Next Issue
Volume 10, May
Previous Issue
Volume 10, March
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
3.5
3.4
Journals
Biomimetics
Volume 10
Issue 4
share announcement

Biomimetics, Volume 10, Issue 4 (April 2025) – 63 articles

Cover Story (view full-size image):  
  • Issues are regarded as officially published after their release is announced to the table of contents alert mailing list.
  • You may sign up for e-mail alerts to receive table of contents of newly released issues.
  • PDF is the official format for papers published in both, html and pdf forms. To view the papers in pdf format, click on the "PDF Full-text" link, and use the free Adobe Reader to open them.
Order results
Result details
Section
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
19 pages, 4347 KiB  
Article
Optimization Method of Human Posture Recognition Based on Kinect V2 Sensor
by Hang Li, Hao Li, Ying Qin and Yiming Liu
Biomimetics 2025, 10(4), 254; https://doi.org/10.3390/biomimetics10040254 - 21 Apr 2025
Viewed by 118
Abstract
Human action recognition aims to understand human behavior and is crucial in enhancing the intelligence and naturalness of human–computer interaction and bionic robots. This paper proposes a method to improve the complexity and real-time performance of action recognition by combining the Kinect sensor [...] Read more.
Human action recognition aims to understand human behavior and is crucial in enhancing the intelligence and naturalness of human–computer interaction and bionic robots. This paper proposes a method to improve the complexity and real-time performance of action recognition by combining the Kinect sensor with the OpenPose algorithm, the Levenberg–Marquardt (LM) algorithm, and the Dynamic Time Warping (DTW) algorithm. First, the Kinect V2 depth sensor is used to capture color images, depth images, and 3D skeletal point information from the human body. Next, the color image is processed using OpenPose to extract 2D skeletal point information, which is then mapped to the depth image to obtain 3D skeletal point information. Subsequently, the LM algorithm is employed to fuse the 3D skeletal point sequences with the sequences obtained from Kinect, generating stable 3D skeletal point sequences. Finally, the DTW algorithm is utilized to recognize complex movements. Experimental results across various scenes and actions demonstrate that the method is stable and accurate, achieving an average recognition rate of 95.94%. The method effectively addresses issues, such as jitter and self-occlusion, when Kinect collects skeletal points. The robustness and accuracy of the method make it highly suitable for application in robot interaction systems. Full article
►▼ Show Figures

Graphical abstract

16 pages, 6610 KiB  
Article
Numerical Simulation of a Bird-Inspired UAV Which Turns Without a Tail Through Proverse Yaw
by Wee-Beng Tay, Timothy Shawn Jie-Sheng Chong, Jia-Qiang Chan, Woei-Leong Chan and Boo-Cheong Khoo
Biomimetics 2025, 10(4), 253; https://doi.org/10.3390/biomimetics10040253 - 21 Apr 2025
Viewed by 110
Abstract
This study numerically explores a bird-inspired tail-less unmanned aerial vehicle (UAV) design which can turn through proverse yaw by using a bell-shaped spanload wing configuration. The research methodology consists of two phases. In the first phase, the objective is to use computational fluid [...] Read more.
This study numerically explores a bird-inspired tail-less unmanned aerial vehicle (UAV) design which can turn through proverse yaw by using a bell-shaped spanload wing configuration. The research methodology consists of two phases. In the first phase, the objective is to use computational fluid dynamics (CFD) simulations to validate that the bell-shaped spanload wing configuration produces proverse yaw, instead of adverse yaw, similar to other typical wing configurations. This allows the UAV to turn without a tail. The solver used is OpenFOAM and a special self-written routine is used to allow the grid to move together with the UAV, which has six degrees-of-freedom (6DOFs) to translate and rotate when its ailerons deflect after reaching steady motion. In the second phase, we investigate the effect of the sweep angle on the proverse yaw. Results show that proverse yaw is indeed produced due to the bell-shaped spanload wing configuration, as CFD simulation shows the UAV turning after aileron deflection. The effect of the sweep angle is more profound on the proverse yaw as simulations show that increasing the sweep angle by 10° increases the turning effect slightly, but decreasing it by 10° instead results in adverse yaw. These findings will have important implications for improving aircraft efficiencies and the development of wing designs. Full article
(This article belongs to the Special Issue Bioinspired Flapping Wing Aerodynamics: Progress and Challenges)
►▼ Show Figures

Figure 1

20 pages, 11980 KiB  
Article
A Bagworm-Inspired Robot That Acquires Its Exterior from External Environments
by Noriko Ishida and Mitsuharu Matsumoto
Biomimetics 2025, 10(4), 252; https://doi.org/10.3390/biomimetics10040252 - 20 Apr 2025
Viewed by 134
Abstract
In this research, we propose a bagworm-inspired robot that can acquire its exterior by incorporating various objects from the surrounding environment into its skin. This study was inspired by the bagworm, the larva of the giant bagworm moth, which wraps itself around straw [...] Read more.
In this research, we propose a bagworm-inspired robot that can acquire its exterior by incorporating various objects from the surrounding environment into its skin. This study was inspired by the bagworm, the larva of the giant bagworm moth, which wraps itself around straw and other materials to use as a nest. When the robot is active outdoors, it is surrounded by natural materials such as sand, fallen leaves, and pieces of wood, and can change its skin by attaching or detaching these materials as needed. In the previous study, the authors developed a camouflage robot that assimilates with the outside world by incorporating natural environmental sand. In this study, by using a water-soluble adhesive as the adhesive material, it is possible to take in a larger number of external substances than before. We also conducted experiments with natural materials, including leaves and pebbles, and confirmed that the robot could pick them up. We expect that by developing these functions, robots will not only have camouflage capabilities but also the ability to reinforce their own skin like bagworms. Full article
(This article belongs to the Special Issue Biorobotics: Challenges and Opportunities)
►▼ Show Figures

