzed on a conservative 100 Hz grid with 20–45 Hz filtering. Sensor subsets and RBF-SVM parameters were selected exclusively within grouped training data using repeated nested validation. Electrode-level NMF, all 28 fixed six-sensor layouts, cyclic re-indexing, channel-block ablation, participant-cluster bootstrap, PCA, and time-domain-only sensitivity analyses were evaluated. Dataset A yielded 97.74% accuracy with six sensors and 97.93% with eight. In Dataset B, accuracy was 75.96% ± 7.47% with six sensors and 77.93% ± 6.49% with eight; the paired difference was −1.97 percentage points (corrected 95% CI, −5.50 to 1.56). The participant-cluster bootstrap interval was −3.70 to −0.31 points. Active-gesture accuracy was 71.67% and 74.35%, respectively. All fixed six-sensor layouts averaged 74.59%. Three NMF components reconstructed 89.95% ± 1.78% of held-out-participant normalized RMS patterns, with no nonconverged folds. One-position cyclic re-indexing reduced accuracy to 40.28%; channel-block ablation caused losses of 0.62–6.71 points. Low-dimensional electrode-level RMS structure did not establish removable sensors across unseen users. Compact biomimetic interfaces require registration, adaptation, or equivariant processing before physical sensor reduction.

Biomimetics doi: 10.3390/biomimetics11090661

Authors: Zhengzhi Yang Yao Yao Qianqian Wang

No standardized workflow currently exists for translating leaf venation—a natural conduction network—into functional surface texture for product design. Taking the basal actinodromous venation of Cinnamomum camphora (camphor tree) leaves as its object of study, this paper proposes a parametric workflow spanning standardized specimen collection, high-resolution imaging, manual layered vector tracing, and a three-stage geometric simplification algorithm. Simplification performance is quantified through three metrics: curve simplification ratio (CSR), topology retention index (TRI), and area-coverage ratio (ACR). To assess how these simplified textures perform in practice, textures from all three stages were fabricated as silicone grip sleeves and evaluated through wet friction testing and tactile direction-recognition experiments. The three stages performed differently. Stage 1 yielded the highest wet friction coefficient. Stage 2 offered the best overall balance among friction enhancement, direction recognizability, and manufacturing feasibility. Stage 3, with the simplest geometry, proved easiest to fabricate. Differences in CSR and ACR across stages were statistically significant (p < 0.001), while TRI is reported separately as a descriptive reference measure. Together, the workflow offers a complete procedural framework and a quantitative basis for designing functional textures in biomimetic grip interfaces.

Biomimetics doi: 10.3390/biomimetics11090660

Authors: Changlong Pang Yukun Wang Wansheng Cheng

This article proposes an improved Moss Growth Optimization (IMGO) algorithm to address the drawbacks of imbalanced exploration and exploitation and susceptibility to local optima in the original MGO. IMGO integrates three-dimensional guidance, elite guidance, adaptive search, and adaptive adversarial learning to achieve a dynamic balance between global exploration and local development. IMGO is compared with eight metaheuristic algorithms on CEC2017 and CEC2022 benchmarks. Results show IMGO ranks first in 30D and 50D CEC2017 with scores of 49 and 64, surpassing second-ranked CFDA (87 and 81). For 20D CEC2022, IMGO takes first place with a score of 27, while original MGO scores 66 and ranks seventh. The Friedman test validates its strong robustness with a statistic of 1.6897, lower than CFDA’s 3.0000. Furthermore, IMGO acquires the optimal average solutions for all six engineering design problems. Experiments verify that IMGO has comprehensive advantages in accuracy and robustness, providing a reliable method for complex optimization problems.

Biomimetics doi: 10.3390/biomimetics11090658

Authors: Jia Yang Chunyang Zhang Ning Li Jie Wen Wenguang Yang Wenyuan Chen

While soft hand exoskeleton robots have approached human-level dexterity in terms of degrees of freedom, precise control methods for fine motor movements remain a significant challenge. Surface electromyography (sEMG) is widely employed in gesture recognition to enable patients to independently control a soft hand exoskeleton. However, individual finger control remains challenging through sEMG-based control due to the complexity of decoupling synergistic muscle activities. In this study, we propose a muscle fiber-based ultrasound perception strategy for fine hand motion recognition and soft hand exoskeleton control. Ultrasound imaging enables non-invasive visualization of forearm muscle morphology and provides information associated with underlying muscle-fiber activity. By reconstructing muscle morphology from ultrasound images, biologically relevant muscle-fiber features are extracted and fused to characterize fine hand movements. A lightweight Random Forest classifier is subsequently employed to map these biologically informed features to discrete hand actions, providing a computationally efficient recognition module for real-time control. To the best of our knowledge, publicly available ultrasound image datasets specifically designed for fine hand gesture recognition in rehabilitation applications remain limited. In the experiments, a dataset containing 21 hand gestures based on muscle ultrasound images was constructed to evaluate the proposed method. All data were collected from healthy participants as a preliminary proof-of-concept investigation. The results show that the proposed approach achieves an average recognition accuracy of 95.24% across three subjects in finger motion recognition. This preliminary study demonstrates the potential of machine learning-based ultrasound perception for improving fine hand gesture recognition and providing an intuitive control interface for soft hand exoskeletons, thereby enhancing their applicability in hand rehabilitation scenarios.

Biomimetics doi: 10.3390/biomimetics11090657

Authors: Mehmet Bilban Onur İnan

Bio-inspired integrity monitors for automated vehicles are reported as single pooled detection figures, which conflates algorithmic performance with evaluation protocol. Transferring the reafference principle to inertial-channel integrity, we identify what actually governs the reported figure. A single-track forward model is identified from the data, and its residual is standardized by Operating-Condition Residual Normalization (OCRN), a bin-wise operator conditioned on an observable operating point; the design is resolved by an exhaustive constrained search and each component is isolated using ablation. Across 224,638 samples spanning five towns and four friction levels, the residual scale varies by a factor of 158, and OCRN recovers 21.8 F1 points, more than the detection rule, the encoder, and the biological attenuation gate combined. A cross-comparison confirms this: changing the detection rule moves the result by 0.09 points, while changing the normalization moves this by 15 to 19. The monitor attains 97.34% precision, 76.35% recall, and 85.57% F1 at a 0.76% false-alarm rate when counting every injection, and 97.30/93.99/95.62% above a 3σ detectability floor when covering 80.1% of them. The floor scales with forward-model error, which is correlated with but not determined by the road friction, and the gate, the most explicitly biological element, is not selected once the residual is conditionally normalized, with its best setting gaining at most one F1 point at nearly twice the false-alarm rate.

Biomimetics doi: 10.3390/biomimetics11090659

Authors: Francesca Pescosolido Silvia Vesco Felicia Carotenuto Andrea Ciammaruconi Florigio Lista Roberto Bei Paolo Di Nardo Federica Trovalusci

Antibacterial coatings are essential for preventing infections and maintaining adequate hygiene standards in high-density and high-risk environments, such as public spaces, public transportation systems, schools, and healthcare facilities. In this context, it is essential that such coatings are easy to apply and compatible with large-scale production processes. Among the various strategies for developing antimicrobial surfaces, the incorporation of antibacterial agents into paints represents a particularly practical and versatile approach. In this work, antimicrobial coatings were developed by incorporating curcumin, a naturally derived antibacterial compound, into a commercially available water-based paint. For comparison, coatings were also developed by incorporating a commercially available silver-ion-based additive, a well-established antibacterial agent, into the same paint. Antimicrobial dispersions were deposited onto polycarbonate (PC) and polymethyl methacrylate (PMMA) substrates using a spray coating technique. The produced coatings were evaluated in terms of adhesion and hardness according to the ASTM D3359 and ASTM D3363 standards, respectively, achieving ratings of 4B for adhesion and 4B/5B for hardness. Moreover, after five days of continuous water immersion, no visible signs of cracking, delamination, or discoloration were observed. The resulting curcumin- and silver-based coatings, under the tested surface-contact assay conditions, markedly reduced viable bacterial recovery against both S. aureus and E. coli, with no colonies recovered at the dilution range used for comparison with the Paint-Blank control. These results support the potential of curcumin as a naturally derived antibacterial additive for the development of water-based antibacterial coatings and provide a basis for further investigation of their long-term stability, durability, and practical applicability.

Biomimetics doi: 10.3390/biomimetics11090656

Authors: Teoman Karadag Emre Gozkaya Arif Basgumus Mustafa Namdar

48 V bus modules must stay regulated under wide input-voltage swings and fast load transients. A proportional-integral (PI) controller and a fuzzy logic controller (FLC) are tuned by the same Particle Swarm Optimization (PSO), a swarm metaheuristic modeled on bird-flock foraging, under an Integral of Time-weighted Absolute Error (ITAE) objective. The novelty is PSO-based membership-boundary optimization: the membership-function limits, analogues of biological decision thresholds, are shaped directly rather than only the scaling gains. A 153 W, 225 kHz synchronous boost converter is simulated with non-ideal parasitics. For input-voltage transitions the PSO-FLC shortens settling time by up to 70.5% (118 versus 400 ms), whereas the PSO-PI gives smaller peak deviation (26.2% versus 40.0%) and faster load-step recovery. Unconstrained, both controllers draw 58–62 A on the worst input-voltage step; a 10 A cycle-by-cycle limit removes these excursions. Under an identical budget and parameter count, boundary tuning attains 12.3% lower ITAE cost and roughly 40% shorter large-signal settling than scaling-gain tuning. Across ten runs the cost varies by under 1.5% of its mean, and the ranking holds at a realizable update rate (4.44 μs, zero-order hold) and under the current limit. The findings are simulation-only comparative design guidance.

Biomimetics doi: 10.3390/biomimetics11090655

Authors: Huachen Su Yuying Yan

Droplet-based electricity generators (DEGs) are usually characterised under dry conditions, although retained water can alter their electrical boundary. We compared blank, bottom-wetted, top-wetted, wetted-and-dried, and partially immersed states using 40 mm × 40 mm FEP/ITO devices. Four separately assembled devices per condition were tested on a separate apparatus, with 100 droplet events per device. The device averages were analysed by Welch’s ANOVA and Holm-adjusted contrasts. Bottom wetting changed the average peak voltage by −0.30% relative to the blank (adjusted p = 0.960), whereas top wetting reduced it by 56.21% (adjusted p = 2.62 × 10−9). The dried-state average was within 0.38% of the blank (adjusted p = 0.960), indicating electrical recovery under the stated protocol without establishing equivalence or complete drying. Immersing 10%, 30%, and 50% of the device length reduced the average by 19.31%, 34.59%, and 47.73%, respectively. The common-state averages agreed within 1.49% with an earlier apparatus series. A process-matched MWCNT:FEP surface reached an apparent contact angle of 151° and a 6° sliding angle while retaining 87.3% of the FEP-only average. These results identify the exposed top surface, rather than the bottom-interface water, as the principal wetting-state sensitivity under the tested conditions and support a lotus-inspired low-adhesion design target.

Biomimetics doi: 10.3390/biomimetics11090654

Authors: Diana V. Portan Panagiotis Zoumpoulakis Vassilis Kostopoulos Ioanna Pitterou Konstantinos Tsiantas Efstathios Michalopoulos Leonard Azamfirei

Background/Objectives: Biomimetic biomaterials should be evaluated not only through their initial properties but also through their interaction with biological environments over time. This study investigated a human cell-based approach integrating flavonoid-induced cellular responses with the temporal evolution of 3D-printed polylactic acid (PLA) scaffolds under physiological-like conditions. Methods: Human Wharton’s Jelly mesenchymal stem cells (WJ-MSCs) were exposed to naringin and hesperidin (250 µg/mL) and evaluated after 3, 7, and 10 days using ALP, TP, OPN, and OC. In parallel, PLA scaffolds’ biodegradation was studied under static and dynamic conditions. Results: Naringin produced a more pronounced early/intermediate osteogenic-related response, whereas hesperidin showed comparatively more sustained ALP- and OPN-related activity at later stages. Dynamic scaffolds exhibited progressive temporal changes, with weight variations ranging from −0.384% to +0.127% at days 1, 3, 7, and 10, respectively, accompanied by progressive morphological modification. Initial AFM analysis showed an average RMS roughness of 22.1 nm. Conclusions: The distinct temporal profiles of the flavonoids, together with the evolving scaffold–medium interface, support a biomimetic strategy based on temporal coordination of biochemical cues. These findings provide a rationale for future experimentally validated sequential delivery systems designed to promote early osteogenic activation followed by sustained cellular and matrix-associated activity.

Biomimetics doi: 10.3390/biomimetics11090653

Authors: Muhammad Hussain Fahman Saeed Sultan Aldera

Deepfake videos currently facilitate extensive financial deception, political misinformation, and unauthorized imagery, with anticipated U.S. losses from deepfake-related fraud surpassing $40 billion by 2027; human evaluators accurately recognize high-quality forgeries merely 25% of the time, highlighting the pressing necessity for automated, widely applicable detection mechanisms. Adaptive-depth architectures offer an intriguing alternative to fixed-depth deepfake detectors when the optimal model capacity is indeterminate in advance. This study presents the Adaptive Entropy-Guided ICA State-Space Model Forgery Detector (AEGIS-FD), a deepfake detection framework at the video level that progressively increases its depth from one to eight layers via an entropy-driven growth mechanism, attaining peak validation performance at a depth of six. The design incorporates a three-dimensional spatiotemporal stem, Sinkhorn-normalized manifold-constrained hyper-coupling (mHC) layers for balanced temporal integration, a selected state-space temporal block for sequence depiction, and FastICA-based initialization for newly introduced layers. Evaluated using Celeb-DF v2, AEGIS-FD achieves a test AUC of 0.9600 and a validation AUC of 0.9607, above the performance of a single-layer Mamba SSM baseline (AUC = 0.8355). In comparison to a fixed-depth-6 baseline, the model demonstrates consistent improvements over five random seeds (96.12 ± 0.28 vs. 94.84 ± 0.44 AUC; p = 0.0022), suggesting that adaptive development provides advantages that exceed mere depth. In a zero-shot cross-dataset evaluation—trained on Celeb-DF v2 and assessed without fine-tuning on a FaceForensics++ (FF++) C23 subset comprising 1000 original and 1000 FaceSwap videos—AEGIS-FD achieves an AUC of 89.2 compared to 88.4 for the corresponding baseline (+0.8 AUC), providing initial proof of cross-dataset transferability. These findings suggest that adaptive-depth growth presents a viable approach for detecting deepfakes at the video level, while further validation across various datasets and modification techniques is essential.

Biomimetics doi: 10.3390/biomimetics11090652

Authors: Maria Ida Rizzo Maria Emiliana Caristo Chiara Ribaldone Simone Faustino Maria Marino Giorgio Spuntarelli Anna Chiara Contini Luigi Dall’Oglio Luigi Tomao Mattia Algeri Stefano Tedesco Gianantonio Pozzato Cristiano De Stefanis Antonello Cardoni Lorenzo Lupoi Camilla Codazzi Lucia Leone Mario Zama Massimiliano Raponi

Cleft palate is a congenital anomaly that causes functional and esthetic challenges, and the hard palate is essential for feeding, speech, and separation of the oral and nasal cavities. Yet, current reconstructive techniques do not restore its bony component. Building on previous in vitro work showing that decellularized palatal mucoperiosteum, microperforated with Quantum Molecular Resonance (QMR®) technology and recellularized with mesenchymal stem cells, preserves the collagen microenvironment, supports engraftment, and shows osteoinductive potential, this study evaluated the early feasibility and regenerative potential of bioengineered mucoperiosteal scaffolds (BEMS) in Landrace pigs model. Bone marrow was collected from recipient pigs to isolate pBM-MSCs. Donor palatal mucoperiosteum was decellularized, microperforated, and recellularized with these cells to generate BEMS. After surgical creation of a cleft palate, four pigs received BEMS, and two controls underwent standard palatoplasty. At one month, scaffold-treated animals showed early mucosal and osseous regeneration, including neo-epithelium, connective tissue, and new bone formation, without clinical or routine histological signs of acute rejection. SPARC (Secreted protein acidic and rich in cysteine) expression supported osteogenic activity. Regenerated palates were stable and fracture-resistant, whereas controls showed incomplete repair and fractures. These findings suggest that BEMS may address limitations of conventional palatal reconstruction and support further investigation for human palatal bone regeneration.

Biomimetics doi: 10.3390/biomimetics11090651

Authors: Xu Xia Ningfang Song Tianze Wang Jian Guo Jingchao Ban Zhenpeng Wang

In Global Navigation Satellite System (GNSS)-denied deep-sea environments, traditional scalar gravity matching navigation methods frequently suffer severe performance degradation in weak-feature, highly repetitive gravity anomaly regions. Inspired by the hippocampal spatial memory mechanism and natural graded foraging behavior of benthic marine organisms, this paper proposes a full-chain bionic framework named the Physics-Consistent Multi-Scale Adaptive Particle Filter for Gravity Matching Navigation (PC-MAPF-GM). This method endows the particle filter with four layers of biologically mimicked autonomous regulation capabilities: quantitative gravity field local suitability assessment, dynamically adjusted time-varying search scope, three-level multi-scale stepwise matching, and along-track trajectory motion physics consistency constraint. The verification of long-term shipborne lake experiments confirms that the proposed method reduces the final gravity matching positioning root mean square error (RMSE) to only 528.2 m, which is more than 41% lower than the classical terrain contour matching (TERCOM) benchmark and 31% lower than iterative closest contour point (ICCP). This biomimetic full-design-chain solution provides a robust new practical navigation paradigm for long-endurance fully autonomous underwater vehicles operating without any external auxiliary positioning information.

Biomimetics doi: 10.3390/biomimetics11090650

Authors: Juan Yu Ping Li Xi Hu

This simulation-based exploratory study examines HGPC-CSA-PSO, a hybrid biomimetic optimization framework integrating the Giza Pyramids Construction Algorithm (GPC), Crow Search Algorithm (CSA), and Particle Swarm Optimization (PSO) for dynamic collaborative grouping in interpreting education. The educational scenario and initialization values are modeling assumptions; no classroom intervention, causal teaching experiment, or independently auditable empirical validation is reported. The scalar proficiency index Qi is used as an education-oriented evaluation-layer measure, whereas the checked-in algorithms optimize their original unweighted coordinate-mean fitness. An illustrative educational-model trajectory reaches its stated scalar threshold after 29 simulated interaction updates. Separately, under the standardized repository configuration, 30 paired runs use the implementation-level termination rule that every coordinate of every simulated student must reach 0.99. Under that configuration, HGPC-CSA-PSO converges in 16.367 ± 1.974 updates and is faster than GPC and CSA (Holm-adjusted p < 0.001 for both), but not significantly different from PSO (Holm-adjusted p = 0.194). These two result layers are not interchangeable and provide model-level computational evidence only.