Figure 1

17 pages, 20945 KiB  
Article
Ultrastructural Changes in Final Instar Larvae of Papilio polytes (Lepidoptera: Papilionidae) Lead to Differences in Epidermal Spreading of Water and Adjuvants
by Zhengyu Lu, Xue Wu, Tingting Zhang and Chufei Tang
Biomimetics 2025, 10(4), 251; https://doi.org/10.3390/biomimetics10040251 - 19 Apr 2025
Viewed by 208
Abstract
Papilio polytes is a cosmopolitan Lepidoptera species of controversial use and management. It remained unclear how its epidermal ultrastructure changes during development and how this affects its wetting properties in relation to water and pesticide adjuvants. In this study, the epidermis of P. [...] Read more.
Papilio polytes is a cosmopolitan Lepidoptera species of controversial use and management. It remained unclear how its epidermal ultrastructure changes during development and how this affects its wetting properties in relation to water and pesticide adjuvants. In this study, the epidermis of P. polytes was systematically examined at the important feeding stage (from 3rd to 5th instar). Its ultrastructure was quantitatively observed by scanning electron microscopy. Its wetting properties towards the three main types of adjuvants and water were evaluated by contact angle. The chemical functional group differences between different instars and different adjuvant treatments were analyzed by mid-infrared spectroscopy. The correlation between the ultrastructural deformation and variations in wetting properties was verified by simulation tests. It was found that the complication of the epidermal structure was the leading factor for the significant increase in hydrophobicity during development. Cationic adjuvants had the best infiltrating effect on complex epidermal structures and organosilicon adjuvants had the best infiltrating effect on simple epidermal structures. The results provide data for biomimetic design for different wetting properties and suggest the feasibility and advantages of selecting pesticide adjuvants based on developmental changes in the structural characteristics of the insect epidermis. Full article
(This article belongs to the Special Issue Functional Morphology and Biomimetics: Learning from Insects)
►▼ Show Figures

Figure 1

22 pages, 6181 KiB  
Article
Research on Adhesion Performance of Track Monomer with Bionic Structure
by Sanling Fu, Xiahua Cui, Le Yang, Xinyue Wang, Zhijun Guo and Fu Zhang
Biomimetics 2025, 10(4), 250; https://doi.org/10.3390/biomimetics10040250 - 18 Apr 2025
Viewed by 160
Abstract
Goats can walk freely and flexibly in complex environments such as concave and convex or soft ground. And their flexible spine has functions such as adjusting balance and providing auxiliary power during movement, while the limbs only have support functions. The spine has [...] Read more.
Goats can walk freely and flexibly in complex environments such as concave and convex or soft ground. And their flexible spine has functions such as adjusting balance and providing auxiliary power during movement, while the limbs only have support functions. The spine has an adjustable and decisive role in the pressure on the sole of the hoof of the goat. Therefore, the goat spine was taken as the bionic prototype, the three-dimensional force distribution of the goat body space was analyzed, and the optimal spinal space curve was explored, combined with the goat gait cycle. Based on the study of spinal curve arrangement and placement, the spinal curve was stretched along the grouser length direction. The soil contact surface structure of the track monomer was constructed based on functional simulation. And the bionic structure of the track monomer with superior adhesion performance was explored. The results of simulation analysis and soil tank test both showed that the attachment performance of bionic structure was better than that of an ordinary structure. It showed that adding bionic curves to the contact surface of the track monomer could significantly improve the adhesion performance, and the bionic structure with a single bionic curve arranged on the complete contact surface of the track monomer had the best adhesion performance. Moreover, the adhesion of the optimal track monomer bionic structure was increased by 19.22 N compared with an ordinary structure in the soil tank test, which verified the superiority of the track monomer bionic structure design. It provides a new method and a new idea for improving the adhesion performance of tracked vehicle in hilly areas. Full article
►▼ Show Figures

Figure 1

14 pages, 10218 KiB  
Article
An Evaluation of the Performance of Low-Cost Resin Printers in Orthodontics
by Fırat Oğuz and Sabahattin Bor
Biomimetics 2025, 10(4), 249; https://doi.org/10.3390/biomimetics10040249 - 18 Apr 2025
Viewed by 195
Abstract
Background/Objectives: This study evaluated the trueness and precision of three low-cost 3D printers compared to a professional-grade printer in fabricating orthodontic models. Methods: Two upper dental models, one crowded and one non-crowded, were designed using Blenderfordental and Autolign. The models were printed with [...] Read more.
Background/Objectives: This study evaluated the trueness and precision of three low-cost 3D printers compared to a professional-grade printer in fabricating orthodontic models. Methods: Two upper dental models, one crowded and one non-crowded, were designed using Blenderfordental and Autolign. The models were printed with Anycubic M3 Premium, Anycubic Photon D2, Phrozen Sonic Mini 8K, and Ackuretta Sol at 45° and 90° using Elegoo orthodontic and Ackuretta Curo resins. A total of 384 models were produced: 256 crowded (128 at 90° and 128 at 45°) and 128 non-crowded (all at 45°). Chitubox Dental Slicer and ALPHA AI slicer were used for slicing. Post-processing involved cleaning with Ackuretta Cleani and curing in Ackuretta Curie. The models were scanned with Smartoptics Vinyl Open Air. Trueness was assessed using RMS deviation analysis in CloudCompare and linear measurements. Results: One-way ANOVA showed significant differences in trueness among the printers at 45° (p < 0.001) and 90° (p < 0.001). The Ackuretta Sol (LCD) exhibited the highest trueness, with the lowest mean RMS values at 45° (0.095 ± 0.008 mm) and 90° (0.115 ± 0.010 mm). The Anycubic M3 Premium (LCD) had the lowest trueness, with RMS values at 45° (0.136 ± 0.015 mm) and 90° (0.149 ± 0.012 mm). The 45° build angle resulted in significantly better trueness than 90° (p < 0.001). In linear measurements, deviations exceeding 0.25 mm were observed only in the R1 distance, except for the Ackuretta SOL, which remained below this threshold. Conclusions: The professional-grade printer demonstrated the best performance overall. Printing at a 45° build angle resulted in improved accuracy. Despite differences among devices, all printers produced results within clinically acceptable limits for orthodontic use. Full article
►▼ Show Figures