Biomimetics doi: 10.3390/biomimetics11090649

Authors: Xinwei Li Irshad Khalil Faisal Rahman Muhammad Nawaz Khan

Images from high-resolution cameras are mapped onto a sparse pattern of low spatial resolution and intensity in the retina, which limits visual perception in retinal prosthetic vision. When the entire scene is converted into phosphenes, it may allow unnecessary background information to be retained and may cause visual clutter, which may make it hard for prosthetic vision users to interpret the scene. In order to tackle this issue, this paper presents a context-, depth-, and user-preference-aware method for selecting the objects of interest in the generation of phosphene images. The proposed method does not show all the objects equally but learns to sort the objects according to their relevance to prosthetic vision. Manual annotation of a subset of COCO images was conducted where the most salient object was selected based on environment type, scene type, user mode, safety, navigation relevance, task importance, and distance. All of the candidate objects are described by full-scene visual features, object-crop features, handcrafted priority features, context embeddings, and monocular depth features. To predict object-level importance scores and identify the Top-1 and Top-4 important objects in unseen scenes, a hybrid deep learning model combining twin ResNet-18 backbones for scene and object feature extraction with embedding-based context encoding was trained. Priority maps and phosphene images were then created using the selected object masks and were depth-weighted. Two types of phosphene representations were also produced: Canny-edge-based and direct full images. The proposed framework is designed to suppress irrelevant background areas and improve important and closer objects in order to obtain a simplified and informative prosthetic-vision representation of the scene. The experimental evaluation, including Top-1 accuracy, Top-3 accuracy, mean reciprocal rank (MRR), and visual comparison, demonstrates the effectiveness of the proposed framework, achieving a Top-1 accuracy of 90.12%, a Top-3 accuracy of 97.45%, and an MRR of 0.9368. Furthermore, the proposed Canny-priority phosphene representation achieved an average human-participant recognition accuracy of approximately 86%. The proposed method offers a user-adaptive strategy for selecting and visualizing the information of a scene under the severe constraint of the bandwidth of retinal prosthetic vision.

Biomimetics doi: 10.3390/biomimetics11090648

Authors: Haktan Altinova Dorothee Hodde José L. Gerardo-Nava Axel Dievernich Lmar Arman Melissa Büchler Andreas Kriebel Jörg Mey Hans Clusmann Joachim Weis Gary A. Brook Pascal Achenbach

The development of bioengineered conduits for the repair of large peripheral nerve defects remains challenging. Here, we introduce a biomimetic fibrin hydrogel scaffold containing stacked arrays of aligned poly(ε-caprolactone) (PCL) nanofibers that mimics the naturally forming fibrin cable seen in transection injuries. We further evaluated the additional encapsulation of Schwann cells (SCs) into the fibrin hydrogel. Nerve regeneration was examined in a 15 mm rat sciatic nerve resection model over a period of 12 weeks, comparing our scaffolds against the autograft, a collagen hollow tube, and a lesion-only control group. Functional and morphometric analyses revealed that the autograft supported the strongest regeneration, followed by the SC-seeded fibrin-nanofiber scaffold and the hollow tube. Unexpectedly, a macrophage-rich core devoid of SCs and regenerated axons formed within both fibrin-nanofiber scaffolds around persisting PCL nanofibers. This core impaired the regenerative potential of the non-SC-seeded fibrin-nanofiber scaffold to the extent that functional regeneration was comparable to that of the lesion-only control group. While the scaffolds were designed to closely mimic and support the natural regenerative environment following nerve transection, our results reveal that these theoretical considerations do not necessarily translate into practice.

Biomimetics doi: 10.3390/biomimetics11090646

Authors: Liancheng Zheng Mohammad Soleimani Amiri Rizauddin Ramli Sharifah Sakinah Syed Ahmad Xiaotian Ma

Precise end-effector positioning of a robotic arm is critical for accurate performance in robotics, which directly impacts the system’s performance in applications requiring high precision. This paper presents an optimized tag-based inverse kinematics approach for a 6 DoF robotic arm using a Modified Biogeography-Based Optimization (MBBO) algorithm, which is an enhanced version of the original Biogeography-Based Optimization (BBO), a population-based evolutionary algorithm inspired by the natural distribution of species across habitats. The target position is identified via AprilTag visual fiducial markers and integrated into the inverse kinematics solver. Performance was evaluated against Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and BBO through 3-dimensional simulations involving ten target points. Results show that MBBO achieves reduced positioning error compared to GA, PSO, and BBO, resulting in lower end-effector position errors. The findings highlight the effectiveness of combining visual tag detection with advanced optimization for precise robotic arm control.

Biomimetics doi: 10.3390/biomimetics11090647

Authors: Ki Ha Min Yi-Rang Jeong Jong Won Mun Kyu Ho Jeon Seung Pil Pack

This review systematically elucidates the rapidly evolving field of biomimetic coacervate coatings, bridging the fundamental thermodynamic principles of liquid–liquid phase separation (LLPS) with advanced biomedical translations. While conventional surface modifications for medical implants frequently fail to maintain structural and functional integrity within dynamic, wet physiological environments, biomimetic coacervation inspired by natural underwater adhesive mechanisms offers a highly versatile, conformable, and robust interfacial strategy. Here, we analyze the critical physicochemical driving forces governing coacervate formation, emphasizing the synergistic interplay of electrostatic, hydrophobic, hydrogen-bonding, and cation–π interactions. We comprehensively discuss diverse macromolecular design principles utilizing marine-derived biopolymers, synthetic or recombinant polypeptides, and hybrid organic–inorganic condensates, alongside key architectural orchestration methodologies including direct deposition, in situ triggerable coacervation, and layer-by-layer (LbL) assembly. Furthermore, we evaluate multi-functional clinical translations, highlighting breakthroughs in wet tissue sealing, bone repair, localized stimuli-responsive drug or nucleic acid delivery, anti-biofouling medical device coatings, and regenerative cell–material interfaces. Ultimately, this review underscores the profound potential of biomimetic coacervates as a cornerstone platform for next-generation multifunctional medical devices and personalized regenerative medicine.

Biomimetics doi: 10.3390/biomimetics11090645

Authors: Broderick Crawford Hugo Caballero Gino Astorga Felipe Cisternas-Caneo Alan Baeza Pablo Puga Lucero Giovanni Giachetti Ricardo Soto

Solving problems associated with the efficient distribution and organization of resources has generated increasing interest in the scientific community. One of the most commonly used approaches consists of approximate solution techniques, which have been able to solve complex covering problems within acceptable computational time and cost. One of the benchmarks used to evaluate these approaches is the Set Covering Problem, which is an NP-hard combinatorial optimization problem. Among the techniques that have been investigated, metaheuristics play an important role. These methods are commonly developed for continuous search spaces and, in order to be applied to covering problems, must be modified to operate in discrete domains. This modification presents an important challenge: finding an appropriate transformation method that translates continuous solutions into binary solutions. This issue has been addressed through two main strategies: binarization using two-step schemes, and, in our proposal, the use of repair operators orchestrated according to their performance through an Adaptive Repair Selection Mechanism based on the multi-armed bandit framework. To evaluate our proposal, we selected the Binary Hunger Games Search metaheuristic because the relative quality of each individual determines its hunger level, which in turn regulates the movement of the population and the influence of the best solution found. Infeasible solutions are handled through a set of Tabu Search-based repair operators. Instead of applying a single repair rule throughout the entire execution, the proposed approach dynamically selects among these operators according to their observed contribution during the search. Each repair operator also incorporates Tabu memory to discourage repetitive decisions during feasibility restoration. The experiments were conducted using the classical Beasley benchmark instances for the Set Covering Problem.

Biomimetics doi: 10.3390/biomimetics11090644

Authors: Tian Mu Junyao Gao Weilong Zuo Leilei Xie

Humanoid robots are highly susceptible to structural damage during irrecoverable falls due to high landing velocity, short impact duration, and high peak impact force. Inspired by human protective strategies, namely instinctive postural adjustment and soft-tissue energy absorption, this paper proposes a passive–active cooperative fall-protection method that combines pre-impact motion regulation with post-impact structural energy absorption. On the passive protection side, high-risk contact regions are identified through multi-directional fall simulations, and a multi-region, multilayer protective suit is optimized considering impact energy absorption, peak-force reduction, anti-bottoming safety, added mass, and thickness constraints. On the active protection side, a variable height inverted pendulum (VHIP) model is used to optimize the center of pressure and center of mass trajectories, reducing the terminal impact energy before ground contact. The residual impact energy is then matched with the absorption capacity of the passive protective layers, forming a unified framework that integrates pre-impact motion unloading and post-impact energy absorption. Numerical validation is performed on a MATLAB–CoppeliaSim co-simulation platform, and physical experiments are conducted on the FCR humanoid robot (approx. 50 kg, 1.65 m, 22 DOF). Compared with the unprotected case, the proposed method reduces the peak equivalent impact force from 4819.1 N to 1038.2 N, i.e., a reduction of 78.5%, demonstrating its effectiveness in attenuating impact loads and enhancing protection capability.

Biomimetics doi: 10.3390/biomimetics11090643

Authors: Ionel Petrescu Valentina-Daniela Băjenaru Daniel-Mircea Popescu Viorel Vulturescu Liviu Marian Ungureanu

Low-cost geared DC motors are widely used in mobile robotics and embedded mechatronic systems; however, their control remains challenging due to dead-zone nonlinearities, friction, low encoder resolution, motor asymmetry, and supply voltage variations. Biological organisms routinely perform motor control in the presence of similar uncertainties by relying on approximate reasoning and adaptive responses rather than precise mathematical models. This paper develops and experimentally validates a practical bio-inspired control architecture for low-cost geared DC motors operating under severe sensing and actuator limitations. The proposed controller combines fuzzy inference, dead-zone compensation, and a ramp-start mechanism to emulate the gradual and adaptive nature of biological motor responses. Instead of relying on an accurate plant model, control actions are generated through linguistic rules that mimic human-like decision-making based on speed error and error variation. The controller is implemented on an Arduino-based differential-drive robotic platform equipped with low-resolution optical encoders. Experimental results demonstrate that the proposed bio-inspired approach effectively mitigates startup stall, reduces oscillatory behavior caused by measurement quantization, and maintains stable speed regulation despite actuator variability and battery voltage fluctuations. The study shows that biologically inspired fuzzy control provides a practical and computationally efficient alternative to conventional PID methods for low-cost robotic systems characterized by significant uncertainty and nonlinear behavior.

Biomimetics doi: 10.3390/biomimetics11090642

Authors: Lizhen Du Dahongnian Zhou Xiaoshuang Xiong Min Shen Hongtao Tang Lianqing Yu Fei Fan

Multi-objective optimization plays an important role in modern design and complex engineering applications. However, achieving an effective balance between the convergence and diversity of Pareto-optimal solutions remains challenging. This paper proposes a Sobol-driven Multi-objective Whale Migration Algorithm (SMOWMA), which extends the Whale Migration Algorithm within a non-dominated sorting and elite-selection framework. A maximin scrambled Sobol initialization scheme is first employed to improve the distribution of the initial population. An archive-guided adaptive Student-t flight mechanism is then incorporated into the leader-whale position update to dynamically balance global exploration and local exploitation. In addition, archive crowding information and archive-entry success feedback are jointly used to adjust the search behavior according to both environmental diversity and recent search performance. SMOWMA is evaluated on five widely used multi-objective benchmark suites, namely ZDT, DTLZ, WFG, UF, and CF, using four performance indicators: generational distance (GD), inverted generational distance (IGD), spacing (SP), and hypervolume (HV). The results, together with Friedman tests and Holm-adjusted Wilcoxon tests, demonstrate that SMOWMA achieves competitive overall performance in terms of convergence, diversity, and objective-space coverage, although its relative advantage remains problem-dependent. The practical applicability of SMOWMA is further examined using multi-objective welded-beam design formulations, a bi-objective four-bar truss design problem, and a five-objective car side-impact design problem. The engineering results show that SMOWMA can obtain competitive and stable approximation sets for constrained design problems with different numbers of objectives, supporting its effectiveness and applicability in multi-objective engineering optimization.

Biomimetics doi: 10.3390/biomimetics11090641

Authors: Seungyeon Lee Hyunseok Yang

Reinforcement learning-based quadruped locomotion policies can exhibit command-tracking errors under terrain variations and unmodeled dynamics. This study proposes an online bounded Gaussian Process-enhanced model predictive control framework, termed Gaussian Process–Model Predictive Control–Reinforcement Learning(GP-MPC-RL), for command-level supervision of a pretrained locomotion policy. A frozen PPO policy generates the low-level locomotion behavior, while an acados-based MPC supervisor adjusts the velocity command using a nominal command-response model. An online Gaussian Process learns the one-step residual between the nominal prediction and measured robot response, and its uncertainty-weighted forward-velocity correction is incorporated into the MPC prediction. The framework was evaluated in Isaac Lab using a Unitree Go2 quadruped robot model over 20 paired rough-terrain trials at target velocities of 0.3, 0.5, and 0.7 m/s; GP-MPC-RL reduced the mean forward-velocity root mean square error (RMSE) relative to PPO by 38.3%, 21.3%, and 10.1%, respectively. Under a 5 kg payload, GP-MPC-RL reduced velocity RMSE by 29.0% relative to MPC-RL and reduced the 0–5 kg payload-induced degradation by 53.0% (p = 0.019). The average supervisor computation time was 0.112 ms. These results indicate that GP residual adaptation is particularly effective when the nominal command-response model becomes inaccurate, improving robustness without retraining the underlying locomotion policy.

Biomimetics doi: 10.3390/biomimetics11090640

Authors: Jure Berce Iztok Golobič

The similarity of droplet impacts on nature-mimicking superhydrophobic surfaces is traditionally compared using the dimensionless Weber number. Yet, maintaining a constant We by decoupling droplet diameter and impact velocity influences secondary forces, challenging this assumption of similarity. In this work, we investigate water droplet impacts on a lotus-leaf-mimicking laser-textured superhydrophobic aluminum surface at two constant Weber number levels (25 and 50), varying droplet diameter from 2.1 to 4.15 mm. Our results confirm that maximum spreading depends on the Reynolds number at a fixed We, as smaller, faster droplets spread less due to increased relative viscous dissipation. We propose a modified empirical scaling model that describes our data with high accuracy and generalizes successfully to external datasets. Crucially, we demonstrate that the contact time of a droplet of a given size is not strictly velocity-independent, unveiling a Weber number-dependent inertia-capillary scaling. We show that this is driven by a shift in rebound dynamics, where the relative timescale of spreading increases over retraction for larger droplets. These findings demonstrate that We is insufficient to characterize droplet rebound across varying scales and that accounting for size-dependent deviations is critical for the precise design of technologies that leverage droplet-surface interactions.

Biomimetics doi: 10.3390/biomimetics11090639

Authors: Lei Nie Yingying Liang Yiran Lin Wei Guo

Rapid hemostasis, oxidative stress resistance, and minimally invasive administration are crucial performance requirements for high-performance wound covering. Inspired by the dynamic remodeling properties of the native extracellular matrix, we fabricated a multifunctional injectable hydrogel through dynamic Schiff-base crosslinking between oxidized hyaluronic acid (OHA) and carboxymethyl chitosan (CMCS), combined with magnesium ion (Mg2+) coordination. The effects of Mg2+ content on hydrogel properties were systematically investigated. The hydrogels gelled rapidly under physiological conditions and showed good injectability, self-healing behavior, and favorable adhesion to moist tissues. Notably, Mg2+ incorporation significantly enhanced hemostatic performance in a mouse tail amputation model, reducing blood loss from 391.7 mg to approximately 75 mg and shortening hemostasis time from 151.7 s to 50.3 s. The 2, 2′-azinobis (3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) radical scavenging efficiency reached approximately 80%, and the hydrogel effectively scavenged intracellular reactive oxygen species (ROS) without compromising cytocompatibility or fibroblast activity. This study presents a biomimetic and easily prepared hydrogel platform that integrates pro-coagulant activity, redox regulation, and on-demand injectability, showing translational potential as bioactive wound covering for bleeding control and oxidative microenvironment regulation.

Biomimetics doi: 10.3390/biomimetics11090638

Authors: Ahmet Bahadır Karlı Murat Ucan Buket Kaya

Stroke is one of the leading causes of mortality and long-term neurological disability worldwide, and early diagnosis through accurate disease classification directly affects treatment success. Rapid differentiation of hemorrhagic and ischemic stroke on computed tomography (CT) images, together with accurate determination of the acute and chronic phase in ischemic cases, is of critical importance in the clinical decision-making process. In this study, StrokeCT-2C5K a two-center dataset comprising 5000 cranial CT images, was assembled specifically for this work. The images were reviewed by radiology specialists and assigned to one of four diagnostic categories: normal, hemorrhagic stroke, acute ischemic stroke, or chronic ischemic stroke. A Squeeze-and-Excitation (SE-Attention) mechanism was then integrated into DenseNet-121, ResNet-50, and EfficientNet-B3. From a biomimetic perspective, this channel-recalibration process provides a functional analogy to biological selective attention by giving greater weight to informative responses while reducing the influence of less relevant ones. All models were trained under the same training, validation, and test protocol; the standard CNN architectures were compared with their SE-Attention-enhanced counterparts. The results showed that the SE-Attention mechanism enables more effective learning of lesion-specific discriminative features by adaptively recalibrating channel-wise information, yielding an average classification accuracy improvement of 0.93 percentage points across all three architectures. The most pronounced improvements were observed in distinguishing ischemic from hemorrhagic stroke, as well as in distinguishing acute from chronic ischemic stroke. These findings show that a selective-information-processing strategy functionally analogous to biological attention can improve multi-class stroke classification across different CNN backbones.

Biomimetics doi: 10.3390/biomimetics11090637

Authors: Yanwei Xie Jiawen He Yi Zhang

To address contact force estimation and compliant control for biomimetic robotic arms interacting with uncertain environments, an adaptive variable-damping impedance control method based on a fuzzy-controlled forgetting-factor strong tracking Kalman filter (FSKF) is proposed. The proposed method improves the conventional strong tracking Kalman filter (SKF) by introducing a fuzzy control strategy to adaptively adjust the forgetting factor, thereby enhancing the filtering performance and improving the accuracy of contact force estimation. The estimated contact force is subsequently incorporated into an adaptive variable-damping impedance controller to achieve simultaneous contact force estimation and compliant control of the biomimetic robotic arm. During biomimetic robotic arm motion, the proposed controller utilizes the estimated contact force to adaptively regulate the damping coefficient, compensating for force-tracking errors caused by environmental uncertainties and thereby improving both force and position tracking performance. The simulation and experimental results demonstrate that the proposed adaptive variable-damping impedance controller has better force and position tracking accuracy compared with the conventional impedance controller. Compared with traditional methods, the estimation accuracy based on FSKF has improved by about 7.3%. These results demonstrate the potential of the proposed method for prosthetic systems and other applications involving compliant robot–environment interaction.

Biomimetics doi: 10.3390/biomimetics11090636

Authors: Luz M. Tobar-Subía-Contento Juan A. Cabrera Anthony Mandow Jesús M. Gómez-de-Gabriel

On-body robots that travel around a human limb must keep a firm enough grip to avoid slipping or detaching, while never pressing hard enough to hurt—a balance that is hardest to strike precisely when the robot is orbiting the limb and gravity continually redistributes the contact loads. This paper presents an open, non-anthropomorphic robot that wraps around a compliant cylindrical surface with a three-contact grasp: a central traction module with two in-line driven wheels, and two lateral spring-loaded arms with distal wheels. Its central contribution is an actuation-space decomposition in which the two lateral wheel torques, expressed in a common-mode/differential basis, simultaneously drive the orbital motion and regulate the central normal force. We show that this basis diagonalises both the rolling kinematics and the static force balance, so the differential (grip-regulating) channel is provably orthogonal to the common-mode (propulsion) channel: a single pair of actuators perform both tasks without mutual interference and without a dedicated force mechanism. A model-based feedforward law derived from the static contact model, corrected by a PI term fed back from the compliant arms—which double as the force sensor—keeps the central force within a safe band; in a full-revolution simulation the differential command reverses sign to counteract the gravitational load swing while leaving the orbit undisturbed. The same compliant arms yield a closed-form estimate of the cylinder radius and contact geometry, accurate to below one millimetre across a 45–87 mm diameter range, from proprioception alone. Preliminary prototype tests reproduce the predicted behaviour, supporting the approach for future wearable and assistive applications.

Biomimetics doi: 10.3390/biomimetics11090635

Authors: Rebecca Pellegrino Maria Rosa Iaquinta Annalia Masi Mauro Pollini Federica Paladini

The management of chronic wounds represents one of the major challenges in regenerative medicine, as the healing process can be compromised by infections, oxidative stress, and persistent inflammation. In this context, wound pH serves as an important biomarker of tissue status, highlighting the need for smart dressings capable of promoting regeneration while simultaneously monitoring the wound microenvironment. In this study, biomimetic silk fibroin films functionalized with curcuminoids extracted from Curcuma longa were developed and characterized through spectroscopic, swelling/degradation, antioxidant, colorimetric, and biological assays, with the aim of obtaining a multifunctional dressing with regenerative properties and pH responsiveness. The results showed that curcuminoids were physically incorporated into the protein matrix without altering its chemical structure. The films exhibited a high absorption ability and antioxidant activity in the initial stages, and a clear and reversible color change in response to pH. Biological assays on 3T3 fibroblasts further confirmed the high cytocompatibility of the materials and their ability to support cell migration and wound closure in vitro. The developed films represent a promising biomimetic platform for advanced wound dressings, capable of combining support for tissue regeneration, antioxidant protection, and visual monitoring of wound status through the detection of pH changes.

Biomimetics doi: 10.3390/biomimetics11090634

Authors: Zhaoyong Fan Xi Li Zhenhua Xiao Lianying Zou Wangming Zhang

Metaheuristic algorithms are widely used to solve complex optimization problems, but the trade-off between exploration and exploitation often limits their performance. The Crested Porcupine Optimizer (CPO) employs four bio-inspired defense mechanisms and achieves competitive performance, but it still tends to converge prematurely, initialize populations poorly, and rely on static parameters that cannot adapt to different phases. This paper proposes a Multi-Strategy Enhanced Crested Porcupine Optimizer (MSCPO) with four phase-targeted enhancement strategies: (1) Kent Chaos Opposition-Based Learning Initialization (KCOL) improves the initial population distribution through chaotic-weighted reflection; (2) Arctic Puffin Optimization (APO)-Inspired Dual-Modal Evasion (APO-DME) reduces dependence on the global best solution and increases population diversity through dual-mode differential perturbation; (3) Adaptive Dual-Differential Perturbation (ADDP) refines the search by combining population diversity and elite guidance information; (4) Periodic Dynamic Adaptive Perturbation (PDAP) enhances exploitation through periodic trigonometric perturbation and a fitness-conditioned update rule. These strategies interact across the optimization process to strengthen each defense mechanism at the appropriate phase. On the CEC 2017 and CEC 2022 benchmark suites, MSCPO achieves the best overall mean rank among all compared algorithms, with an overall rank of 2.638 on CEC 2017 and 2.375 on CEC 2022. A full-factorial ablation over all 16 strategy combinations confirms that each strategy contributes positively: removing any single strategy degrades the overall mean rank, and the complete MSCPO achieves the best mean rank (3.34), significantly outperforming all single-strategy variants (Wilcoxon signed-rank test, p < 0.001). To verify the practical applicability of MSCPO, the algorithm is further applied to three engineering design problems: step-cone pulley design, hydrostatic thrust bearing design, and robotic gripper design. MSCPO ranks first on the hydrostatic thrust bearing problem, second on the step-cone pulley problem, and third on the robotic gripper problem. Future work will explore adaptive population sizing to further improve the scalability of MSCPO on very high-dimensional problems.

Biomimetics doi: 10.3390/biomimetics11090633

Authors: Zhiquan Wei Xinlan Hu Xinran Hu Yaozhe Yu

The inherent trade-off between strength and toughness in structural materials remains a critical challenge. Inspired by the hierarchical architecture of fish scales, this study proposes a novel overlapping helical composite structure. Multi-material three dimensional (3D) printing technology was employed to fabricate single-edge notched bending specimens. Quasi-static three-point bending experiment was conducted to investigate the mechanical performance of a fish scale-inspired structure. The results show that compared to the stiff bulk structure, the bio-inspired design exhibits a 60.4% enhancement in apparent fracture toughness and a 157.5% increase in energy absorption despite a reduction in flexural modulus and strength. The significant improvement may be attributed to the synergistic effects of crack deflection, which transform the fracture mode from catastrophic brittle failure to progressive damage with a stable post-peak deformation stage. Furthermore, parametric studies reveal that both the linear helical angle and its nonlinear gradient distribution critically govern the toughening efficiency. An optimal linear angle of 19° provides the best overall performance, while a nonlinear gradient (e = 1.75) further shifts energy dissipation towards the post-peak deformation stage, achieving a higher toughening efficiency. This work establishes a fundamental understanding of an overlapping helical coupling toughening strategy and provides a promising design route for high-damage-tolerance composite structures.

Biomimetics doi: 10.3390/biomimetics11090632

Authors: Jeong Yun Lee Hyun Ho Shin Seongyeon Jo Da Han Hyun Ji Hyun Ryu

Postsurgical adhesions are common complications of abdominal and pelvic surgeries, often leading to bowel obstruction, chronic pain, and increased risks during reoperation. Although various physical barrier materials have been developed to prevent postsurgical adhesions, retaining anti-adhesive materials at surgical sites is challenging. In this study, we developed thermosensitive injectable gallic acid-conjugated hyaluronic acid (HA-GA) and Pluronic F127 (PluF) composite hydrogels to retain anti-adhesive materials and prevent postsurgical adhesion. The incorporation of HA-GA reduced the critical gelation concentration of PluF from approximately 18 to 12 wt% and decreased the gelation temperature from 28.5 °C for PluF to 25.6 °C for HA-GA (2 wt%)/PluF hydrogels with the left-shift in sol–gel curves. The G′ values increased from 10.1 ± 1.7 kPa for PluF to 24.8 ± 3.6 kPa for HA-GA (2 wt%)/PluF hydrogels at 37 °C. In addition, the HA-GA/PluF hydrogels exhibited enhanced mass retention as a function of time compared to PluF hydrogels alone. HA-GA (2 wt%)/PluF hydrogels retained 33.4 ± 3.5% of their initial mass after 7 d, whereas PluF was completely eroded within 3 d. The anti-adhesion efficacy of the HA-GA/PluF hydrogels was evaluated using a rat cecum abrasion model. Notably, the HA-GA (2 wt%)/PluF hydrogels showed reduced adhesion scores, with an adhesion extent score of 0.67 and 2.33 on postoperative day 7 and 21, respectively, compared with the untreated control group (3 and 3 on postoperative day 7 and 21). In addition, the adhesion severity scores of HA-GA (2 wt%)/PluF hydrogel groups were 0.33 and 1.33 on postoperative day 7 and 21, respectively, compared with the untreated control groups (1.67 and 2.33 on postoperative day 7 and 21). Therefore, HA-GA/PluF hydrogels provide reversible thermosensitive properties with enhanced stability and retention, supporting their potential as injectable physical barriers for postoperative adhesion prevention.

Biomimetics doi: 10.3390/biomimetics11090631

Authors: Jun Huang Mengying He Guanghui Yang Xiuyi Ao Yupin Zhang Natalia Hartono Duc T. Pham

End-of-life (EoL) electric vehicle (EV) batteries comprise numerous interconnected components with complex topological and precedence relationships. These constraints significantly increase the difficulty of disassembly sequence planning (DSP), as feasible sequences must satisfy multiple dependency requirements. Moreover, the large number of possible disassembly alternatives creates a vast search space, making sequence optimisation susceptible to combinatorial explosion and convergence to local optima. Therefore, effective DSP requires both robust constraint-handling mechanisms to ensure sequence feasibility and efficient optimisation strategies to identify high-quality solutions. To address these challenges, this paper proposes a disassembly sequence optimisation method that integrates a hard-constraint rule base, the linear upper confidence bound (LinUCB) algorithm, and the Bees Algorithm (BA). First, a disassembly-oriented hard-constraint rule base is developed to standardise the identification of component topological relationships and precedence constraints, thereby ensuring the generation of feasible disassembly sequences. A LinUCB-based contextual adaptive operator-selection mechanism is subsequently introduced to dynamically select neighbourhood operators according to the current search state. A weighted multi-criteria evaluation function incorporating disassembly time, payment cost, and human–robot utility is integrated into the BA. Two representative EoL-EV battery case studies with different levels of structural complexity are used for validation. Across 50 independent runs, LinUCB-BA reduced the mean normalised weighted objective value by 39.01% and 28.12% relative to simplified swarm optimisation (SSO) and teaching–learning-based optimisation (TLBO), respectively, in the 27-component case, and by 7.19% and 2.33% in the 16-component case. Compared with the enhanced discrete Bees Algorithm (EDBA) ablation baseline, further reductions of 1.98% and 0.43% were achieved, together with lower run-to-run variability. These results indicate that the proposed framework is effective for the two investigated battery disassembly scenarios, while broader validation across additional battery architectures and operating conditions remains necessary.

Biomimetics doi: 10.3390/biomimetics11090630

Authors: Bo Yan Hongyuan Liu Decai Tang

Fish-like biomimetic robots increasingly combine compliant structures, soft and variable-stiffness actuation, distributed sensing, and autonomous control, but cross-study comparison remains difficult because biological inspiration, robotic embodiment, test boundaries, and mission definitions are heterogeneous. We synthesize the field through a mechanism-to-evidence framework that links biological mechanisms to measurable descriptors, robotic embodiment, controlled interventions, task-oriented evidence, and conditional design principles. Across the literature, the transferable unit is not external resemblance but a functional mechanism whose advantage remains measurable after robotic integration. Three conclusions recur: dynamic performance depends on matching stiffness, actuation frequency, damping, and fluid loading within the intended operating range; morphology, sensing, control, power, and payload must be co-designed; and component, free-swimming, controlled-task, and field studies support different scopes of inference. We translate these findings into nine evidence-informed conditional design principles with explicit applicability limits and discriminating validation tests. Major gaps remain in wet-state dynamic characterization, complete reporting of power boundaries and kinematics, uncertainty and failures, matched task-level comparisons, and long-duration field validation. The resulting framework shifts evaluation from peak metrics and taxonomic labels toward task-conditioned, evidence-bounded design decisions.

Biomimetics doi: 10.3390/biomimetics11090629

Authors: Dongbao Sui Zongwei Zhang Tianshuo Wang Sikai Zhao Yu Zhang Xinzui Wang

Architectured soft structures have unlocked new possibilities for designing continuum manipulators with tailored mechanical performance. This work introduces a bioinspired soft manipulator based on modular wave spring units, leveraging the high elasticity of wave spring structures to enable compliant deformation and axial extensibility of the soft manipulator. To achieve precise control of the soft manipulator, a neural network-based inverse kinematics framework is developed to establish an efficient mapping from desired end poses to tendon actuation lengths. An iterative learning control strategy is further integrated to compensate for material hysteresis, friction, and external disturbances. A prototype is fabricated using flexible 3D-printable material (TPU), and comprehensive experiments are conducted to validate extensible performance, point positioning accuracy, trajectory-tracking accuracy, and compliant interaction capability. The results demonstrate that the proposed wave spring-based soft manipulator achieves a high extension ratio and reliable control precision.

Biomimetics doi: 10.3390/biomimetics11090628

Authors: Zongkun Li Shanfa Tang

Complex optimization tasks in data-driven prediction and engineering applications often involve nonlinear, multimodal, and ill-conditioned objective functions. This study proposes an Enhanced Moss Growth Optimization algorithm (EMGO), an improved variant of the baseline MGO framework, to enhance exploratory step-size control and local covariance exploitation. EMGO incorporates two key algorithmic augmentations: a budget-adaptive jump regulation mechanism that balances global dispersal and fine-grained refinement, and a shrinkage-regularized covariance-guided sampling operator with relative eigenvalue flooring to exploit correlation structures among elite individuals without rank deficiency. The proposed algorithm is evaluated on the CEC2017 benchmark suite across 50 and 100 dimensions with 29 test functions, 30 independent runs, and a budget of 3×105 function evaluations per run, compared against ten state-of-the-art optimizers including CMA-ES, L-SHADE, SBO, and baseline MGO. Nonparametric Friedman ranking, Holm-adjusted Wilcoxon signed-rank tests, and runtime-matched analyses demonstrate that EMGO achieves highly competitive performance across high-dimensional landscapes. Furthermore, EMGO is applied to tune support vector regression (SVR) hyperparameters for effluent suspended solid (SS) prediction using the UCI Water Treatment Plant dataset under an expanding-window rolling-origin cross-validation scheme. EMGO-SVR achieves superior predictive accuracy (RMSE=5.58±0.64, MAE=3.97±0.46, R2=0.889±0.028), outperforming standard SVR, tree-based ensembles, and Bayesian optimization baselines. SHAP-based feature importance analysis confirms the physical and process consistency of the model predictions.

Biomimetics doi: 10.3390/biomimetics11090627

Authors: Yuhang Wang Yaohui Wang Wenshan Wang Huan Shen Eize J. Stamhuis Lining Sun Qian Wang Chao Liu

The legs of flying insects play a critical role in enabling seamless transitions between aerial and terrestrial environments. These appendages serve multiple functions, including landing, walking, jumping, and transitioning from jumping to flight (takeoff). Such capabilities have inspired engineers to seek similar multimodal mechanisms in Flapping-Wing Aerial Robots (FWARs) to expand their operational versatility across diverse environments. However, designing multimodal mechanisms with distinct kinematic and propulsive characteristics remains challenging, particularly in the domain of autonomous jump takeoff for FWARs, where research remains relatively sparse. In this study, inspired by the jumping takeoff strategy and hindleg kinematics of the Asian migratory locust (Locusta migratoria), we propose a functional bio-inspired jumping takeoff mechanism that extracts selected mechanical principles of the locust jumping system, including elastic energy accumulation, temporary mechanical locking, and rapid energy release. The mechanism employs a gear–crank–slider transmission system and utilizes one-way bearings to regulate the locking and disengaging states, enabling the storage and rapid release of energy for jump takeoff, thereby achieving autonomous takeoff of the robot. Adams dynamic simulations show that at a torsion spring angle of 40°, the mechanism achieves a maximum resultant velocity of 1.955 m/s, a jump height of 168.2 mm, and a horizontal displacement upon landing of 134.6 mm. Ansys Fluent (2024 R2) simulations under multiple operating conditions further confirm that the aerodynamic performance is optimal at a takeoff angle of attack(α) of 5° with a torsion spring angle(β) of 40°, yielding a lift-to-drag ratio of 3.005. This work presents a functional bio-inspired jumping takeoff mechanism based on selected mechanical principles of locust jumping, providing a potential approach for improving the autonomous takeoff capability of small-scale FWARs.

Biomimetics doi: 10.3390/biomimetics11090626

Authors: Xiubo Xia Jian Sun Xiaoyu Geng Pu Zhang Yongling Fu

Metaheuristics require sustained global search without sacrificing local refinement, yet many variable-structure methods change operators through one-way iteration schedules. We introduce the Weather State Ants Optimizer (WSAO), in which a discrete-time Markov chain recurrently selects one of three population updates. Sunny, cloudy, and rainy states correspond to global exploration, movement toward nests, and local refinement, respectively. An archive-based mechanism also maintains several spatially separated nests as concurrent search centers. Thirty independent runs compared WSAO with 11 algorithms on 29 CEC2017 and 12 CEC2022 functions. WSAO achieved the lowest Friedman mean rank on both suites, at 2.48 and 2.33. Across five constrained design cases, it joined the leading group by mean objective value on four cases and ranked second on pressure-vessel design. Targeted CEC2022 controls showed that no alternative transition matrix dominated the baseline. Eliminating the trial perturbation worsened every selected function, whereas the contribution of multiple nests depended on the landscape structure. The combined evidence supports recurrent state-controlled search as a competitive framework for continuous numerical and constrained optimization.

Biomimetics doi: 10.3390/biomimetics11090625

Authors: Xinyao Tang Xupeng Wang Xinying Xue Hongyan Liu Mengyuan Pu

Human augmentation is an important branch of robotics research aimed at reducing metabolic energy consumption, delaying fatigue, and increasing body speed. However, existing evaluation protocols lack systematic frameworks for unpowered hip devices. This study aims to reduce the energy consumption of human movement without providing additional power and to develop a hip-assisted exoskeleton device. Through gait, plantar pressure, and electromyography tests, the system studied the assistance performance of exoskeletons in three wearing states: “No exo.”, “Exo. on”, and “Exo. off”. A comprehensive evaluation method of unpowered lower limb wearable exoskeleton (CE-ULLWE) is established, featuring the novel three-condition design that isolates the structural mass effect from true assistance via the “Exo. off” condition, and integrates multi-indicator metrics including kinematics, dynamics, plantar pressure, EMG, and metabolic simulations. Combining wearing and exercise testing to obtain the human–machine compatibility of exoskeletons and the subjective comfort of users when wearing exoskeletons. Experimental results demonstrate that wearing the exoskeleton increases peak hip and knee angular velocities by 18.5% and 9.0%, reduces joint power, decreases plantar pressure center excursion by 32.3%, and lowers total metabolic energy consumption by 16.0%, confirming its effectiveness in reducing metabolic cost and delaying fatigue. This achievement has important theoretical guidance and practical application value for the design, function, and performance evaluation of wearable assistive exoskeleton products. The proposed CE-ULLWE offers a replicable, multi-indicator framework that clarifies assistive efficacy and guides future exoskeleton optimization.