Graphical abstract

17 pages, 7106 KiB  
Article
Effect of Biomimetic Fish Scale Texture on Reciprocating Friction Pairs on Interfacial Lubricating Oil Transport
by Tao Sun, Zhijun Yan, Lixia Xue, Yuanyuan Jiang and Shibo Wu
Biomimetics 2025, 10(4), 248; https://doi.org/10.3390/biomimetics10040248 - 17 Apr 2025
Viewed by 212
Abstract
Focusing on the difficulty of lubrication in the scavenging port area of a cylinder liner of an actual marine two-stroke diesel engine, the transportation of interface lubricating oil was studied. In this paper, a biomimetic fish scale texture composed of fan-shaped and arc-shaped [...] Read more.
Focusing on the difficulty of lubrication in the scavenging port area of a cylinder liner of an actual marine two-stroke diesel engine, the transportation of interface lubricating oil was studied. In this paper, a biomimetic fish scale texture composed of fan-shaped and arc-shaped curves is designed, and the numerical simulation model is established according to this texture. Through simulation research, the variation rules of pressure distribution, interfacial velocity, and outlet volume flow rate on the biomimetic fish scale texture surface at different velocities and temperatures are obtained. Moreover, the biomimetic fish scale texture is machined on the surface of a reciprocating friction pair by laser etching, and the oil transport speed of the interface is tested under different conditions. The results show that the existence of the biomimetic fish scale texture on the friction pair can effectively improve the pressure difference between interfaces during reciprocating motion. The pressure difference enhances the flow properties of interfacial lubricating oil, thereby improving its mass transport capacity. In addition, increasing the movement speed and oil temperature can increase the oil transport speed of interfacial lubricating oil. The results of the experiment suggest that, under continuous and discontinuous interface conditions, compared with a friction pair without texture, the improvement rate of the lubricating oil transport speed at the interface of the friction pair with the biomimetic fish scale texture can reach 40.7% and 69.1%, respectively. Full article
(This article belongs to the Section Biomimetic Surfaces and Interfaces)
►▼ Show Figures

Figure 1

16 pages, 3024 KiB  
Article
Establishing a Xanthan Gum–Locust Bean Gum Mucus Mimic for Cystic Fibrosis Models: Yield Stress and Viscoelasticity Analysis
by Rameen Taherzadeh, Nathan Wood, Zhijian Pei and Hongmin Qin
Biomimetics 2025, 10(4), 247; https://doi.org/10.3390/biomimetics10040247 - 17 Apr 2025
Viewed by 171
Abstract
Airway mucus plays a critical role in respiratory health, with diseases such as cystic fibrosis (CF) being characterized by mucus that exhibits increased viscosity and altered viscoelasticity. In vitro models that emulate these properties are essential for understanding the impact of CF mucus [...] Read more.
Airway mucus plays a critical role in respiratory health, with diseases such as cystic fibrosis (CF) being characterized by mucus that exhibits increased viscosity and altered viscoelasticity. In vitro models that emulate these properties are essential for understanding the impact of CF mucus on airway function and for the development of therapeutic strategies. This study characterizes a mucus mimic composed of xanthan gum and locust bean gum, which is designed to exhibit the rheological properties of CF mucus. Mucus concentrations ranging from 0.07% to 0.3% w/v were tested to simulate different states of bacterial infection in CF. Key rheological parameters, including yield stress, storage modulus, loss modulus, and viscosity, were measured using an HR2 rheometer with strain sweep, oscillation frequency, and flow ramp tests. The results show that increasing the concentration enhanced the mimic’s elasticity and yield stress, with values aligning with those reported for CF mucus in pathological states. These findings provide a quantitative framework for tuning the rheological properties of mucus in vitro, allowing for the simulation of CF mucus across a range of concentrations. This mucus mimic is cost-effective, readily cross-linked, and provides a foundation for future studies examining the mechanobiological effects of mucus yield stress on epithelial cell layers, particularly in the context of bacterial infections and airway disease modeling. Full article
(This article belongs to the Special Issue Mechanical Properties and Functions of Bionic Materials/Structures)
►▼ Show Figures