Biomimetics doi: 10.3390/biomimetics11090624

Authors: Bojia Hou Fei Yu Ya Zhang Baojin Ping Zhaoxu Wang

Bionic intelligent robots operating in unstructured dynamic environments require perception systems that regulate uncertain observations according to their reliability and avoid converting transient disturbances into persistent spatial references. Inspired by reliability-weighted multisensory integration and by a functional abstraction of persistence-based evidence consolidation in biological navigation, this paper proposes a bio-inspired reliability-constrained LiDAR–inertial odometry framework. Each point-to-map observation is evaluated using residual consistency, local geometric quality, and voxel-level temporal stability. The fused reliability score regulates both the ESIKF state update and incremental map maintenance: low-confidence observations are continuously down-weighted, highly reliable points are admitted to the persistent map, ambiguous points are retained in short-term candidate memory for multi-frame verification, and unreliable points are rejected. The framework translates biological design principles into an engineering perception–memory architecture rather than reproducing a specific neural circuit. Repeated experiments on public datasets and a wheeled mobile robot platform show comparable accuracy in two normal sequences. Across five dynamic sequences, the complete method reduces localization RMSE by 14.08–28.88% relative to Fast-LIO2 and achieves lower mean RMSE than Dynamic-LIO on all five evaluated dynamic sequences. Frozen-map evaluation further yields 7.77–15.71% lower point-to-map RMSE together with higher consistent-correspondence ratios and coverage, while the maximum mean RMSE deviation in the parameter-sensitivity study remains below 7.2%. The maximum measured processing time is 18.69 ms per scan, maintaining real-time operation for a 10 Hz LiDAR.

Biomimetics doi: 10.3390/biomimetics11090623

Authors: Lidong Sun Ye Wang Hao Cha Fuchun Sun

Compliant muscle–tendon mechanics can improve terrain adaptation in musculoskeletal robots, but heterogeneous terrains impose competing requirements on compliance, propulsion, foot clearance, and support transfer. This paper proposes Value-Gradient Generalist Design (VGGD), a framework for selecting one fixed physiological muscle–tendon parameterization for cross-terrain locomotion while preserving skeletal topology and muscle routing. VGGD searches a compact PCA-based manifold that coordinates bounded scale factors for muscle strength, contraction-velocity capacity, and passive elastic response. A design- and terrain-conditioned value proxy is learned from sampled latent designs and then optimized by proximity-regularized projected value-gradient ascent under a soft-worst objective. Independent proxy validation uses 50 random designs solely for calibration and 100 separately sampled test designs excluded from fitting and Stage-B optimization. On the 100-design test set, the proxy shows positive agreement with mean cross-terrain return (Pearson r=0.580, Spearman ρ=0.565) and worst-terrain return (r=0.583, ρ=0.551), with all bootstrap intervals above zero and Holm-adjusted permutation p=0.0006. A separate paired local-direction test uses 60 previously unused evaluation seeds and equal feasible-space perturbation radii; at the nominal, midpoint, and selected designs, the proxy-gradient direction agrees with improvements in mean and seed-wise worst-terrain rollout return. After design selection, the muscle–tendon parameters are fixed, and a terrain-aware controller is trained with variational information-bottleneck regularization and auxiliary expert distillation. Checkpoint-resolved evaluation records identify 216/300 successes and an overall mean distance of 12.14 m for the complete pipeline, compared with 57/300 and 6.20 m for nominal-body PPO. The three checkpoint success rates are 84%, 78%, and 54% for Proposed and 49%, 0%, and 8% for PPO; exact two-sided policy-level permutation tests yield p=0.10 for success rate and p=0.20 for mean distance. All methods receive the same nominal 100-million-step final-controller budget per training run, while the 10 M Stage-A budget and expert-pretraining costs are reported separately.

Biomimetics doi: 10.3390/biomimetics11090622

Authors: Jincheng Sun Yang Luo Yaqi Chu Xiao Wang Yunxiang Jiang Xingang Zhao Yiwen Zhao

Long-horizon robotic manipulation requires coordinating multiple motor primitives under uncertainty, especially in contact-rich and changing environments. Existing end-to-end visuomotor policies often lack explicit temporal structure, causing brittle execution and limited recovery after disturbances. Inspired by biological motor control, where reusable primitives are organized through phase decomposition and feedback-dependent transitions, we propose a bio-inspired phase-aware framework that represents execution as structured transitions over perception-grounded motor phases. The framework integrates three modules: a Multimodal Phase-and-Primitive Detector that extracts semantically and physically consistent phases from visual, proprioceptive, and force–torque signals; a Multimodal Perception Skill Graph (MPSG) that encodes feasible phase transitions and supports skipping, rollback, and recovery; and Promptable Phase Control, which converts language instructions into graph-level ordering constraints for task reordering without retraining low-level policies. Experiments on four multi-stage tasks show improved robustness, increasing the average disturbed-condition success rate from 25.0% to 73.8% relative to the monolithic Action Chunking with Transformers (ACT) baseline.

Biomimetics doi: 10.3390/biomimetics11090621

Authors: Xiaohu Liu Hongke Pan Xiaogang Yu Jun Xi Yujun Peng

Nighttime driving safety remains a critical challenge in modern transportation systems: insufficient ambient lighting significantly degrades visual perception quality, adversely affecting both human drivers and advanced driver-assistance systems (ADAS) and directly threatening road users’ safety. Traditional image enhancement methods often suffer from color distortion and visual artifacts, whereas existing deep learning approaches typically require paired training data and incur substantial computational overhead. To address these limitations, this paper presents BLEN (bio-inspired low-light enhancement network), a zero-reference deep learning framework that integrates biological vision principles with efficient convolutional architectures. Specifically, BLEN leverages Retinex theory for illumination–reflectance decomposition, is inspired by and functionally approximates lateral inhibition mechanisms for edge enhancement, and incorporates a Large-Kernel Convolution with Attention (LKCA) module that reduces the parameter count of the LKCA encoder block by 76% (0.56 M vs. 2.34 M for a standard 13 × 13 convolution) relative to standard large-kernel operations. Extensive experiments on the SICE and LOL benchmarks demonstrate that BLEN achieves state-of-the-art performance among real-time, edge-deployable zero-reference methods on the SICE benchmark, yielding a peak signal-to-noise ratio (PSNR) of 23.67 ± 0.14 dB and a structural similarity index measure (SSIM) of 0.891 ± 0.004 on SICE while maintaining 2.10 M parameters (2.1 MB in INT8, 8.4 MB in FP32). Furthermore, the proposed method enables real-time inference at 31 frames per second (FPS) on embedded platforms, including the HiSilicon SS928 and Jetson Nano, demonstrating that the proposed method is an efficient and effective front-end for camera-based ADAS perception on automotive-grade edge hardware.

Biomimetics doi: 10.3390/biomimetics11090620

Authors: Wiktor Nisterenko Yash Raj Singh Katarzyna Ewa Greber Karolina Jagiełło Krzesimir Ciura

Developmental and adult neurotoxicity are hard to assess with conventional in vitro assays, which rarely account for how a chemical actually distributes once inside the body. Biomimetic chromatography offers a practical way to fill that gap, using stationary phases that mimic phospholipid membranes and major plasma proteins to yield experimental descriptors of lipophilicity, membrane affinity, and protein binding without needing radiolabeled compounds or large sample amounts. Working within the Partnership for the Assessment of Risk from Chemicals, we profiled 67 neurotoxicology-relevant chemicals on immobilized artificial membrane, human serum albumin, and alpha-1-acid glycoprotein columns alongside lipophilicity measured at three pH values. The chromatographic lipophilicity scale matched literature log p values closely. Principal component and hierarchical cluster analyses of the seven descriptors pointed to one dominant hydrophobicity axis and a smaller, second axis tied to ionization and albumin binding. The panel’s overall chromatographic profile did not separate neurotoxic from non-neurotoxic reference compounds, but the most membrane- and protein-avid compounds were disproportionately positive controls and warrant closer attention. The resulting dataset gives New Approach Methodologies a ready, experimentally grounded input for in vitro-to-in vivo extrapolation and baseline toxicity assessment.

Biomimetics doi: 10.3390/biomimetics11090619

Authors: Qichao Song Siyi Chen Huiling Zhang

Bio-inspired architectural design increasingly relies on generative artificial intelligence to expand early-stage concept exploration, yet current workflows often suffer from vague requirement definition, subjective proposal selection, and weak connections between user expectations, visual generation, and design evaluation. This study proposes a closed-loop human–AI decision-support framework for biomimetic architectural concept generation. The framework first uses Kansei-oriented requirement analysis and the KANO model to identify and classify stakeholder expectations concerning morphology, structural rationality, environmental integration, cultural narrative, visual novelty, interactivity, and sustainability. On the basis of 282 valid questionnaire responses, the most influential requirement categories are further translated into an Analytic Hierarchy Process (AHP) hierarchy, where expert judgement from a five-member specialist panel is used to derive criterion and sub-criterion weights. These weights are then converted into structured prompts—via a formally specified weight-to-language conversion strategy—to guide a diffusion-based image-generation system toward more targeted biomimetic concepts. Finally, TOPSIS is applied to rank generated design alternatives according to the same weighted criteria, thereby creating a traceable link from requirement discovery to generation and decision-making. Case studies involving eagle-, manta ray-, and cheetah-inspired architectural concepts indicate that the framework improves the explicitness of design objectives, supports more consistent comparison among alternatives, and reduces reliance on purely intuitive aesthetic judgement. An ablation comparison confirms that each stage of the framework contributes incrementally to the quality of final outcomes. This study proposes an integrated workflow that combines requirements modeling, multi-criteria evaluation, and AI-assisted visual design.

Biomimetics doi: 10.3390/biomimetics11090618

Authors: Kyohei Wakabayashi Tetsuya Oda

Manual soldering requires phase-dependent coordination of posture, hand movement, tool position, and visual attention. We propose a depth-camera framework that estimates the work phase and detects deviations using three-dimensional upper-body and hand features, task-related object positions, and gaze-related approximation features derived from facial orientation and head posture. The system estimates three predefined phases: preparation, active soldering, and cleanup. Windows with insufficient phase confidence are assigned to Uncertain Phase and routed to review rather than treated as deviation labels. In the evaluation, normal trials showed stable process sequences, whereas trials with scripted simulated unsafe-like movements produced local increases in the deviation score and review-required intervals associated with reduced phase-estimation reliability. These findings suggest that the framework may support retrospective safety-related assessment and the identification of process-inconsistent operations in seated manual soldering under controlled laboratory conditions. The bio-inspired contribution is a functional abstraction of phase-dependent perceptual-motor coordination into context-dependent engineering reference patterns and an uncertainty-aware review mechanism.

Biomimetics doi: 10.3390/biomimetics11090617

Authors: Zhimin Cao Shuilin Wang

Ice accumulation severely threatens the safe operation of aerospace, wind power and power transmission facilities, while traditional deicing technologies suffer high energy consumption and secondary pollution. Bioinspired superhydrophobic photothermal coatings integrate micro-nano bionic architectures and light-to-heat conversion media to realize synergistic passive ice suppression and solar-driven active deicing, emerging as an eco-friendly anti-icing route. This critical review systematically sorts scattered experimental findings from existing literature and clarifies that most observed performance correlations are restricted by non-uniform test conditions, rather than universal mechanistic laws applicable to all service scenarios. All comparative observations between different photothermal material systems and biomimetic structures are derived from discrete experimental datasets without harmonized measurement frameworks, so definitive cross-group performance rankings cannot be generalized across all icing environments. This work classifies mainstream photothermal filler categories and corresponding microfabrication techniques, analyzes inter-study data discrepancies caused by the lack of unified ice characterization standards, and elaborates multi-dimensional practical limitations of lacquer-based anti-icing coatings, including weak mechanical robustness and heavy dependence on solar irradiation. Finally, we propose targeted breakthrough directions involving multi-mode energy synergy, computational structural optimization and standardized characterization protocols, to provide targeted mechanistic guidance for developing high-performance, industrially viable bionic anti-icing lacquers.

Biomimetics doi: 10.3390/biomimetics11090616

Authors: Burak Aggul Amir Seyyedabbasi

Information in the Glider Snake Optimizer (GSO) propagates through a leader–predecessor chain. This structure is simple, but it can lose diversity when adjacent agents converge to the same region. The proposed Multi-Strategy Glider Snake Optimizer (MSGSO) retains the original GSO update and subsequently applies nonlinear dynamic polynomial mutation (NDM), elite opposition-based learning (EOBL), and quadratic interpolation (QI). MSGSO was evaluated on CEC 2019 and CEC 2022 with 30 matched random seeds and 15,000 objective evaluations for every algorithm–problem pair. The experiments included seven alternative optimizers and all single, pairwise, and three-operator GSO variants. Across the 34 benchmark problems, MSGSO significantly outperformed GSO on 31 and showed no significant loss. It nevertheless ranked third in each external comparison: L-SHADE led CEC 2019 and CEC 2022 at D=10, while CMA-ES led CEC 2022 at D=20. The ablation attributed most of the gain to NDM; NDM–GSO led the GSO variants on CEC 2019, and NDM–QI–GSO led at both CEC 2022 dimensions. In the UAV study, MSGSO returned 29 feasible paths in 30 runs in Scenario 1 and feasible paths in every run in the other two scenarios. It led the feasibility-first ranking in Scenarios 1 and 2 and placed third in Scenario 3. Thus, MSGSO improves its parent algorithm, although the full three-operator sequence is not consistently the best configuration.

Biomimetics doi: 10.3390/biomimetics11090615

Authors: Zhiyuan Liang Jie Xue Haiming Mou Qingdu Li Jianwei Zhang

Robust blind locomotion over complex unstructured terrains relies on accurate estimation of robot states and surrounding terrain geometry. However, under real-world deployment conditions without exteroceptive perception, it remains challenging to accurately estimate robot states and infer surrounding terrain structures solely from noisy proprioceptive observations. Existing methods commonly learn single-scale implicit terrain representations from historical proprioceptive observations and explicitly estimate robot states. However, they lack explicit terrain estimation and may lose critical geometric details. Moreover, single-scale terrain information is insufficient to capture both local geometric structures and global terrain trends. To address these issues, we propose a Terrain and State Implicit-Explicit Estimation (TSIE) framework to improve the locomotion capability of bipedal robots over complex terrains. TSIE encodes long-horizon proprioceptive observations using a Long Short-Term Memory (LSTM) network and introduces a dual-branch architecture consisting of a Terrain Implicit-Explicit Estimator (Terrain-IE) and a State Implicit-Explicit Estimator (State-IE). Terrain-IE performs multi-scale terrain implicit-explicit estimation by explicitly estimating a local high-resolution height map and implicitly reconstructing a global low-resolution height map. By preserving gradient connections between the two terrain branches, Terrain-IE enables joint implicit-explicit training of multi-scale terrain representations, improving terrain understanding and estimation accuracy. State-IE explicitly estimates the base linear velocity and foot-centered height map, while implicitly reconstructing future proprioceptive states to further improve tracking performance and locomotion robustness. We validate TSIE through simulation and real-world experiments on a full-sized bipedal robot platform with a height of 170cm and a mass of 35kg. Experimental results show that TSIE outperforms baseline methods in complex-terrain traversal capability, terrain and state estimation accuracy, and velocity-tracking stability. Real-world deployment further demonstrates the robustness of TSIE across indoor and outdoor complex terrains, achieving a 95% success rate in continuous stair ascent and descent tasks.

Biomimetics doi: 10.3390/biomimetics11090613

Authors: Junhong Luo Jianbin Yu Hui Cao Qiyue Tan Jinheng Chen Jing Xiao

Traditional P300 brain–computer interface (BCI) systems rely on multi-channel EEG acquisition, causing cumbersome setup, lengthy preparation, and high user workloads, which limits their real-world application. To enhance practicality, this paper proposes a fixed few-channel selection framework based on a heuristic algorithm to balance decoding performance and user experience. We integrated a genetic algorithm (GA) with Bayesian linear discriminant analysis (BLDA) to identify a strongly generalizable few-channel combination from a traditional eight-channel system, avoiding costly subject-specific recalibration. Validating this method, 48 healthy subjects completed rigorous offline and online virtual reality (VR) experiments. Results showed that the proposed three-channel system maintained highly comparable accuracy and information transfer rates to the eight-channel system, showing no significant performance degradation. Crucially, the few-channel scheme reduced equipment preparation time by 90% (from 30 to 3 min). Furthermore, NASA-TLX workload evaluations confirmed a significant reduction in users’ psychological and physical burdens (p < 0.05). Ultimately, while preserving core interaction performance, this few-channel strategy vastly improves user experience and system practicality, offering key theoretical and practical support for implementing lightweight, user-friendly BCI systems.

Biomimetics doi: 10.3390/biomimetics11090614

Authors: Alaeddin Koska

The industrial relevance of biomimetic design rests on whether a biological principle can be converted into an engineering solution that is manufacturable, verifiable, scalable, and traceable in operation. This critical integrative review considers that problem from a production-management perspective and examines the contribution of Industry 4.0 technologies to design, validation, manufacture, and industrial use. The evidence base consists of 50 purposively selected Gold Open Access full texts (n = 50) from the Web of Science Core Collection. Each study was coded for its biological model, design principle, digital technology, material and manufacturing route, manufacturability, scale, evidence level, and reported performance. CAD, modeling, simulation, machine learning, optimization, and digital manufacturing support different parts of the transformation process, whereas sensors, robotics, and IIoT supply functional and operational feedback. The direct studies report laboratory- or prototype-level improvements in strength, stiffness, material or support use, drag, thermal and energy performance, sensing, and actuator durability. None, however, demonstrates E4–E5 evidence across the complete chain. Evidence on serial production, process capability, stable quality, total cost, standardization, certification, and life-cycle performance is still limited. Drawing on these findings, the review proposes a technology–production–performance framework and a set of measurable Industry 5.0 indicators. Its main limitations are the Gold Open Access and English-language boundaries, purposive rather than exhaustive selection, unavailable exact search strings and record-level exclusion counts, single-researcher coding, and outcome heterogeneity that prevented meta-analysis.

Biomimetics doi: 10.3390/biomimetics11090612

Authors: Fabrice Saffre Hanno Hildmann

We present a nature-inspired approach for self-organizing logistics in contested environments characterized by uncertainty. The presented approach leverages territorial partitioning principles observed in natural collectives and systems. The logistics challenge addressed involves autonomously deploying a fleet of uncrewed mobile robots tasked with dynamically adjusting their positions to optimally service spatially and temporally fluctuating demands. Unlike traditional logistics, our approach is largely decentralized, relying almost exclusively on local interactions and decision-making. Monte Carlo simulations are conducted to evaluate performance across different scenarios, varying in client distribution (from uniform to highly clustered) and the range of the local perception of the logistic platform. Results demonstrate that the decentralized allocation strategy can deliver logistics support at a performance on par with a traditional clustering method, such as k-means, at runtime and without preliminary offline calculations.

Biomimetics doi: 10.3390/biomimetics11090611

Authors: Andrés Pérez Broderick Crawford Eduardo Rodriguez-Tello Jorge Mendoza Gino Astorga Ricardo Soto

Bio-inspiration has mainly been used to represent organisms, behaviors, or natural processes in the design of metaheuristics. This study proposes a different use: drawing on a biological phenomenon to model the temporal evolution of an internal parameter in an existing algorithm. Logarithmic Mean Optimization (LMO) is adopted as a case study, focusing on β, which scales the stochastic perturbation term and regulates the balance between exploration and exploitation. Inspired by the progressive transition observed during bird landing, a normalized arctangent trajectory controlled by m and k is proposed. Both hyperparameters were tuned through Bayesian optimization using the Tree-structured Parzen Estimator (TPE) implemented in Optuna. The 23 benchmark functions were divided into 12 tuning functions and 11 independent test functions. Nine β configurations were evaluated through 31 runs per function. The Friedman test showed significant differences among the variants (p=3.8269×10−10), and the proposed formulation achieved the best average rank (1.7273). Post-hoc Wilcoxon tests with Holm correction found significant differences against two of the eight alternatives. Overall, the results suggest that bio-inspiration, when used as a criterion for designing adaptive parameter-control mechanisms, can yield improvements and be considered a potential alternative in the design of new metaheuristic algorithms.

Biomimetics doi: 10.3390/biomimetics11090610

Authors: Lei Jiang Jinyi Chen Kaixin Lan Xianwei Liu Yongbin Jin Hongtao Wang

Robotic piano playing is a challenging benchmark for biomimetic dexterous manipulation, requiring precise timing, coordinated multi-finger motion, and stable key contact. This study proposes a robotic piano-playing framework based on a two-stage reinforcement learning curriculum. Musical Instrument Digital Interface (MIDI) data are converted into target-key and fingering grids to provide future musical goals for policy learning in a parallel MJLab simulation environment. A Soft Actor–Critic (SAC) agent takes a 2106-dimensional observation vector, including joint states, previous actions, musical phase, future key targets, fingering assignments, and piano-key states, and outputs a 21-dimensional continuous action vector for wrist, finger, and global hand-positioning control. Stage 1 weakens physical regularization to facilitate key-pressing acquisition, whereas Stage 2 strengthens power, velocity, acceleration, collision, posture, and finger-speed constraints to improve the regularity of policy outputs and readiness for real-world deployment. Simulation experiments on 30 s right-hand excerpts from Für Elise, Canon, and Beethoven’s Symphony No. 5 achieve frame-wise key-state F1 scores above 0.99 on the first two excerpts and approximately 0.945 on Beethoven. Real-world deployment uses open-loop playback of policy-generated high-level trajectories with low-level joint-position feedback and achieves F1 scores of 0.95, 0.91, and 0.83, respectively, while reproducing representative piano techniques such as chords, octaves, mixed black-and-white-key patterns, overlapping finger actions, and rapid sequential movements. These physical results demonstrate the feasibility of the proposed sim-to-real pipeline for complete 30 s executions; they are not intended as a statistical repeatability study. The results further show that biomimetic robotic hands can learn complex piano-playing skills from MIDI-based task objectives without relying on human motion demonstration trajectories.

Biomimetics doi: 10.3390/biomimetics11090609

Authors: Jingyuan Luo Siquan Wu Li Shen Chuliang Chi Hao Wu Yongquan Chen

High-voltage grid maintenance and chemical processing demand dexterous manipulation in confined, visually occluded environments. Existing teleoperation systems present a severe cost–fidelity trade-off: industrial platforms exceed USD 60,000 yet lack spatial directional resolution, while affordable alternatives omit fingertip force sensing entirely. We present STRP (Spatial Teleoperation Research Platform), a low-cost teleoperation system with concurrent motion capture and directional haptic feedback that encodes fingertip six-axis force vectors into four-directional vibrotactile cues (left, right, dorsal, volar). The master interface is a wearable 15-DoF exoskeleton glove with Hall-effect joint sensing and a 16-channel LRA array; the slave end is a 17-DoF dexterous hand with integrated micro six-axis force/torque sensors. Experimental validation demonstrates 97.5% blind discrimination accuracy for directional cues and an end-to-end latency of approximately 28 ms. In teleoperated pick-and-place tasks, multi-directional tactile feedback reduced task completion time by 41% (31.2 s to 18.4 s) and peak grip force by 32% (4.11 N to 2.79 N) versus no-feedback baselines. In visually occluded wall exploration, directional feedback enabled systematic location of all target walls, whereas single-site volar feedback permitted detection of only volar-facing surfaces and no-feedback conditions precluded any spatial reasoning. These results identify spatial directional vibrotactile feedback as a distinct sensory channel that cannot be substituted by magnitude-only vibration or vision alone. STRP establishes a foundation for affordable, high-fidelity haptic teleoperation in hazardous environments.

Biomimetics doi: 10.3390/biomimetics11090608

Authors: Nanhao Zhou Han Huang

Inspired by the natural curvature and jointed configuration of tortoise limbs, this paper presents a magnetically controlled soft robot based on a “programmable intrinsic curvature” design paradigm. When corrugated flexural notches are introduced into a single-material elastomeric leg, localized stress concentration enables each leg to acquire a predefined curved shape during fabrication. Combined with corrugated-straw-based sacrificial molding and magnetic-field-assisted curing, the strategy allows independent tuning of bending angle and leg length. Four legs are radially integrated onto a soft torso. Experimental results show that the stepping–pushing asymmetry of a single leg is governed by the predefined curvature, with the snap-through state corresponding to the equilibrium between magnetic attraction and paramagnetic deflection. The robot achieves speeds of 1.1 mm/s on dry ground, 9.5 mm/s in semi-submerged water, and 56 mm/s in fully submerged water (at 0.8 Hz, 340 mT), a ~50-fold increase in swimming speed compared with its terrestrial locomotion speed. It also demonstrates sharp turning, payload transport, rough surface traversal, and self-righting. This work elevates intrinsic curvature from passive geometry to an active design variable for soft robots in complex multi-environment scenarios.

Biomimetics doi: 10.3390/biomimetics11090607

Authors: Wencong Gan Jiehui Chen Qingdu Li Haiming Mou Jianwei Zhang

Tendon-cable transmission can reduce distal-limb inertia in full-size humanoids, but its elasticity, hysteresis, backlash, and multi-joint coupling introduce state-dependent joint-to-motor discrepancies. We present a hierarchical whole-body tracking framework for the 28-DoF Droid X3 that separates high-level motion learning from transmission compensation. A reference-residual policy is trained in simulation by single-stage proximal policy optimization (PPO) using a unified robot-space motion representation, globally anchored tracking rewards, hierarchical hard-example sampling, and tendon-oriented domain randomization. In simulation checkpoint evaluation, more than 90% of 12,674 tested reference motions are completed. Independently, a state-conditioned mapper is trained offline through a differentiable motor–joint forward model identified from physical motor-excitation data and connected in series between the frozen policy and the low-level motor controller. Randomized repeated Mapping-OFF/ON trials are conducted on two nominally identical Droid X3 units. Within every robot–motion block, the frozen PPO checkpoint, reference trajectory, controller settings, safety bounds, and frozen mapper weights are held fixed; complete trials are the statistical units. OFF converts desired joint positions with the robot-specific fixed static calibration, whereas ON feeds the complete policy-level desired-joint vector and measured plant state to the frozen mapper, which directly outputs the complete motor-position command. Across the complete physical trials, the aggregate action-completion rate is 68% with Mapping OFF and 79% with Mapping ON, an increase of 11 percentage points. Representative walk, squat, and dance trajectories illustrate lower tracking errors under Mapping ON, while individual frames and selected temporal fragments are used only for visualization.

Biomimetics doi: 10.3390/biomimetics11090606

Authors: Xu Zheng Chaobo Li Chenxi Gao

Grounded in structural biomimetics, this study extracts the multi-segmented tapered geometry from the 10-segmented lamellate antenna of Allomyrina dichotoma and maps it to microwave antenna design. Through biological characterization and parametric modeling, key geometric features—multi-segmented configuration, irregular contour, and bilateral symmetry—were extracted. Along the 2D pathway, a 10 × 10 × 1 mm3 PCB microstrip antenna was designed and fabricated, achieving 111% fractional bandwidth from 4.86 to 17.05 GHz with a peak gain of 2.15 dBi and a radiation efficiency of 67–72% across the operating band, plus two additional bands at 24.76–28.58 GHz and 32.66–37.73 GHz. The multiple resonance valleys on S11 curves and frequency-dependent surface current evolution indicate that multi-mode resonant coupling, perimeter increment, and symmetric aperture efficiency together underpin the ultra-wideband performance. Along the 3D pathway, a dipole antenna replicated via metallic 3D printing attains an electrical length of 0.17λ, with 66% bandwidth and 1.45 dBi gain, confirming the same geometric principle in a shape-preserving form. The 2D route favors planar integration and bandwidth, while the 3D route offers extreme miniaturization. This work provides experimental validation of cross-domain geometric mapping from biology to electromagnetics within structural biomimetics, and offers engineering evidence for the intrinsic versatility of this morphology across physical domains.

Biomimetics doi: 10.3390/biomimetics11090605

Authors: Deniz Korkmaz Gonca Ozmen Koca Cafer Bal Mustafa Ay Zuhtu Hakan Akpolat

Terrain types significantly affect the dynamics and locomotion performance of hexapod robots during walking. Terrain classification is a key solution to modify gait patterns in different terrains and recognize hazardous conditions. Perceiving the terrain with proprioceptive sensing is a robust and reliable approach in extreme conditions. In this paper, an efficient terrain classification approach for a hexapod robot is proposed. The proposed method combines a deep classification framework including the long short-term memory (LSTM) network and an effective statistical feature extraction. Proprioceptive inertial measurement unit (IMU) data is only used as the sensing system for the robot–terrain interaction. In the feature extraction process, four meaningful characteristic features, namely, the mean, median, Lomb–Scargle periodogram power spectral density (LPSD), and Welch’s power spectral density (WPSD), are extracted from the body orientation data using a sliding-window method. These features are combined and fed into the network to perform the training and testing processes. In the experiments, the proposed method is evaluated with commonly used soft computing and deep learning models. The classification performance for the concrete, pebble, and waxed tile terrains reaches 100% with the proposed method. The overall accuracy, precision, sensitivity, specificity, F1-score, and Matthew correlation coefficient are recorded as 95.45%, 96.36%, 95.28%, 98.86%, 95.49%, and 94.61%, respectively. These results demonstrate that the proposed approach delivers reliable classification performance with a low-cost and easy-to-implement solution.

Biomimetics doi: 10.3390/biomimetics11090604

Authors: Chaochuan Jia Feilong Yu Ting Yang Yu Liu Maosheng Fu Fang Wang Ling Li

Adaptive Reinforced Gray Langur Optimization (ARGLO), an enhanced variant of the Gray Langurs Optimizer, is developed for high-dimensional, multimodal, and nonlinear search landscapes susceptible to local trapping. Although the original GLO performs multi-population cooperative search by simulating the social structures of gray langurs, it still suffers from uneven random initialization, insufficient adaptive population partitioning, weak local perturbation, and premature convergence. ARGLO incorporates three strategies: good point set-based oppositional and quasi-oppositional learning initialization, hierarchical equilibrium adaptive population partitioning, and elite-guided hybrid mutation. Collectively, these mechanisms generate a higher-quality starting population, coordinate global search with local refinement, and reduce the risk of entrapment in suboptimal regions. Evidence from component-wise experiments together with the CEC test suite indicates that ARGLO delivers higher solution precision, steadier convergence, as well as more consistent performance, especially as dimensionality increases. Moreover, ARGLO is applied to near-infrared spectral feature selection and SVR parameter optimization for polysaccharide content prediction in Dendrobium huoshanense. Compared with unoptimized SVR, ARGLO-SVR reduces RMSE by 35.35% and improves R2 by 21.92%; compared with GLO-SVR, it reduces RMSE by 6.05% and improves R2 by 2.30%. These results demonstrate the effectiveness and application potential of ARGLO in complex optimization and rapid nondestructive quality detection of traditional Chinese medicinal materials.

Biomimetics doi: 10.3390/biomimetics11090603

Authors: Qunwei Zhu Tao Jiang Zirong Luo Yiming Zhu Guanhai Huang

High nonlinearity and limited computational resources create persistent challenges for achieving autonomous, stable, and accurate locomotion in insect-scale crawling robots, hindering their practical deployment. In this study, we build the mathematical models of the insect-scale micro crawling robot named PLioBot and propose a locomotion control strategy that eliminates the need for gait transitions. The locomotion transformation of the PLioBot prototype, from driving inputs to mechanical motion outputs, is decomposed into multiple motion processes. This study analyzes the theoretical models underlying these motion processes and develops the locomotion simulation model for the PLioBot based on the mathematical models and the multibody dynamics simulation tool. The locomotion control strategy with low computational requirements is designed based on a closed-loop PID controller, which regulates the robot’s locomotion by independently adjusting the step lengths of its left and right legs. The locomotion co-simulation system is established to validate the control strategy. The simulation results confirm that this control strategy enables the PLioBot to perform straight-line locomotion and turning without requiring any gait transition.

Biomimetics doi: 10.3390/biomimetics11090602

Authors: Abdullah Alabdulatif Shahd Alqaan Rana Abdulaziz Alhusika Talah Abdullah Almuhawwis Raghad Ahmed Almujaydil

Authenticated key exchange has been built for decades on a small number of hard mathematical problems, and the elliptic curve discrete logarithm is one of them. The structural complexity of biological macromolecules, however, has rarely been used as a source of secret material in such protocols. In this paper, a protein-derived authenticated key exchange protocol, called PDKE-2, is proposed to address this gap. The protocol keeps elliptic curve Diffie–Hellman at its core and adds two ingredients from structural biology. Firstly, a per-session base point G* is generated by applying RFC 9380 hash-to-curve to a session mapping table. Secondly, a long-term secret shared between the two parties is derived from the SHA-256 digest of an AlphaFold2 inter-residue distance matrix. In contrast to an earlier version of this design, the two parties run ephemeral Diffie–Hellman in each session and the session key is never sent over the channel; only transcript-based confirmation tags are exchanged. Thus, the protocol provides forward secrecy and avoids a confidentiality weakness that existed in the earlier version. The security of PDKE-2 is analysed in the Real-or-Random model, where session key secrecy, forward secrecy and mutual authentication are proved under the Gap Diffie–Hellman, PRF-HKDF and EUF-CMA-HMAC assumptions. In addition, the protocol is modelled in ProVerif against a Dolev–Yao attacker, and the secrecy of the session key, the secrecy of the shared secret and the injective agreement in both directions are all confirmed. A Python reference implementation is also provided. In the implementation, key agreement and mutual authentication succeed in every session, and the complete handshake takes about 15 ms in the reference code. It should be noted that the protein layer does not increase the elliptic curve security level; it works as an authentication secret whose strength depends on the difficulty of guessing the protein sequence when this sequence is kept private.

Biomimetics doi: 10.3390/biomimetics11090601

Authors: Xu Dai Guoqiang Lu Jiaxue Li

With the rapid growth of social media, online news has become increasingly multimodal, combining textual and visual information, posing new challenges for fake news detection. Existing methods often suffer from redundant features, distribution differences across modalities, and insufficient modeling of semantic interactions. To address these issues, this paper proposes an Adaptive Binary Osprey Optimization Algorithm and Cross-modal Disentangled Fusion model (ABOOA-CDF). First, an Adaptive Binary Osprey Optimization Algorithm (ABOOA) is developed for multimodal feature selection by integrating chaotic initialization, adaptive search, and binary mapping strategies to identify informative feature subsets. Then, a Cross-modal Relation Disentanglement Module (CRDM) is introduced to decompose multimodal representations into shared, discrepant, and complementary components, thereby enhancing semantic relationship modeling. Furthermore, an Adaptive Semantic Fusion Module (ASFM) dynamically learns fusion weights to generate discriminative multimodal representations. Experimental results demonstrate that ABOOA-CDF effectively improves detection performance. Compared with MFO and OOA, the proposed method achieves Accuracy improvements of 1.02 and 2.66 percentage points, respectively, verifying its effectiveness in feature optimization, cross-modal relation modeling, and semantic fusion.

Biomimetics doi: 10.3390/biomimetics11080600

Authors: Seçil Kulaç

Human–robot collaborative assembly line balancing and scheduling constitutes an NP-hard combinatorial optimization problem involving the simultaneous optimization of task assignment, resource allocation, processing mode selection, station-level scheduling, and ergonomic constraints. This study proposes a Q-learning-based hyper-heuristic genetic algorithm (QLHH-GA) to solve the cost-oriented ergonomic mixed-model human–robot collaborative assembly line balancing and scheduling problem. The proposed approach integrates bio-inspired evolutionary mechanisms of population variation and selection with adaptive, Q-learning-guided low-level heuristic selection. The Q-learning layer uses performance feedback to adapt the search strategy to different solution states while maintaining solution feasibility. A mixed-integer linear programming (MILP) model is also developed to minimize the total operating cost, including station opening, labor, robot operation, and energy consumption costs, while enforcing station-level energy expenditure (EE) limits. Computational experiments conducted using benchmark instances of varying sizes and a literature-based industrial case study demonstrate that QLHH-GA produces solutions comparable to those obtained by the MILP model on small-scale instances and maintains strong solution quality on larger instances, for which exact optimization becomes computationally prohibitive. These findings demonstrate the scalability and effectiveness of reinforcement-learning-guided hyper-heuristic search for designing cost-efficient and ergonomically constrained human–robot collaborative assembly lines.

Biomimetics doi: 10.3390/biomimetics11080599

Authors: Haiguang Zhang Feng Jiang Qianmin Gao Qingxi Hu Jiaxuan Feng

Vascular stent implantation is a major treatment for vascular diseases, yet postoperative infection and persistent inflammation increase the risk of in-stent restenosis. Herein, core–shell structured PCL/PEO-PVP fiber membranes co-loaded with ciprofloxacin hydrochloride (CIP) and curcumin (CUR) were fabricated via coaxial electrospinning. Orthogonal experiments were conducted to optimize critical spinning parameters through multi-index comprehensive evaluation. Characterizations confirm intact core–shell architecture and stable polymeric backbone structure. In vitro release tests reveal sequential drug-delivery behavior: a rapid initial release of hydrophilic CIP and a delayed sustained release of hydrophobic CUR were observed, contributing to early-stage antibacterial and long-term anti-inflammatory effects, respectively. The “antibacterial zone” method verifies favorable antibacterial activity against Staphylococcus aureus (S. aureus) and Escherichia coli (E. coli). ELISA results demonstrate the enhanced anti-inflammatory capacity of the dual-drug-loaded coaxial fiber membrane. Cellular assays confirm satisfactory cytocompatibility, and endothelial cells achieve normal proliferation and exhibit typical polygonal morphology on the membrane surface. This dual-drug-loaded coaxial-fiber membrane realizes coordinated sequential antibacterial and anti-inflammatory properties, which provides a feasible strategy for developing functional coatings toward vascular stents.