Figure 1

19 pages, 7825 KiB  
Article
Jump Control Based on Nonlinear Wheel-Spring-Loaded Inverted Pendulum Model: Validation of a Wheeled-Bipedal Robot with Single-Degree-of-Freedom Legs
by Jingsong Gao, Hongzhe Jin, Liang Gao, Yanhe Zhu, Jie Zhao and Hegao Cai
Biomimetics 2025, 10(4), 246; https://doi.org/10.3390/biomimetics10040246 - 17 Apr 2025
Viewed by 195
Abstract
Jumping is a fundamental capability for wheeled-bipedal robots (WBRs) navigating unstructured terrains, with jump height and stability serving as indicators of the robot’s environmental adaptability. However, existing trajectory planning methods demand high output capacity from the joints and fail to balance computational load [...] Read more.
Jumping is a fundamental capability for wheeled-bipedal robots (WBRs) navigating unstructured terrains, with jump height and stability serving as indicators of the robot’s environmental adaptability. However, existing trajectory planning methods demand high output capacity from the joints and fail to balance computational load with trajectory tracking performance. This limitation hinders most robots from experimental validation. To address these challenges, this study presents an optimized virtual model, trajectory planning strategy, and control method. These solutions enhance both the height and stability of jumps while ensuring real-time execution on physical robots. Firstly, inspired by the human jumping mechanism, a Nonlinear Wheel-Spring-Loaded Inverted Pendulum (NW-SLIP) model was originally proposed as the virtual model for trajectory planning. The jump height is increased by 3.4 times compared to the linear spring model. Then, cost functions are established based on this virtual model, and the trajectory for each stage is iteratively optimized using Quadratic Programming (QP) and a bisection method. This leads to a 21.5% increase in the maximum jump height while reducing the peak joint torque by 14% at the same height. This significantly eases execution and enhances the robot’s trajectory-tracking ability. Subsequently, a leg statics model is introduced alongside the kinematics model to map the relationship between the virtual model and joint space. This approach improves trajectory tracking performance while circumventing the intricate calculation of the dynamics model, thereby enhancing jump consistency and stability. Finally, the proposed trajectory planning and jump control method is validated through both simulations and real-world experiments, demonstrating its feasibility and effectiveness in practical robotic applications. Full article
►▼ Show Figures

Graphical abstract

22 pages, 4865 KiB  
Article
An Unsupervised Fusion Strategy for Anomaly Detection via Chebyshev Graph Convolution and a Modified Adversarial Network
by Hamideh Manafi, Farnaz Mahan and Habib Izadkhah
Biomimetics 2025, 10(4), 245; https://doi.org/10.3390/biomimetics10040245 - 17 Apr 2025
Viewed by 223
Abstract
Anomalies refer to data inconsistent with the overall trend of the dataset and may indicate an error or an unusual event. Time series prediction can detect anomalies that happen unexpectedly in critical situations during the usage of a system or a network. Detecting [...] Read more.
Anomalies refer to data inconsistent with the overall trend of the dataset and may indicate an error or an unusual event. Time series prediction can detect anomalies that happen unexpectedly in critical situations during the usage of a system or a network. Detecting or predicting anomalies in the traditional way is time-consuming and error-prone. Accordingly, the automatic recognition of anomalies is applicable to reduce the cost of defects and will pave the way for companies to optimize their performance. This unsupervised technique is an efficient way of detecting abnormal samples during the fluctuations of time series. In this paper, an unsupervised deep network is proposed to predict temporal information. The correlations between the neighboring samples are acquired to construct a graph of neighboring fluctuations. The extricated features related to the temporal distribution of the time samples in the constructed graph representation are used to impose the Chebyshev graph convolution layers. The output is used to train an adversarial network for anomaly detection. A modification is performed for the generative adversarial network’s cost function to perfectly match our purpose. Thus, the proposed method is based on combining generative adversarial networks (GANs) and a Chebyshev graph, which has shown good results in various domains. Accordingly, the performance of the proposed fusion approach of a Chebyshev graph-based modified adversarial network (Cheb-MA) is evaluated on the Numenta dataset. The proposed model was evaluated based on various evaluation indices, including the average F1-score, and was able to reach a value of 82.09%, which is very promising compared to recent research. Full article
(This article belongs to the Special Issue Biomimicry for Optimization, Control, and Automation: 3rd Edition)
►▼ Show Figures

Figure 1

30 pages, 5159 KiB  
Article
Snake Optimization Algorithm Augmented by Adaptive t-Distribution Mixed Mutation and Its Application in Energy Storage System Capacity Optimization
by Yinggao Yue, Li Cao, Changzu Chen, Yaodan Chen and Binhe Chen
Biomimetics 2025, 10(4), 244; https://doi.org/10.3390/biomimetics10040244 - 16 Apr 2025
Viewed by 249
Abstract
To address the drawbacks of the traditional snake optimization method, such as a random population initialization, slow convergence speed, and low accuracy, an adaptive t-distribution mixed mutation snake optimization strategy is proposed. Initially, Tent-based chaotic mapping and the quasi-reverse learning approach are [...] Read more.
To address the drawbacks of the traditional snake optimization method, such as a random population initialization, slow convergence speed, and low accuracy, an adaptive t-distribution mixed mutation snake optimization strategy is proposed. Initially, Tent-based chaotic mapping and the quasi-reverse learning approach are utilized to enhance the quality of the initial solution and the population initialization process of the original method. During the evolution stage, a novel adaptive t-distribution mixed mutation foraging strategy is introduced to substitute the original foraging stage method. This strategy perturbs and mutates at the optimal solution position to generate new solutions, thereby improving the algorithm’s ability to escape local optima. The mating mode in the evolution stage is replaced with an opposite-sex attraction mechanism, providing the algorithm with more opportunities for global exploration and exploitation. The improved snake optimization method accelerates convergence and improves accuracy while balancing the algorithm’s local and global exploitation capabilities. The experimental results demonstrate that the improved method outperforms other optimization methods, including the standard snake optimization technique, in terms of solution robustness and accuracy. Additionally, each improvement technique complements and amplifies the effects of the others. Full article
(This article belongs to the Special Issue Bio-Inspired Optimization Algorithms and Designs for Engineering Applications: 3rd Edition)
►▼ Show Figures