Biomimetics doi: 10.3390/biomimetics11080597

Authors: Sercan Demirci Durmuş Özkan Şahin Gülcan Yıldız Doğan Yıldız Samad Hasanlı

Classification is a fundamental technique in data mining that predicts categorical labels by analyzing input features. However, training Artificial Neural Networks (ANNs) using traditional methods often encounters challenges, such as getting stuck in local minima and slow convergence. To address these issues, this study proposes a novel hybrid model, IPA-ANN, which integrates the Immune Plasma Algorithm (IPA) to optimize the ANN’s connection weights and biases. The IPA, inspired by the immune plasma treatment process, utilizes a unique donor-receiver mechanism to balance exploration and exploitation in the search space. The proposed model was evaluated on nine benchmark datasets from the UCI repository and compared with 18 state-of-the-art metaheuristic algorithms, including Grey Wolf Optimization (GWO), Differential Evolution (DE), and Particle Swarm Optimization (PSO). Experimental results were analyzed using accuracy, F1-score, confusion matrices, and convergence graphs. The findings indicate that IPA-ANN achieves competitive and stable classification performance across different datasets while demonstrating favorable convergence characteristics in several cases. Furthermore, the study investigates the influence of donor–receiver parameters on the optimization process, highlighting the adaptability of the proposed framework. The reliability of these findings was further examined through repeated stratified 5-fold cross-validation and paired Wilcoxon signed-rank tests with Holm–Bonferroni correction on representative datasets, confirming that a subset of the observed performance differences are statistically significant, and through a computational cost analysis showing that IPA-ANN incurs no additional overhead relative to the majority of the compared algorithms. This study contributes to the literature by presenting the first documented application of IPA in ANN training and by providing a modular infrastructure for future metaheuristic-based ANN optimization studies.

Biomimetics doi: 10.3390/biomimetics11080598

Authors: Meiqi Qin Yongmei Ding

Bioinspired swarm optimizers are widely used for constrained engineering optimization, yet whole-vector updates can disrupt well-evolved components and insufficient elite refinement can limit late-stage accuracy. This study proposes DSD-LI-PO, an enhanced Parrot Optimizer (PO) integrating dynamic swarm dimension hybridization (DSD) with stagnation-triggered variable precision lens imaging (LI). DSD selectively inherits components from the current best solution to preserve useful dimensional information, whereas LI generates a progressively contracted elite-derived candidate when stagnation occurs. DSD-LI-PO is evaluated on CEC2017, six representative functions at 30, 50, and 100 dimensions, ablation variants, and five constrained engineering design problems. On CEC2017, it ranks second overall with an average rank of 1.8966 and records 28/1/0 wins/ties/losses against PO and 29/0/0 against IPO. In the scalability tests, it remains second in all three dimensions, with average ranks of 1.9167, 1.9167, and 1.7500. The complete method ranks first in the ablation study and second in the engineering tests, with feasible solutions in all runs. These results demonstrate that DSD-LI-PO provides a consistent and targeted improvement over PO while maintaining competitive performance against strong non-PO optimizers.

Biomimetics doi: 10.3390/biomimetics11080596

Authors: Gyeongsu Sim Hojin Jin Sangyoon Woo Won-Gyu Bae

Despite the rapid proliferation of unmanned aerial vehicles (UAVs) across industrial, agricultural, and scientific domains, their deployment remains constrained by limited endurance, aerodynamic inefficiency, and acoustic emissions, all mediated by a shared onboard energy budget. Existing biomimetic UAV reviews have generally treated aerodynamics, structures, sensing, control, and energy systems as parallel topics rather than as interacting components of a unified aerial architecture. Drawing primarily on literature published between 2015 and June 2026 and identified through searches of Web of Science, Scopus, and Google Scholar, this review addresses this gap by examining bio-inspired technologies across six principal domains: aeroacoustic and passive flow control, aerodynamic efficiency, multifunctional structural composites, neuromorphic sensing and control, ionic energy storage, and energy harvesting. Its principal contribution is a cross-domain synergy analysis identifying five performance couplings and one structural enabling architecture through which these domains interact physically and functionally. Representative examples include serration-based propeller geometries that can simultaneously reduce noise and power demand; morphing wing surfaces that serve as both aerodynamic structures and triboelectric harvesting substrates; and neuromorphic spiking neural networks that have been reported, in specific event-vision inference benchmarks, to reduce inference energy by three to four orders of magnitude relative to embedded graphics processing unit (GPU)-based implementations. Mechanical harvesting outputs nonetheless remain orders of magnitude below propulsion requirements and are thus positioned as supplementary. Four systemic barriers (unquantified mass–energy balance, undocumented durability, aeroelastic co-design gaps, and heterogeneous metrics) are evaluated, and the resulting synthesis indicates that advancing bio-inspired UAVs requires a transition from structural imitation to functional, system-level biomimetics.

Biomimetics doi: 10.3390/biomimetics11080595

Authors: Ganbayar Batchuluun Sung Jae Lee Su Jin Im Kang Ryoung Park

Although deep learning methods have shown high performance in chest X-ray classification, high accuracy alone does not guarantee reliable reasoning. A model may still exhibit pathological behavior, such as unstable evidence usage under harmless input changes, inconsistent reasoning across augmented views, excessive dependence on surrounding frame information, and appearance bias dominance, where prediction relies too heavily on intensity while neglecting texture and shape. In this paper, we propose a bio-inspired pathology-aware, factor-aware framework for explainable and reliable chest X-ray classification, inspired by biological vision principles such as figure–ground separation, selective attention, and balanced use of complementary visual cues. During training, the method regularizes appearance bias dominance through evidence-guided counterfactual perturbations that mimic cue-suppression analysis in biological perception, thereby revealing and penalizing excessive factor dependence. During testing, it evaluates model behavior using four criteria: reasoning stability, augmentation inconsistency, appearance bias dominance, and framing sensitivity. This combination enables the framework to go beyond conventional inference-time explanation by both correcting pathological behavior during training and exposing it during evaluation. From a biomimetic perspective, the framework encourages the model to separate relevant foreground anatomy from surrounding background and to avoid over-reliance on a single dominant cue. The proposed approach improves interpretability and reliability without modifying the backbone architecture or increasing model size or inference-time cost. The proposed training process improved the F1-score of DenseNet-121 from 0.899 to 0.931, while also producing more stable and balanced reasoning.

Biomimetics doi: 10.3390/biomimetics11080594

Authors: Zhen Tang Xingze Chen Xin Wang Yangfan Yang Shanhong Tang Linpeng Liu

Flexible pressure sensors with porous structures are essential for wearable electronics and robotic perception. However, traditional porous flexible sensors suffer from poor stability and long recovery times. To address these challenges, a strategy integrating bionic architectures inspired by the rigid–flexible coupling structure of Bambusa textilis and the micro-protrusion structure of Salvia plebeia R.Br. is proposed. An aluminum sheet serves as the support layer, and the sensing layer is prepared through mold replication, material impregnation, and layer-by-layer assembly, offering a simple and scalable fabrication route. The sensor exhibits a broad effective pressure range of 0–31.5 kPa, with a minimum resolvable force of 0.2 N, and within its operating range (0–23.5 kPa), the relationship between resistance and pressure exhibits a trend that can be fitted to a quadratic term, with a coefficient of determination of 0.9986. It achieves a minimum response time of 350 ms and maintains stable signals under dynamic loading at different frequencies, indicating reliable detection of low-frequency weak pressures. After 5000 loading–unloading cycles, the device shows no obvious performance degradation. When mounted on a robotic foot, the sensor successfully distinguishes land, sponge, sand, and pebbles surfaces, demonstrating its potential for intelligent environmental perception.

Biomimetics doi: 10.3390/biomimetics11080593

Authors: Hee Sook Hwang Chung-Sung Lee

Localized therapeutic delivery has gained increasing attention as an effective strategy in enhancing treatment efficacy while at the same time minimizing systemic side effects. However, conventional hydrogel-based therapeutic delivery systems often suffer from poor tissue retention and insufficient control over delivery. This drawback is highly pronounced in wet and dynamic biological environments. So, catechol-based adhesive hydrogels have emerged as promising biomaterials for localized therapeutic applications and are inspired by the remarkable wet-adhesion capability of marine mussels. As a highlight, catechol chemistry enables robust tissue adhesion through multiple intermolecular interactions, including hydrogen bonding, metal coordination, and covalent coupling. At the same time, it contributes to hydrogel cohesion and structural stability. Recent advances in hydrogel engineering have expanded the functionality of these systems through integration of injectable formulations, self-healing networks, nanocomposite reinforcement, and stimuli-responsive biointerfaces. These developments have transformed adhesive hydrogels from tissue sealants into multifunctional therapeutic platforms capable of enhancing tissue retention, regulating therapeutic release, and dynamically interacting with biological microenvironments. Here, we review molecular mechanisms underlying catechol-mediated adhesion and discuss recent progress in advanced adhesive hydrogel systems. We further highlight their therapeutic applications in wound healing, musculoskeletal regeneration, exosome and gene delivery, immunomodulatory therapies, and localized cancer therapy. Finally, current translational challenges and future opportunities in developing next-generation smart biointerfaces for precision regenerative medicine are discussed.

Biomimetics doi: 10.3390/biomimetics11080592

Authors: Zhongyue Lu Zhichao Zhu Shanjun Chen Tao Jiang Zirong Luo

This paper presents a load-capacity-constrained arm-angle planning method for a centrally driven humanoid robotic arm. Joint-range and singularity constraints are projected into the one-dimensional arm-angle domain to form a geometric feasible set. A static load-capacity constraint is derived from gravity torque, the Jacobian-transpose mapping of a known endpoint load, and individual actuator-torque limits, and is projected into the same domain. Continuous arm-angle values are then selected along a prescribed Cartesian path within the intersection of the geometric and mechanical feasible sets. In a heavy-load simulation, the maximum output-power metric decreased by 13.3%, and the energy value decreased from 30.60 J to 22.95 J (25.0%). In a prototype proof-of-principle test with a 13 N payload (approximately 1.33 kg), each trajectory was executed three times; the controller-recorded shoulder peak was approximately 1.83% lower, and the recorded energy value decreased from 30.378 J to 28.85 J (5.03%). The prototype values are descriptive because run-level statistics and measurement uncertainty are unavailable, and the experiment covers only one payload and one path. The results support the proposed static, load-capacity-aware planning principle for the tested slow-motion conditions but do not establish general performance or real-time suitability.

Biomimetics doi: 10.3390/biomimetics11080591

Authors: Dekel Maroz Adi Aharonov Hod Hoenig Mirit Sharabi

Replicating the nonlinear, pressure-dependent mechanical behavior of native arteries remains a central challenge in small-diameter vascular graft design, where compliance mismatch between synthetic grafts and host vessels is strongly linked to graft failure. Building on a silk fiber-reinforced alginate–polyacrylamide interpenetrating polymer network (IPN) hydrogel platform, we investigated whether biomimetic vascular structural motifs, specifically fiber reorientation and crimp, can be used as programmable design parameters to tune the tensile and pressure-dependent mechanical response of tubular constructs toward native coronary artery behavior. Three architectures were fabricated: a cross-plied (CP) baseline, a 25° reoriented configuration, and a crimped CP configuration. Under internal pressurization, fiber reorientation and crimp significantly increased compliance at low physiological pressures, with a consistent directional trend across the full pressure range, by extending the low-stiffness toe region preceding fiber recruitment while preserving tensile stiffness. Across the investigated pressure range, all architectures exhibited compliance within the reported range of native coronary arteries, with crimped constructs producing pressure–diameter behavior that most closely resembled young coronary arteries, whereas the CP baseline more closely resembled aged coronary arteries. These findings demonstrate that biomimetic structural motifs, without altering material composition, can serve as programmable design parameters for engineering vascular mechanics, enabling a single material platform to reproduce distinct physiological mechanical phenotypes through architecture alone.

Biomimetics doi: 10.3390/biomimetics11080590

Authors: Yu Liu Feilong Yu Yaqi Yang Xingyu Gao Maosheng Fu Chaochuan Jia Zhengyu Liu

Dendrobium polysaccharide is an important indicator for evaluating the quality of Dendrobium huoshanense. To improve the prediction accuracy of polysaccharide content, this study proposes an improved Artificial Lemming Algorithm (IALA) optimized BP neural network model. In IALA, a periodic mutation strategy and a fast hybrid opposition learning strategy (FHOBL) are introduced to enhance population diversity, improve global search ability, and avoid premature convergence. The proposed IALA was first evaluated on CEC2017 and CEC2020 benchmark functions. Experimental results show that IALA achieves better or competitive performance compared with seven other algorithms in terms of mean fitness, best fitness, and standard deviation. Statistical tests, including Wilcoxon rank-sum and Friedman tests, further verify the significant superiority and robustness of IALA. Then, IALA was used to optimize the initial weights and thresholds of BP neural networks for Dendrobium polysaccharide content prediction. The results show that IALA-BP achieves the best overall prediction performance, with an R2 of 0.8731, RMSE of 2.1581, and MSE of 4.6683. Compared with standard BP and other optimized BP models, IALA-BP provides more accurate and stable prediction results. Therefore, the proposed IALA-BP model is effective for rapid prediction of Dendrobium polysaccharide content.

Biomimetics doi: 10.3390/biomimetics11080589

Authors: Wenzhi Zhou Tiantian Li

Integrating traditional cultural heritage, such as Chinese paper-cutting, into modern product design is a critical challenge. To address the lack of a systematic method, this study proposes the PPI-SCAMPER method: a structured Design-by-Analogy (DbA) method that integrates knowledge bases (a Paper-Cutting Element Category and a Flat Process Category) with SCAMPER heuristics. Subsequently, an empirical study involving 18 experts and 146 users was conducted to evaluate the method’s outcomes. Multimodal data—including expert ratings of product attributes, user ratings of attractiveness, and eye-tracking data—were collected, and hierarchical regression and Random Forest models were employed to analyze the underlying mechanisms of attractiveness. The results indicate that products designed via PPI-SCAMPER achieved significantly higher ratings for user attractiveness, novelty, and culture attribute. Eye-tracking data revealed that the improved products elicited greater overall visual exploration. The exploratory analysis of associative patterns showed that product attributes constituted the primary basis for attractiveness judgments, with eye-tracking data providing significant yet limited incremental explanatory power. This study not only constructs an innovative design pathway for integrating paper-cutting culture from a DbA perspective but also provides a comprehensive framework for evaluating culturally inspired innovations through multi-dimensional empirical data.

Biomimetics doi: 10.3390/biomimetics11080588

Authors: Sumeyye Altunok Gunes Kibar Fatma Aylaz Dide Su Demirel Veli Cengiz Ozalp

Prostate cancer remains a major clinical challenge, partly due to tumour–immune interactions that contribute to immune evasion and therapeutic resistance. The CD47–SIRPα axis is a key macrophage-associated immune recognition pathway; and materials-oriented platforms that allow preliminary investigation of tumour–macrophage interaction patterns remain valuable. Here, we report an optically traceable biohybrid nanoplatform based on CD47 aptamer-functionalized fluorescent carboxyl-functional MMES-POSS hybrid nanoparticles. The nanoparticles were synthesized within 5 min using UV-induced free-radical emulsion polymerization in an ethanol–water system and exhibited spherical morphology, SEM-derived dry-state mean diameters below 100 nm, negative surface charge, and retained fluorescence due to RITC encapsulation within the crosslinked hybrid matrix; however, DLS measurements revealed pronounced aggregation and polydispersity in aqueous dispersion. SEM and EDX provided complementary evidence of nanoparticle morphology and elemental composition, whereas zeta potential and Nanodrop measurements provided evidence consistent with surface functionalization, and FTIR confirmed retention of the core POSS, carbonyl, and RITC-associated chemistry. In a PC-3/THP-1 macrophage co-culture model, Annexin V–FITC/PI flow cytometry showed a concentration-dependent redistribution of cell populations, with the most pronounced early apoptotic enrichment observed at 2 µg/mL. However, pairwise exploratory comparisons among concentrations did not reach statistical significance, and this is discussed as a limitation of the present exploratory dose–response characterization. These findings indicate preliminary interaction-associated response patterns rather than definitive receptor-specific targeting or therapeutic CD47–SIRPα blockade. Overall, this modular POSS-based biohybrid interface provides a materials-oriented platform for probing prostate cancer–macrophage interaction profiles and guiding future mechanistic validation studies.

Biomimetics doi: 10.3390/biomimetics11080587

Authors: Jingya Zhang Yu Liu Chaochuan Jia Maosheng Fu Xinyu Gao Yubao Zhu Qiqi Zhang

The Kangaroo Escape Optimization Technique (KET) combines escape and safe-area searches, but its fixed stage allocation, restricted guidance range, limited refinement of low-ranked individuals, and insufficient use of population-state information can reduce its performance on complex problems. This study develops a Multi-Strategy Kangaroo Escape Optimization Technique (MSKET) through iteration-dependent stage switching, population-proportion-based candidate guidance, selective greedy DE/rand-to-best/1 refinement, and beta-distribution opposition-based global-best guidance. The contribution lies in assigning established mechanisms to specific KET limitations and coordinating them across different stages and population subsets. MSKET was evaluated through 30 independent runs on the 100-dimensional CEC2017 and CEC2020 suites. It ranked first on 22 of 29 CEC2017 functions and 8 of 10 CEC2020 functions, with the best average rank on both suites. Ablation, diversity, and nonparametric statistical analyses further supported the observed performance gains. MSKET also achieved the best overall repeated-run results on the piston–lever and three-bar truss design problems. For near-infrared prediction, MSKET-BP obtained an R2 of 0.86440, an RMSE of 2.1377, and a MAPE of 5.0083% on the testing set. These results indicate improved search and prediction performance within the examined tasks, although at a higher computational cost than KET.

Biomimetics doi: 10.3390/biomimetics11080586

Authors: Luis Fernando Rosas-Ordaz Estefania Ruiz-Muñoz Saúl Zapotecas-Martínez Leopoldo Altamirano-Robles Raquel Díaz-Hernández José de Jesús Velázquez Arreola

Sinogram images represent projection-domain data acquired during digital breast tomosynthesis (DBT), preserving angular information prior to reconstruction. However, their low contrast and noise-related degradation limit their direct use in downstream tasks. This work proposes a budget-constrained hybrid biomimetic optimization framework for projection-domain contrast enhancement based on the integration of Particle Swarm Optimization (PSO) and the Nelder–Mead (NM) simplex method. The approach combines global exploration with local refinement under a fixed number of objective function evaluations (NFE), enabling fair and computationally efficient comparisons with standalone optimizers. Experiments were conducted on 222 labeled mammographic images (136 benign and 86 malignant), which were transformed into sinograms via the Radon transform. The proposed method achieves competitive performance in terms of PSNR, SSIM, and FSIM, while exhibiting faster convergence and reduced computational cost compared to several state-of-the-art swarm-based approaches. Additionally, the impact of the enhanced sinograms was evaluated through a downstream classification task using a Vision Transformer-based knowledge distillation scheme, demonstrating improved discrimination between benign and malignant cases. These results demonstrate that the proposed budget-constrained hybrid biomimetic strategy provides an effective and computationally efficient solution for contrast enhancement in projection-domain medical imaging.

Biomimetics doi: 10.3390/biomimetics11080584

Authors: Ridvan Firat Cinar

This study proposes a spiral random walk enhanced puma optimizer (SRW-PO) for array antenna design. The method preserves the hunting and ambush search structure of the classical puma optimizer (PO) while embedding two complementary mechanisms into its update process. The adaptive random walk operator improves population diversity and enables weak candidate solutions to escape unproductive regions. The spiral local search operator strengthens exploitation around the current best solution. This structure provides a balanced search process for highly nonlinear antenna synthesis problems involving side lobe suppression, null control, and excitation dynamic range. SRW-PO is evaluated on representative linear array antenna design cases and further assessed through a focused CEC2022 benchmark comparison with classical PO, GWO, WOA, and GA. The results indicate stronger convergence, more consistent objective cost reduction, and competitive computational efficiency. In the CEC2022 benchmark, SRW-PO achieves the best average rank and ranks first for most benchmark functions. In antenna synthesis experiments, it provides accurate radiation pattern shaping, effective side lobe reduction, and deep null placement. These findings indicate that the embedded spiral and random walk mechanisms improve the search quality and optimization consistency of classical PO for array antenna design.

Biomimetics doi: 10.3390/biomimetics11080585

Authors: Borut Gregorčič Armin Hadžić Jure Berce Matevž Zupančič Matic Može Iztok Golobič

Natural water-repellent surfaces use hierarchical texture and low surface energy to minimize liquid adhesion, and this principle has inspired engineered superhydrophobic surfaces for passive anti-icing. However, whether such bioinspired water-repellent architectures remain beneficial during freezing and ice detachment depends on the stability of the wetting state and on the interaction between surface texture and coating chemistry. This study evaluates the anti-icing performance of smooth and laser-textured 1050A aluminum surfaces functionalized with different hydrophobic agents. Freezing delay measurements at −18 °C and ice adhesion strength measurements at −20 °C were conducted, together with wettability, surface free energy, roughness, and morphology analyses, to compare different coatings on identical morphologies and to isolate the effect of laser-generated texture for the same coating chemistry. On smooth surfaces, fluorinated alkyl phosphonic acid coating provided the largest reduction in ice adhesion strength, decreasing it by approximately 72% relative to the non-functionalized reference, while alkyl phosphonic acid coating reduced it by approximately 50%. In contrast, polydimethylsiloxane showed the longest freezing delay, with a mean value of 907 s, whereas the fatty acid-based coatings exhibited shorter freezing delays than the bare reference surface. On laser-textured surfaces, all coatings initially produced highly water-repellent wetting states. However, the differences in ice adhesion strength were markedly reduced and no longer followed the same ranking as on smooth surfaces. Polydimethylsiloxane again exhibited the most favorable freezing delay, while fluorinated alkyl phosphonic acid showed the poorest performance on the textured substrate. These results show that the bioinspired superhydrophobic state created by laser texturing does not by itself guarantee improved anti-icing performance, as under icing conditions, texture-mediated wetting, local liquid penetration, condensation or frost formation inside the texture, and mechanical interlocking can dominate over the nominal low-surface-energy chemistry.

Biomimetics doi: 10.3390/biomimetics11080583

Authors: Amanda-Ioana Răduţă Andreea-Ramona Treteanu Octavian Andronic Ștefan Busnatu Roxana Nicoleta Silişte Elena Bălășescu

Myocardial infarction remains a leading cause of heart failure because current reperfusion therapies cannot prevent adverse ventricular remodeling or restore lost cardiomyocytes. Regenerative strategies based on stem cells and extracellular vesicles (EVs) have emerged as promising approaches; however, their clinical efficacy is limited by poor retention, rapid clearance, and the hostile post-infarction microenvironment. This narrative review critically examines the role of biomaterial-assisted delivery systems in enhancing stem cell and EV-based cardiac regeneration, with particular emphasis on the distinction between biomimetic and bioactive biomaterials, mechanisms of action, preclinical and clinical evidence, translational barriers, and emerging regenerative technologies. Current evidence demonstrates that injectable hydrogels, extracellular matrix-derived scaffolds, cardiac patches, conductive biomaterials, and multifunctional delivery platforms improve therapeutic retention, prolong paracrine signaling, and actively modulate inflammation, angiogenesis, fibrosis, and extracellular matrix remodeling, resulting in superior functional recovery compared with conventional delivery approaches in preclinical models. Nevertheless, robust clinical evidence remains limited because few biomaterial-assisted strategies have advanced beyond early-phase studies. Future progress will depend on integrating smart biomaterials with engineered extracellular vesicles, gene editing, and personalized regenerative approaches, together with standardized manufacturing, harmonized regulatory frameworks, and adequately powered clinical trials.

Biomimetics doi: 10.3390/biomimetics11080582

Authors: Jinping Feng Nan Xu Xi Li Zhongtao Fu Zhenhua Xiao Zhenghua Huang

Biomimetic perception systems integrate heterogeneous cues selectively rather than treating all available information as equally reliable. Inspired by this principle, this study proposes QGFace-LLaVA, a multimodal large language model (MLLM)-centered framework for robust face analysis under imperfect metadata. A pretrained MLLM serves as the shared prompt-conditioned visual–language reasoning backbone, while structured side information, including age, gender, confidence cues, and availability indicators, is regulated through task-aware quality estimation, reliability-guided metadata calibration, gated residual correction, and counterfactual metadata reliability regularization (CMRR). Experiments on FER2013, CelebA-40, and UTKFace cover facial expression recognition, facial attribute recognition, and age estimation under clean, noisy, missing, shuffled, naturally erroneous, and counterfactual metadata conditions. The results show that metadata utility depends jointly on task relevance, metadata reliability, and fusion strategy, and that clean-setting gains do not necessarily imply robustness. QGFace-LLaVA reduces harmful dependence on unreliable metadata, while CMRR provides additional stability under corruption and mismatch. Overall, the framework transfers biomimetic selective cue integration into MLLM-based face analysis by treating metadata as reliability-controlled auxiliary evidence rather than a uniformly beneficial input.

Biomimetics doi: 10.3390/biomimetics11080581

Authors: Jinting Zhou Ruiyu Gao Shijie Zhou Fengli Zhang

Automatic semantic parsing of architectural floor plans provides a two-dimensional semantic layer for building information modeling (BIM)-related workflows, intelligent plan checking, renovation, and large-scale drawing management. Unlike natural images, floor plans contain sparse textures, dense linework, small architectural symbols, and strong axis-aligned geometric regularities. These properties make conventional segmentation networks vulnerable to small-symbol dilution during downsampling, noisy skip-feature fusion, and fragmented predictions along long wall boundaries. Inspired by principles of hierarchical visual processing, selective attention, and spatial encoding, this study presents PCP-Net, an end-to-end Planar Component Parsing Network for room, icon, and boundary-aware floor plan parsing. PCP-Net uses a Grouped Residual Encoder to extract multi-scale local patterns, a Cross-Scale Feature Recalibration (CSFR) pipeline to recalibrate skip features through Adaptive Channel Gating, Spatial Response Amplification, and Axis-Decompositional Positional Embedding, and a Structural Gradient Propagation branch to provide training-time boundary regularization. Experiments on CubiCasa5K and three external datasets (R3D, CVC-FP, and ROBIN) evaluate PCP-Net under in-domain and zero-shot cross-domain protocols. On CubiCasa5K, PCP-Net attains 71.3% mIoU, 90.1% overall accuracy, and 78.9% mean accuracy; in the current ablation setting, ADPE improves mIoU by 0.8 percentage points over the ADPE-ablated configuration. These results indicate that cross-scale feature recalibration with axis-aware positional cues can improve floor plan semantic parsing within the evaluated datasets and pixel-level metrics, while downstream BIM generation still requires additional vectorization, topology graph construction, and attribute extraction.

Biomimetics doi: 10.3390/biomimetics11080580

Authors: Hermann Ehrlich Teofil Jesionowski

Extreme biomimetics represents a novel multidisciplinary direction within classical biomimetics and bioinspired materials science that draws inspiration from extremophiles and extreme natural habitats to create unusual approaches for the design of next-generation materials, including composites never reported or predicted before. This review is divided into several sections covering achievements and fundamental challenges in the following fields: sources of inspiration—organisms and locations; biological materials for extreme biomimetics based on examples of biosilica, cellulose, chitin, and structural proteins (collagen, byssus, silk, keratin). Additionally, conchixes of molluscan shell origin, fish scales, and spongin as a skeletal biocomposite of industrial sponge origin are represented and discussed from the viewpoint of extreme biomimetics. Finally, scientifically based but daring experimental decisions in modern extreme biomimetics, taking the examples of extreme carbonization of naturally pre-structured biomaterials, galvanobiomimetics, in-flame biomimetics, pyro-biomimetics, and ultra-high pressure mimetics are represented for the first time as current trends with challenging goals.

Biomimetics doi: 10.3390/biomimetics11080579

Authors: Avinash S Pramod Sejun Park Jeongho Choo Giuk Lee

Cable-driven hip exosuits are designed to deliver assistive torques during locomotion. However, most studies focus on assistance in a single plane, even though hip mechanics are inherently multiplanar. This study investigates whether modulation of proximal and distal cable anchor-point locations can systematically alter the distribution of hip assistive torque across anatomical planes. To characterize and estimate three-dimensional torque, we propose a geometric model that incorporates empirically identified correction terms. The model was evaluated using a hip–thigh phantom across 12 anchor configurations and 17 joint-angle conditions, resulting in 204 experimental cases. The results show that anchor-point modulation systematically redistributes assistive torque among the sagittal, frontal, and transverse planes, demonstrating that changes in cable routing can generate distinct multiplanar torque profiles. The model captured major trends and directional characteristics of the torque, with strongest agreement observed for the Mz component, for which the mean absolute percentage error ranged from 4.7% to 10.1%. Estimation errors increased under conditions of greater cable wrap, reflecting the influence of torsion and compliance effects not explicitly represented. These findings establish anchor-point modulation as a practical method for achieving multiplanar hip assistance and provide a foundational empirical baseline for extending cable-driven assistance beyond conventional sagittal-plane operation.

Biomimetics doi: 10.3390/biomimetics11080578

Authors: Wangshuang Zang Congrong Xiao Dongkwon Seong

Background: Early-stage elderly-care robot design requires biological analogies to be translated without turning qualitative inspiration into unvalidated numerical evidence. Methods: We audited 15 initial variables and retained a four-dimensional exploratory space: nominal shell-edge radius (V06), pre-braking trigger distance (V09), translational speed (V11), and commanded deceleration (V12). Latin hypercube samples were filtered by V09 − V112/(2V12) ≥ 0. A derived warning margin proxy, I5 = 1 − [V112/(2V12)]/V09, was evaluated with fixed-seed feasibility, distribution, coverage, cluster, coordinate stability, and distance-sensitivity diagnostics. Results: The pooled pre-check acceptance rate was 0.8621. I5 descriptive and distributional stability passed at N = 768→1024, but four-dimensional coverage passed in only 10/30 seeds; no sufficient N was established up to 1024. Natural cluster structure was not detected, exact representative coordinates were seed-sensitive, and selection overlap under an alternative distance definition was 0.53. Two fixed-seed points were therefore retained only as illustrative boundaries. AI-assisted images and legacy Rhino studies were used for qualitative design communication, not as validated realizations of the computation. Conclusions: The evidence-tiered workflow supports traceable exclusion of infeasible combinations and transparent product design translation while preserving explicit limits: no physical safety, usability, manufacturing, or optimality claim is made.

Biomimetics doi: 10.3390/biomimetics11080577

Authors: Muhammad Shahid Abdullah Zulaikha Fatima Wasif Feroze Miguel Jesús Torres Ruiz Magdalena Saldaña-Pérez Carlos Guzmán Sánchez-Mejorada Rolando Quintero Tellez

Reinforcement learning (RL) has emerged as a promising approach for autonomous surgical robotic subtasks. Recent advances include deep reinforcement learning (DRL), imitation learning (IL), and vision–language–action (VLA) models. However, current evidence remains fragmented across simulation benchmarks, task-specific demonstrations, and limited clinical studies. Existing reviews primarily focus on RL algorithms, while the broader pathway from algorithm development to clinically deployable surgical autonomy has not been comprehensively synthesised. This PRISMA 2020-guided systematic review examines RL, IL, safe RL, simulation-to-real (sim-to-real) transfer, foundation models, VLA systems, and regulatory readiness in surgical robotics. We searched IEEE Xplore, PubMed/MEDLINE, Embase, Scopus, Web of Science, the Cochrane Library, ACM Digital Library, arXiv, and medRxiv for studies published between January 2015 and March 2026, with additional studies identified through backward citation tracing. Eligible studies proposed novel RL, imitation learning, or foundation-model approaches for surgical robotics with empirical validation in simulation or on physical robotic platforms. Two reviewers independently extracted data using a predefined coding scheme, and a third reviewer resolved disagreements. Owing to substantial heterogeneity in platforms, tasks, and outcome measures, a quantitative meta-analysis was not feasible; therefore, the evidence was synthesised narratively using a comparative framework. A total of 220 studies met the inclusion criteria, covering eleven active surgical RL platforms, seven paired sim-to-real studies, emerging foundation-model architectures, and three FDA-cleared robotic systems exhibiting Level 3 autonomy. Available comparative studies suggest that hierarchical approaches can outperform flat policies in long-horizon tasks, while language-conditioned models demonstrated promising multi-step surgical capabilities. Seven paired simulation-to-real studies were identified, encompassing tissue retraction, guidewire navigation, and surgical cutting tasks. Sim-to-real performance gaps varied substantially by task and metric, with success-rate gaps ranging from −10 to 50 percentage points (negative values indicating better real-world than simulated performance), while paired mean spatial errors differed by at most 0.61 mm. Most studies employed domain randomization or visual domain adaptation; hierarchical reinforcement learning demonstrated advantages over flat policies in multi-step surgical tasks. Explicit safety-constrained methods (CPO, CBF, and SER), formal verification, and regulatory-aligned evaluation were reported in fewer than 3% of applied studies. Most evidence remained simulation-based, with no reported autonomous RL execution in vivo in humans. Overall, RL-based surgical robotics appears mature at the simulation stage but remains preclinical for autonomous clinical deployment. Future progress requires stronger sim-to-real validation, multimodal safety-aware architectures, alignment with IEC 62304, ISO 14971, FDA guidance, and the EU AI Act, and open benchmarks that jointly evaluate performance, safety, and surgeon trust.

Biomimetics doi: 10.3390/biomimetics11080576

Authors: Yoslandy Lazo Broderick Crawford Gino Astorga Felipe Cisternas-Caneo José Barrera-Garcia Giovanni Giachetti Ricardo Soto

Bio-inspired swarm metaheuristic algorithms constitute a widely used tool for solving complex optimization problems. However, the experimental characterization of their emergent behavior remains a methodological challenge. This study proposes an experimental framework for characterizing emergent behavior through swarm collective dynamics. The framework integrates complementary dynamic indicators and establishes relative diversity loss as a homogeneous criterion for defining equivalent comparison states across different search processes. The framework was evaluated using the Reptile Search Algorithm (RSA) and Draco Lizard Optimizer (DLO) as case studies, with Particle Swarm Optimization (PSO) serving as a reference algorithm. The results showed that swarm collective dynamics were associated with both the mathematical properties of the search landscape and the search mechanisms of each metaheuristic. Furthermore, relative diversity loss enabled the comparison of different metaheuristics within a common reference framework. In RSA, swarm reorganization occurred during the first iterations. DLO exhibited a more gradual evolution, whereas PSO showed an intermediate behavior between both dynamics. The proposed experimental framework provides a methodological basis for the experimental characterization of emergent behavior in swarm metaheuristics.

Biomimetics doi: 10.3390/biomimetics11080575

Authors: Sanjana Challa Alynah J. Adams Laken Anderson Bria Fiebiger Rikin Soni Athena Ye Jocelyn Hunt Charlotte Thomas Dorien I. Schonebaum Jose A. Foppiani Umar Choudry Samuel J. Lin

Cartilage regeneration remains a major clinical challenge because cartilage has limited healing capacity and must withstand substantial mechanical loading. Silk fibroin, a cocoon-derived biomaterial with tunable mechanical properties, has emerged as a promising scaffold material for cartilage tissue engineering. This systematic review evaluates the biomechanical performance, fabrication approaches, and regenerative potential of silk fibroin bioprinted scaffolds for cartilage repair. A systematic review was conducted according to PRISMA guidelines. Primary outcomes included scaffold mechanical properties relevant to cartilage function. Secondary outcomes included scaffold composition, fabrication methods, biological performance, clinical applications, and reported limitations. Of 918 identified records, 24 studies published between 2014 and 2025 met the inclusion criteria. Most were preclinical studies using silk fibroin-based hydrogels, often combined with hyaluronic acid, gelatin, collagen, or synthetic polymers. Mechanical properties varied according to scaffold composition and crosslinking methods. Reinforced scaffolds generally demonstrated improved stiffness, structural support, and load distribution. Silk fibroin scaffolds consistently supported chondrogenesis and cartilage-like tissue formation. However, challenges remain, including mechanical mismatch with native cartilage, construct instability, cell loss, and suboptimal degradation profiles. Overall, silk fibroin bioprinted scaffolds demonstrated support for chondrogenesis and cartilage-like tissue formation in preclinical studies for cartilage regeneration. Future research should focus on standardized biomechanical testing, long-term in vivo evaluation, and clinically relevant scaffold designs to facilitate translation toward functional cartilage repair.

Biomimetics doi: 10.3390/biomimetics11080574

Authors: Yi Zhang Changyi Feng Yong Xu

This study proposes an adaptive Support Vector Machine (SVM) classification method based on an enhanced Starfish Optimization Algorithm (SFOAE-SVM) for hyperspectral image (HSI) classification. HSI classification remains a critical challenge in remote sensing due to the high dimensionality of spectral features, spectral mixing, scarcity of labeled samples, and complex land-cover distributions. The SFOAE algorithm is used for global hyperparameter optimization of SVMs, accounting for the distributional characteristics of the target HSI data. The approach aims to improve search capability and reduce the likelihood of convergence to local optima by combining multi-dimensional topology-oriented expansion with global exploration. Experimental results demonstrate that SFOAE-SVM achieves competitive classification accuracy and stable performance compared with conventional SVM parameter selection strategies and other optimization-based methods across three benchmark hyperspectral remote-sensing datasets. These results indicate that the proposed method offers a promising optimization-assisted SVM framework for hyperspectral remote-sensing image classification.