Figure 1

29 pages, 26512 KiB  
Article
Radar-Based Activity Recognition in Strictly Privacy-Sensitive Settings Through Deep Feature Learning
by Giovanni Diraco, Gabriele Rescio and Alessandro Leone
Biomimetics 2025, 10(4), 243; https://doi.org/10.3390/biomimetics10040243 - 15 Apr 2025
Viewed by 268
Abstract
Human activity recognition in privacy-sensitive environments, such as bathrooms, presents significant challenges due to the need for non-invasive and anonymous monitoring. Traditional vision-based methods raise privacy concerns, while wearable sensors require user compliance. This study explores a radar-based approach for recognizing the activities [...] Read more.
Human activity recognition in privacy-sensitive environments, such as bathrooms, presents significant challenges due to the need for non-invasive and anonymous monitoring. Traditional vision-based methods raise privacy concerns, while wearable sensors require user compliance. This study explores a radar-based approach for recognizing the activities of daily living in a bathroom setting, utilizing a BGT60TR13C Xensiv 60 GHz radar, manufactured by Infineon Technologies AG (Munich, Germany, EU), to classify human movements without capturing identifiable biometric features. A dataset was collected from seven volunteers performing ten activities which are part of daily living, including activities unique to bathroom environments, such as face washing, teeth brushing, dressing/undressing, and resting on the toilet seat. Deep learning models based on pre-trained feature extractors combined with bidirectional long short-term memory networks were employed for classification. Among the 16 pre-trained networks evaluated, DenseNet201 achieved the highest overall accuracy (97.02%), followed by ResNet50 (94.57%), with the classification accuracy varying by activity. The results highlight the feasibility of Doppler radar-based human activity recognition in privacy-sensitive settings, demonstrating strong recognition performance for most activities while identifying lying down and getting up as more challenging cases due to their motion similarity. The findings suggest that radar-based human activity recognition is a viable alternative to other more invasive monitoring systems (e.g., camera-based), offering an effective, privacy-preserving solution for smart home and healthcare applications. Full article
►▼ Show Figures

Figure 1

19 pages, 11348 KiB  
Article
Vision-Based Grasping Method for Prosthetic Hands via Geometry and Symmetry Axis Recognition
by Yi Zhang, Yanwei Xie, Qian Zhao, Xiaolei Xu, Hua Deng and Nianen Yi
Biomimetics 2025, 10(4), 242; https://doi.org/10.3390/biomimetics10040242 - 15 Apr 2025
Viewed by 246
Abstract
This paper proposes a grasping method for prosthetic hands based on object geometry and symmetry axis. The method utilizes computer vision to extract the geometric shape, spatial position, and symmetry axis of target objects and selects appropriate grasping modes and postures through the [...] Read more.
This paper proposes a grasping method for prosthetic hands based on object geometry and symmetry axis. The method utilizes computer vision to extract the geometric shape, spatial position, and symmetry axis of target objects and selects appropriate grasping modes and postures through the extracted features. First, grasping patterns are classified based on the analysis of hand-grasping movements. A mapping relationship between object geometry and grasp patterns is established. Then, target object images are captured using binocular depth cameras, and the YOLO algorithm is employed for object detection. The SIFT algorithm is applied to extract the object’s symmetry axis, thereby determining the optimal grasp point and initial hand posture. An experimental platform is built based on a seven-degree-of-freedom (7-DoF) robotic arm and a multi-mode prosthetic hand to conduct grasping experiments on objects with different characteristics. Experimental results demonstrate that the proposed method achieves high accuracy and real-time performance in recognizing object geometric features. The system can automatically match appropriate grasp modes according to object features, improving grasp stability and success rate. Full article
(This article belongs to the Special Issue Human-Inspired Grasp Control in Robotics 2025)
►▼ Show Figures

Figure 1

17 pages, 3452 KiB  
Article
A Categorical Model of General Consciousness
by Yinsheng Zhang
Biomimetics 2025, 10(4), 241; https://doi.org/10.3390/biomimetics10040241 - 14 Apr 2025
Viewed by 484
Abstract
Consciousness is liable to not be defined in scientific research, because it is an object of study in philosophy too, which actually hinders the integration of research on a large scale. The present study attempts to define consciousness with mathematical approaches by including [...] Read more.
Consciousness is liable to not be defined in scientific research, because it is an object of study in philosophy too, which actually hinders the integration of research on a large scale. The present study attempts to define consciousness with mathematical approaches by including the common meaning of consciousness across multiple disciplines. By extracting the essential characteristics of consciousness—transitivity—a categorical model of consciousness is established. This model is used to obtain three layers of categories, namely objects, materials as reflex units, and consciousness per se in homomorphism. The model forms a framework that functional neurons or AI (biomimetic) parts can be treated as variables, functions or local solutions of the model. Consequently, consciousness is quantified algebraically, which helps determining and evaluating consciousness with views that integrate nature and artifacts. Current consciousness theories and computation theories are analyzed to support the model. Full article
(This article belongs to the Special Issue Exploration of Bio-Inspired Computing)
►▼ Show Figures

Figure 1

19 pages, 592 KiB  
Article
A Reinforced, Event-Driven, and Attention-Based Convolution Spiking Neural Network for Multivariate Time Series Prediction
by Ying Li, Xikang Guan, Wenwei Yue, Yongsheng Huang, Bin Zhang and Peibo Duan
Biomimetics 2025, 10(4), 240; https://doi.org/10.3390/biomimetics10040240 - 13 Apr 2025
Viewed by 167
Abstract
Despite spiking neural networks (SNNs) inherently exceling at processing time series due to their rich spatio-temporal information and efficient event-driven computing, the challenge of extracting complex correlations between variables in multivariate time series (MTS) remains to be addressed. This paper proposes a reinforced, [...] Read more.
Despite spiking neural networks (SNNs) inherently exceling at processing time series due to their rich spatio-temporal information and efficient event-driven computing, the challenge of extracting complex correlations between variables in multivariate time series (MTS) remains to be addressed. This paper proposes a reinforced, event-driven, and attention-based convolution SNN model (REAT-CSNN) with three novel features. First, a joint Gramian Angular Field and Rate (GAFR) coding scheme is proposed to convert MTS into spike images, preserving the inherent features in MTS, such as the temporal patterns and spatio-temporal correlations between time series. Second, an advanced LIF-pooling strategy is developed, which is then theoretically and empirically proved to be effective in preserving more features from the regions of interest in spike images than average-pooling strategies. Third, a convolutional block attention mechanism (CBAM) is redesigned to support spike-based input, enhancing event-driven characteristics in weighting operations while maintaining outstanding capability to capture the information encoded in spike images. Experiments on multiple MTS data sets, such as stocks and PM2.5 data sets, demonstrate that our model rivals, and even surpasses, some CNN- and RNN-based techniques, with up to 3% better performance, while consuming significantly less energy. Full article
(This article belongs to the Special Issue Exploration of Bio-Inspired Computing)
►▼ Show Figures

Figure 1

20 pages, 9631 KiB  
Article
Performance Evaluation of Rapid Entire Body Assessment Using AI-Assisted Ergonomic Analysis in Dentistry
by Benhar Arvind Manohar, Jebakani Devaraj, Chellapandian Maheswaran and Selvan Pugalenthi
Biomimetics 2025, 10(4), 239; https://doi.org/10.3390/biomimetics10040239 - 13 Apr 2025
Viewed by 313
Abstract
This study seeks to automate the Rapid Entire Body Assessment (REBA) in dentistry with Artificial Intelligence (AI) technologies, notably MediaPipe, to improve accuracy and obviate the necessity for expert judgment. This research utilizes time-synchronized videos and averages across frames to mitigate mistakes resulting [...] Read more.
This study seeks to automate the Rapid Entire Body Assessment (REBA) in dentistry with Artificial Intelligence (AI) technologies, notably MediaPipe, to improve accuracy and obviate the necessity for expert judgment. This research utilizes time-synchronized videos and averages across frames to mitigate mistakes resulting from visual occlusion and over- or underestimation, respectively. The REBA scores of the observed dentists were evaluated and compared with the conventional single image-based method. Among the evaluated dentists, 83% of dentists are at high risk, and the other 17% of dentists are at very high risk, requiring solutions to lower their REBA scores and prevent musculoskeletal disorders (MSDs). The individual REBA point profiles differed, necessitating a collective study through response surface methodology (RSM) utilizing Design Expert software. The RSM model exhibited substantial results, as indicated by R2 = 0.9055 and p = < 0.0001 values. A linear regression equation was established, and contour graphs depicted the relative variation of REBA points. The optimized REBA score profile establishes a maximum attainable threshold for dentists, directing them towards the lower scores. This streamlined contour functions as a design restriction for creating ergonomic solutions in dental practice. Full article
►▼ Show Figures

Figure 1

20 pages, 3509 KiB  
Article
A Neural Network-Accelerated Approach for Orthopedic Implant Design and Evaluation Through Strain Shielding Analysis
by Ana Isabel Lopes Pais, Jorge Lino Alves and Jorge Belinha
Biomimetics 2025, 10(4), 238; https://doi.org/10.3390/biomimetics10040238 - 13 Apr 2025
Viewed by 261
Abstract
The design of orthopedic implants is a complex challenge, requiring the careful balance of mechanical performance and biological integration to ensure long-term success. This study focuses on the development of a porous femoral stem implant aimed at reducing stiffness and mitigating stress shielding [...] Read more.
The design of orthopedic implants is a complex challenge, requiring the careful balance of mechanical performance and biological integration to ensure long-term success. This study focuses on the development of a porous femoral stem implant aimed at reducing stiffness and mitigating stress shielding effects. To accelerate the design process, neural networks were trained to predict the optimal density distribution of the implant, enabling rapid optimization. Two initial design spaces were evaluated, revealing the necessity of incorporating the femur’s anatomical features into the design process. The trained models achieved a median error near 0 for both conventional and extended design spaces, producing optimized designs in a fraction of the computational time typically required. Finite element analysis (FEA) was employed to assess the mechanical performance of the neural network-generated implants. The results demonstrated that the neural network predictions effectively reduced stress shielding compared to a solid model in 50% of the test cases. While the graded porosity implant design did not show significant differences in stress shielding prevention compared to a uniform porosity design, it was found to be significantly stronger, highlighting its potential for enhanced durability. This work underscores the efficacy of neural network-accelerated design in improving implant development efficiency and performance. Full article
(This article belongs to the Special Issue Dendritic Neuron Model: Theory, Design, Optimization and Applications)
►▼ Show Figures