Biomimetics doi: 10.3390/biomimetics11080573

Authors: Meihong Gao Zhenjiang Wei Zhengyang Wu Chengchun Zhang Chun Shen

In engineering fields such as aviation, shipping, and high-speed rail, system operational efficiency and energy consumption are largely determined by turbulent drag. The drag reduction design of shark skin-inspired micro-groove structures have opened up new avenues for improving aerodynamic efficiency, optimizing flow field characteristics, and saving energy, representing a highly promising research hotspot in the field of functional micro-structured surfaces. Addressing the issue of drag reduction for V-shaped grooves (frictional Reynolds number 85–695), this paper employs Design of Experiments (DOE) combined with high-precision numerical simulation to clarify the influence of groove height (h), groove width (s), and inflow velocity (U) on the drag reduction rate for bionic microgroove surfaces. The drag reduction mechanism is further revealed through the analysis of vorticity distribution, vortex core position, boundary layer velocity distributions, pulsating velocity fields, and Reynolds stress distributions. When the dimensionless height h+ and width s+ range from 8.50 to 29.75, these grooves can effectively reduce resistance. A maximum drag reduction rate of 12.33% is achieved at h+ = s+ = 25.29 and a flow velocity of 80.7 m/s (frictional Reynolds number 599). At low flow velocities, larger groove dimensions are favorable for drag reduction. In contrast, smaller groove dimensions are required under medium-to-high flow velocity conditions. The optimal microstructural dimensions of V-shaped grooves decrease as the inflow velocity increases. V-shaped grooves can lift turbulent vortex coherent structures, reduce pulsating velocities in the streamwise, normal, and spanwise directions, and decrease the peak values of Reynolds stress in the near-wall region. The results can provide a quantitative basis for the design and engineering applications of biomimetic riblet drag-reducing surfaces.

Biomimetics doi: 10.3390/biomimetics11080572

Authors: Ding Xu Hanpeng Liu Zijun Hu Shutao Wei Kefeng Wang Zhaoyang Li Caideng Yuan Xiang Ge

Wound healing, particularly chronic wound healing, is a complex and dynamically regulated process that is commonly challenged by wound infection caused by the invasion of various bacteria. The emergence of nanofibers has overcome the limitations of traditional wound dressings in clinical applications, and has further evolved into smart responsive wound dressings with specific capability to perceive both endogenous and exogenous stimuli and produce dynamic responses. Accordingly, this article reviews the latest research advances in smart responsive nanofibers as wound dressings, and systematically elaborates the types of substrate materials and various preparation methods for nanofibers. In particular, the nanofibers prepared by electrospinning (such as ultrasound (US) responsive nanofiber dressing) have demonstrated unique potential in the construction of smart responsive wound dressings due to their high specific surface area and porosity characteristics. Most importantly, the incorporation of different types of stimuli constructs smart responsive nanofiber dressings, which not only provides a novel therapeutic strategy for wound healing, but also outlines future research directions for the development of smart responsive nanofibers.

Biomimetics doi: 10.3390/biomimetics11080571

Authors: Suk-Dae Lim Hyun Cho Sun-Ho Choi Hong-Gun Kim Young-Soon Kim

Inadequate cleaning of flexible endoscope channels remains a major cause of healthcare-associated infections despite established reprocessing protocols. We developed biomimetic carbon balls inspired by mustard seed surface features as a brush-free adjunct for endoscope channel cleaning, combining mild mechanical action with adsorption-based removal of organic debris. Activated carbon (AC) balls and carbon fiber (CF) balls with diameters of 1.8–3.0 mm were fabricated to fit 2.5–3.5 mm working channels and characterized by TGA, BET, SEM, and EDS. Cleaning performance was evaluated using tomato powder residue and microbiological validation, and computational fluid dynamics was used to assess flow behavior in a 2.7 mm channel containing a 2.6 mm ball under suction. CF balls showed more uniform thermal degradation (8.9% residual mass at 1000 °C) and a ribbed fibrous morphology. In contrast, AC balls exhibited porous, irregular structures with strong adsorption capacity, characterized by high thermal stability (86.4–89.3% residual mass) and pore volumes of 0.087–0.108 cm3/g and BET surface areas of 63.00 m2/g and 21.65 m2/g. After residue exposure, AC surfaces retained more particulate matter, while CF surfaces remained cleaner and promoted more pronounced wall interaction. Microbiological testing showed effective decontamination, with bacterial counts reduced below the detection limit. CFD analysis demonstrated pressure-driven gap flow, elevated local velocity, and vortex-induced wall shear stress sufficient to enhance debris removal without apparent channel damage. These results suggest that mustard seed-inspired carbon balls provide a promising brush-free strategy for endoscope reprocessing by integrating adsorption, hydrodynamic agitation, and gentle scouring to improve cleaning safety and efficacy.

Biomimetics doi: 10.3390/biomimetics11080570

Authors: Hao Jiang Yuheng Lin Zhihan Li Liguo Shuai

Adaptive locomotion across continuous terrain transitions remains difficult for hexapod robots because contact timing, body attitude, support height, and load distribution change simultaneously along a route. This paper presents a unified proprioception-driven deep reinforcement learning and central pattern generator (DRL-CPG) framework for terrain-transition locomotion without visual terrain classification, explicit terrain labels, or terrain-specific controller switching. A high-level proximal policy optimization policy maps a 46-dimensional proprioceptive observation to a three-dimensional CPG modulation action comprising oscillation amplitude, swing-phase frequency, and turn modulation. A coupled six-node Hopf oscillator network then expands these modulated parameters into phase-coordinated rhythmic commands, which are mapped to the 18 joint targets of a JetHexa hexapod and executed by a low-level proportional-derivative controller. The observation space contains body linear velocity, body angular velocity, relative joint positions, relative joint velocities, the previous three-dimensional policy action, and inertial measurement unit (IMU)yaw/heading relative to the initial track direction. A continuous route consisting of flat ground, uphill stairs, irregular terrain, downhill stairs, and a recovery segment is defined to evaluate transition-aware locomotion using route completion, velocity-tracking error, lateral deviation, and roll/pitch fluctuation. Compared with the fixed-parameter CPG and end-to-end DRL baselines, the proposed method increased the full-distance success rate at 4.7 m from 9% and 20%, respectively, to 88%, while maintaining smoother velocity, lateral deviation, and roll/pitch responses. The framework preserves the rhythmic prior of CPG control while reducing the exploration burden of reinforcement learning, providing a compact formulation for adaptive hexapod locomotion across terrain transitions.

Biomimetics doi: 10.3390/biomimetics11080569

Authors: Xiangdong Bu Hongru Jiang Tianruo Guo Heng Li Yao Chen

Task-driven recurrent neural networks (RNNs) have been widely employed as tools for investigating neural dynamics in neural motor control research by modeling the motor cortex. RNNs are often implicitly assumed to learn the underlying computational mechanisms in accordance with biological neural circuits. However, the brain network has a highly structured and specific network connectivity and during individual development the motor cortex has acquired a rich repertoire of behavioral primitives via continuous learning of body control. Considering that the task-driven RNNs are often initialized randomly and trained directly on the specific task, how much these models can truly reveal about the motor cortex is still a crucial question awaiting further research. In this study, we propose a method for modeling the motor cortex pretrained on single reaching skills. Specifically, we use an RNN, receiving sensory feedback and task inputs, as the controller to produce motor commands that drive a musculoskeletal arm model. This model can perform reaching movements along a mini-jerk trajectory between arbitrary points in the workspace, prior to training on specific tasks. The model pretrained on single-reach task has more similarity with real neural data both on a neural geometry and neural dynamics level in center-out (CO) and random target touch (RTT) tasks than models directly trained on these tasks. Surprisingly, we observed the opposite pattern in a double-reach (DR) task, in which two targets appeared simultaneously, rather than presenting the next target after the completion of the prior movement as in the RTT task. This suggests that sequential movements are planned as an integrated unit, and this capability may be implemented at the level of motor cortical circuits. In summary, our results suggest that endowing the network with capabilities beyond the immediate task demands—through more systematic training or other methods—can help better understand the dynamics of biological neural circuits.

Biomimetics doi: 10.3390/biomimetics11080568

Authors: Emre Bendeş

Population-based bio-inspired metaheuristics are the dominant tools for tuning hybrid decomposition–deep-learning forecasters, yet their relative behavior on a common problem is rarely assessed with a leakage-free, physically meaningful protocol. We benchmark eight metaheuristics on the joint nine-dimensional hyperparameter optimization of a discrete-wavelet-transform bidirectional-LSTM (DWT–BiLSTM) architecture for short-term wind speed forecasting, using 409,152 hourly observations from eight meteorological stations. The set comprises six nature-inspired methods (Artificial Bee Colony, ABC; genetic algorithm, GA; Particle Swarm Optimization, PSO; Grey Wolf Optimizer, GWO; Hippopotamus Optimization, HO; and the Raindrop Optimizer) together with two recent metaphor-free or social variants (the Farthest-better Nearest-worse Optimizer, FNO; and the Tuckman Optimization Algorithm, TOA). A multi-stage protocol covers 30 independent runs per algorithm, a joint-versus-sequential comparison, a genuine rolling-origin out-of-sample evaluation, and component ablation. Friedman testing reveals significant differences (χ2 = 49.76; p < 10−8), with the Grey Wolf Optimizer attaining the best mean rank (2.27) and Pareto-dominant run-time; ablation shows the DWT front-end is essential (Cohen’s d = 13.09) and bidirectionality negligible at the one-hour horizon (p = 0.674). Critically, evaluating forecasts in reconstructed physical units reveals that the per-component advantage does not persist: at the one-hour horizon the reconstructed forecast does not exceed a naive persistence baseline (skill ≈ −0.5 in m/s versus +0.44 in normalized component space), a discrepancy independent of decomposition leakage that we report transparently. This work thus contributes a rigorous, leakage-controlled bio-inspired benchmark and a cautionary evaluation methodology.

Biomimetics doi: 10.3390/biomimetics11080567

Authors: Jiawen Wang Jiayue Cai Xuefei Xie Yang Shen Fanxing Meng Yanxiu Yu Dongman Cao

Sand Cat Swarm Optimization (SCSO) is a swarm intelligence algorithm characterized by a simple structure and a small number of control parameters. However, when solving complex optimization problems, SCSO suffers from several limitations, including an uneven initial population distribution, excessive dependence on the current best individual during the search process, insufficient local exploitation accuracy, and susceptibility to local optima. To address these limitations, a Collaborative Multi-Strategy Sand Cat Swarm Optimization algorithm (CMSCSO) is proposed. The good point set method is adopted to generate a uniformly distributed initial population. An adaptive random reuse strategy is designed to selectively inherit dimensional information from the best individual according to differences in individual fitness. A prey impatience coefficient is introduced to dynamically adjust the local search intensity according to the distance between the population and the current best solution. In addition, a refractive-mechanism-based opposition-based learning strategy for the worst individuals is incorporated to update low-quality individuals and improve the ability of the algorithm to escape from local optima. CMSCSO was evaluated using the 30-dimensional CEC2017 and 10-dimensional CEC2022 benchmark suites. Its performance was compared with that of SCSO and several recently developed metaheuristic algorithms. The experimental results show that CMSCSO achieved the best mean values on 24 of the 29 CEC2017 benchmark functions and on 10 of the 12 CEC2022 benchmark functions. In the Wilcoxon tests conducted on CEC2017 and CEC2022, CMSCSO achieved 220 and 90 statistically significant wins, respectively. It also ranked first in the Friedman tests for both benchmark suites. For engineering optimization problems, the results obtained from six types of engineering design problems demonstrate that CMSCSO can consistently obtain high-quality feasible solutions that satisfy the specified constraints. In two-dimensional and three-dimensional wireless sensor network coverage optimization problems, coverage rates of 96.30% and 89.54% were achieved. For photovoltaic model parameter identification, CMSCSO achieved the highest identification accuracy. The numerical and engineering test results demonstrate that CMSCSO provides high optimization accuracy, strong stability, and good adaptability to complex engineering problems. It can therefore serve as an effective solution method for optimization tasks in structural design, mechanical engineering, and other related fields.

Biomimetics doi: 10.3390/biomimetics11080566

Authors: José Pérez-Rigueiro Atocha Guedán-Durán Fivos Panetsos Gianna Arencibia Gustavo V. Guinea Luis Colchero Miriam Quero Jaime Espinosa Alessandro Rizzi Tando Maduna Anna Pancho Marsela Hakani Andreas Vlachos Julia Sepúlveda-Díaz Alan Morin Michele Papa Giovanni Cirillo Assunta Virtuoso Ciro De Luca

The European Pathfinder THOR project envisages the creation of a vascularized fragment of tissue that can be implanted in a patient using regenerated silk fibers as its building blocks. The selection of regenerated silk as the main building block of the scaffold relies heavily on its outstanding biocompatibility in comparison with either other artificial polymeric fibers or even natural silk fibers. Additionally, regenerated fibers produced through the Dynamic Dope Destabilization Spinning (D3STM) process are shown to exhibit high mechanical performance as reflected in values of strain at breaking and work to fracture comparable to those of the natural material. It is further shown that these fibers are endowed with the unique property of self-adhesion whereby hydrated fibers attach to one another and may sustain detachment forces of up to a few tens of MPa, a property that facilitates the generation of the scaffold with the fibers as its basic building block. Lastly, regenerated silk fibers are shown to be efficiently decorated with either peptides or small proteins, such as the vascular endothelial growth factor (VEGF), or with antibodies. The performance of both non-functionalized and decorated silk fibers is assessed in two different in vitro biological systems: (1) endothelial cell cultures, and (2) organotypic brain slice cultures. Together, these results support the use of regenerated silk fibers as versatile building blocks for biofunctional tissue scaffolds and provide experimental validation of the tissue engineering strategy established by the THOR project.

Biomimetics doi: 10.3390/biomimetics11080565

Authors: Kaixin Lan Lei Jiang Yucheng Tao Chaojie Fu Yongbin Jin Hongtao Wang

Existing quadruped robots typically place their center of mass (CoM) near the geometric center of the body to achieve structural symmetry and simplify control design. In contrast, many quadrupedal animals capable of agile running exhibit a pronounced anteriorly biased mass distribution, with the CoM located closer to the front of the body. This biological characteristic motivates a re-examination of whether a geometrically centered CoM necessarily corresponds to dynamically balanced loading between the fore- and hindlimbs during high-speed locomotion. To address this question, this study investigates the influence of longitudinal CoM position on load distribution during high-speed straight-line locomotion of quadruped robots. A unified analytical framework is established by combining whole-body force and pitch moment equilibrium, sagittal-plane kinematics, and Jacobian-based force-to-torque mapping, thereby linking longitudinal CoM position, foot-end support forces, and joint loads. Simulation validation is conducted on the Black Panther 2 quadruped robot using four central-body CoM configurations, denoted as ×0, ×5, ×10, and ×15. In the primary evaluation at 5 m/s, shifting the CoM forward from ×0 to ×15 reduces the absolute median fore–hindlimb differences in support force and joint torque by approximately 86.6% and 93.4%, respectively, indicating a transition from hindlimb-dominated loading toward cooperative load sharing between the fore and hindlimbs. Independent training runs with multiple random seeds further confirm the robustness of this load-redistribution trend to reinforcement learning variability. Consistent behavior is also observed at 8 m/s, while no evident degradation in turning response or locomotion stability is found under the tested turning and randomly generated rough-terrain conditions. These results demonstrate that a moderate forward shift of the longitudinal CoM can alleviate hindlimb load concentration and promote a more balanced fore–hindlimb load distribution, providing a theoretical basis for the morphological design and control optimization of high-speed quadruped robots.

Biomimetics doi: 10.3390/biomimetics11080564

Authors: Yoann Matagne Guillaume Marchal Damien Coibion Sébastien Blasutig Fanny Lambert Frederic Boschini Rudi Cloots Nicolas Somers

As the population ages, the demand for customizable, resorbable bone implants in tissue engineering has intensified, outstripping the limitations of traditional autografts and allografts. While silicate-based bioactive glasses dominate bioactive glass research, borate-based bioactive glasses (BBGs) present distinct biomimetic advantages due to their accelerated degradation kinetics and superior ion-release profiles. However, producing highly pure, homogeneous BBG powders tailored for additive manufacturing remains a severe bottleneck. This study reports the development of a highly efficient synthesis protocol and subsequent Digital Light Processing (DLP) 3D printing of BBG scaffolds. An aqueous-based precursor mixture was processed via spray drying and a customized multi-stage thermal pretreatment sequence up to 800 °C to mitigate material loss, minimize oxide evaporation, and completely eliminate carbonates. Subsequent “flash melting” at 1150 °C for 20 min yielded an amorphous, high-purity borate–phosphate glass network (68.1B2O3-3.8Na2O-18.9CaO-4.9MgO-4.3P2O5, in wt%). Differential scanning calorimetry (DSC) revealed a glass transition temperature (Tg) of 625 °C, while in situ X-ray diffraction localized the onset of crystal nucleation between 706 °C and 723 °C. Following fine planetary milling to achieve a highly dense particle packing distribution (Dv50 = 5.4 µm, Dn50 = 0.6 µm), the optimized BBG powder was successfully loaded into an acrylate-based photosensitive slurry (51.2 wt% solid loading) to manufacture complex 3D biomimetic gyroid scaffolds via DLP. While the structural feasibility of printing high-resolution gyroid porous architectures is validated, post-printing evaluation highlighted a narrow thermal processing window; sintering at 660 °C optimized particle coalescence while minimizing microstructural de-densification caused by closed porosity expansion (which reaches 48.4% at 675 °C). This scalable synthesis-to-printing workflow offers a crucial steppingstone toward next-generation fully resorbable bone tissue scaffolds.

Biomimetics doi: 10.3390/biomimetics11080563

Authors: Sophie C. Liegenfeld Niklas P. Straub Nicolas Krueger Mandy Dittmer Arianna Delle Coste Jan T. Strenge Markus Geissen Sophie C. Rhode Wolfgang R. Streit Ralf Smeets Ewa K. Stuermer

Biodegradable biomaterials are promising candidates for regenerative wound care, yet their performance under infection-driven conditions remains poorly understood. This study evaluated the regenerative efficacy and structural persistence of a silk fibroin membrane and electrospun nonwoven matrix using a human ex vivo full-thickness skin wound model under non-infected and bacterially infected conditions. Complementary in vitro degradation assays assessed matrix durability following exposure to Staphylococcus aureus, Pseudomonas aeruginosa, bacterial culture supernatants and clinically relevant antiseptic solutions. In non-infected wound conditions, both matrices enhanced wound regeneration, resulting in increased re-epithelialization and proliferative activity compared with untreated controls; membrane-treated wounds achieved approximately 95% re-epithelialization after 15 days compared with approximately 20% in untreated controls, indicating accelerated wound closure, whereas nonwoven matrices supported sustained cellular proliferation over time. In contrast, bacterial infection was associated with almost complete absence of re-epithelialization and proliferative activity irrespective of matrix architecture. Matrix persistence was pathogen-dependent. While S. aureus induced moderate degradation, P. aeruginosa caused pronounced structural deterioration in ex vivo and in vitro models. Exposure to P. aeruginosa culture supernatants produced similar effects. These findings demonstrate that regenerative efficacy and material persistence are distinct biomaterial properties that are strongly influenced by the wound microbiological environment and are profoundly compromised under conditions of high bacterial burden.