Figure 1

20 pages, 2636 KiB  
Article
Research on Assembly Sequence Planning of Large Cruise Ship Cabins Based on Improved Genetic Algorithm
by Liyang Ju, Xiaoyuan Wu, Yixi Zhao, Jianfeng Liu and Kun Liu
Biomimetics 2025, 10(4), 237; https://doi.org/10.3390/biomimetics10040237 - 11 Apr 2025
Viewed by 146
Abstract
In the construction process of large cruise ships, there are numerous cabin components, and the number of assembly sequences will experience a “combinatorial explosion”, which will become a complex NP hard problem. This article proposes an assembly sequence planning method based on practical [...] Read more.
In the construction process of large cruise ships, there are numerous cabin components, and the number of assembly sequences will experience a “combinatorial explosion”, which will become a complex NP hard problem. This article proposes an assembly sequence planning method based on practical engineering problems in the construction process of large cruise ships. The cabin components are modularized, and an optimization algorithm is designed for multi-objective problem solving to obtain the optimal assembly sequence of cabin components. This article analyzes the impact of six constraint conditions on the assembly plan, including geometric constraints, sequence constraints, number of assembly reversals, number of tool replacements, stable connection relationships, and selection of reference components. A fitness function is designed and a mathematical model is established. On this basis, a genetic greedy combination algorithm is proposed to solve the optimal assembly sequence. Compared with traditional genetic algorithms, this improves computational efficiency and solves complex problems in a better manner. Multiple unique optimal solutions can be obtained in one solution process. The feasibility and effectiveness of this method were verified through examples. Full article
►▼ Show Figures

Figure 1

39 pages, 3588 KiB  
Article
A Hybrid Black-Winged Kite Algorithm with PSO and Differential Mutation for Superior Global Optimization and Engineering Applications
by Xuemei Zhu, Jinsi Zhang, Chaochuan Jia, Yu Liu and Maosheng Fu
Biomimetics 2025, 10(4), 236; https://doi.org/10.3390/biomimetics10040236 - 11 Apr 2025
Viewed by 207
Abstract
This study addresses the premature convergence issue of the Black-Winged Kite Algorithm (BKA) in high-dimensional optimization problems by proposing an enhanced hybrid algorithm (BKAPI). First of all, BKA provides dynamic global exploration through its hovering and dive attack strategies, while Particle Swarm Optimization [...] Read more.
This study addresses the premature convergence issue of the Black-Winged Kite Algorithm (BKA) in high-dimensional optimization problems by proposing an enhanced hybrid algorithm (BKAPI). First of all, BKA provides dynamic global exploration through its hovering and dive attack strategies, while Particle Swarm Optimization (PSO) enhances local exploitation via its velocity-based search mechanism. Then, PSO enables efficient local refinement, and Differential Evolution (DE) introduces a differential mutation strategy to maintain population diversity and prevent premature convergence. Finally, the integration ensures a balanced exploration–exploitation trade-off, overcoming BKA’s sensitivity to parameter settings and insufficient local search capabilities. By combining these mechanisms, BKAPI achieves a robust balance, significantly improving convergence speed and computational accuracy. To validate its effectiveness, the performance of the enhanced hybrid algorithm is rigorously evaluated against seven other intelligent optimization algorithms using the CEC 2017 and CEC 2022 benchmark test functions. Experimental results demonstrate that the proposed integrated strategy surpasses other advanced algorithms, highlighting its superiority and strong application potential. Additionally, the algorithm’s practical utility is further confirmed through its successful application to three real-world engineering problems: welding beam design, the Himmelblau function, and visible light positioning, underscoring the effectiveness and versatility of the proposed approach. Full article
►▼ Show Figures

Figure 1

12 pages, 1529 KiB  
Article
Synthesis and Application of Sol-Gel-Derived Nano-Silica in Glass Ionomer Cement for Dental Cementation
by Mohammad Dharma Utama, Nina Ariani, Edy Machmud, Acing Habibie Mude, Muhammad Akira Takashi Dharma, Aksani Taqwim and Risnawati Risnawati
Biomimetics 2025, 10(4), 235; https://doi.org/10.3390/biomimetics10040235 - 10 Apr 2025
Viewed by 188
Abstract
Although glass ionomer cements (GIC) are widely used in dental restorations, their long-term performance remains limited by their mechanical properties, including surface roughness and fracture resistance. This study investigates the synthesis of nano-silica from Thalassiosira sp. diatoms through the sol-gel process and its [...] Read more.
Although glass ionomer cements (GIC) are widely used in dental restorations, their long-term performance remains limited by their mechanical properties, including surface roughness and fracture resistance. This study investigates the synthesis of nano-silica from Thalassiosira sp. diatoms through the sol-gel process and its application in influencing the mechanical and physical properties of GIC luting materials. A control group and three experimental groups of different nano-silica concentrations (1%, 3%, and 5%) were compared. Several analyses, including confocal laser scanning microscopy (CLSM), scanning electron microscopy (SEM), and universal testing machines (UTM), were used to determine layer thickness, surface roughness, compressive strength, and tensile strength. Statistical analysis exhibited significant differences between the groups (p < 0.05). The 3% nano-silica group indicated an optimal compromise between mechanical strength and surface smoothness, while the 5% group showed increased thickness and roughness with slightly lower strength. These findings emphasize that the sol-gel-derived nano-silica from Thalassiosira sp. potentially enhances certain characteristics of GIC for possible dental cementation. Further research is needed to determine the long-term durability and bioactivity of these modified materials. Full article
►▼ Show Figures

Figure 1

15 pages, 4898 KiB  
Article
Bio-Inspired Highest Lift-to-Drag-Ratio Fin Shape and Angle for Maximum Surfboard Stability: Flow Around Fish Fins
by Megan S. MacNeill and Brian D. Barkdoll
Biomimetics 2025, 10(4), 234; https://doi.org/10.3390/biomimetics10040234 - 9 Apr 2025
Viewed by 270
Abstract
Wave surfing is a multi-billion dollar industry involving both maneuverability and speed, yet little research has been performed regarding the highest lift-to-drag-ratio fin shape for these competing qualities. Numerical modeling and laboratory experiments were performed here to identify a bio-inspired fin shape that [...] Read more.
Wave surfing is a multi-billion dollar industry involving both maneuverability and speed, yet little research has been performed regarding the highest lift-to-drag-ratio fin shape for these competing qualities. Numerical modeling and laboratory experiments were performed here to identify a bio-inspired fin shape that maximized lateral stability and minimized drag forces, in order to increase surfing maneuverability. Nine fins based on dorsal fins of real fish were tested. Both the CFD and laboratory experiments confirmed that the fin of the same shape as that of the Short-Finned Pilot Whale at an angle of attack of 10° had the greatest lift-to-drag ratios. Flow patterns around fins at a low angle of attack were smooth with negligible flow separation, while at any angle of attack greater than 25°, flow-separation-induced drag forces became excessive. Full article
(This article belongs to the Special Issue Advances in Sensing, Dynamics, and Control for Bio-Inspired Underwater Systems)
►▼ Show Figures

Figure 1

38 pages, 9376 KiB  
Article
IA-DTPSO: A Multi-Strategy Integrated Particle Swarm Optimization for Predicting the Total Urban Water Resources in China
by Zheyu Zhu, Jiawei Wang and Kanhua Yu
Biomimetics 2025, 10(4), 233; https://doi.org/10.3390/biomimetics10040233 - 8 Apr 2025
Viewed by 217
Abstract
In order to overcome the drawbacks of low search efficiency and susceptibility to local optimal traps in PSO, this study proposes a multi-strategy particle swarm optimization (PSO) with information acquisition, referred to as IA-DTPSO. Firstly, Sobol sequence initialization on particles to achieve a [...] Read more.
In order to overcome the drawbacks of low search efficiency and susceptibility to local optimal traps in PSO, this study proposes a multi-strategy particle swarm optimization (PSO) with information acquisition, referred to as IA-DTPSO. Firstly, Sobol sequence initialization on particles to achieve a more uniform initial population distribution is performed. Secondly, an update scheme based on information acquisition is established, which adopts different information processing methods according to the evaluation status of particles at different stages to improve the accuracy of information shared between particles. Then, the Spearman’s correlation coefficient (SCC) is introduced to determine the dimensions that require reverse solution position updates, and the tangent flight strategy is used to improve the inherent single update method of PSO. Finally, a dimension learning strategy is introduced to strengthen individual particles’ activity, thereby ameliorating the entire particle population’s diversity. In order to conduct a comprehensive analysis of IA-DTPSO, its excellent exploration and exploitation (ENE) capability is firstly validated on CEC2022. Subsequently, the performance of IA-DTPSO and other algorithms on different dimensions of CEC2022 is validated, and the results show that IA-DTPSO wins 58.33% and 41.67% of the functions on 10 and 20 dimensions of CEC2022, respectively. Finally, IA-DTPSO is employed to optimize parameters of the time-dependent gray model (1,1,r,ξ,Csz) (TDGM (1,1,r,ξ,Csz)) and applied to simulate and predict total urban water resources (TUWRs) in China. By using four error evaluation indicators, this method is compared with other algorithms and existing models. The results show that the total MAPE (%) value obtained by simulation after IA-DTPSO optimization is 5.9439, which has the smallest error among all comparison methods and models, verifying the effectiveness of this method for predicting TUWRs in China. Full article
(This article belongs to the Special Issue Advances in Swarm Intelligence Optimization Algorithms and Applications)
►▼ Show Figures

Figure 1

31 pages, 7050 KiB  
Article
mESC: An Enhanced Escape Algorithm Fusing Multiple Strategies for Engineering Optimization
by Jia Liu, Jianwei Yang and Lele Cui
Biomimetics 2025, 10(4), 232; https://doi.org/10.3390/biomimetics10040232 - 8 Apr 2025
Viewed by 295
Abstract
A multi-strategy enhanced version of the escape algorithm (mESC, for short) is proposed to address the challenges of balancing exploration and development stages and low convergence accuracy in the escape algorithm (ESC). Firstly, an adaptive perturbation factor strategy was employed to maintain population [...] Read more.
A multi-strategy enhanced version of the escape algorithm (mESC, for short) is proposed to address the challenges of balancing exploration and development stages and low convergence accuracy in the escape algorithm (ESC). Firstly, an adaptive perturbation factor strategy was employed to maintain population diversity. Secondly, introducing a restart mechanism to enhance the exploration capability of mESC. Finally, a dynamic centroid reverse learning strategy was designed to balance local development. In addition, in order to accelerate the global convergence speed, a boundary adjustment strategy based on the elite pool is proposed, which selects elite individuals to replace bad individuals. Comparing mESC with the latest metaheuristic algorithm and high-performance winner algorithm in the CEC2022 testing suite, numerical results confirmed that mESC outperforms other competitors. Finally, the superiority of mESC in handling problems was verified through several classic real-world optimization problems. Full article
(This article belongs to the Special Issue Bio-Inspired Optimization Algorithms and Designs for Engineering Applications: 3rd Edition)
►▼ Show Figures

Figure 1

23 pages, 433 KiB  
Article
Performance Guarantees of Recurrent Neural Networks for the Subset Sum Problem
by Zengkai Wang, Weizhi Liao, Youzhen Jin and Zijia Wang
Biomimetics 2025, 10(4), 231; https://doi.org/10.3390/biomimetics10040231 - 8 Apr 2025
Viewed by 285
Abstract
The subset sum problem is a classical NP-hard problem. Various methods have been developed to address this issue, including backtracking techniques, dynamic programming approaches, branch-and-bound strategies, and Monte Carlo methods. In recent years, researchers have proposed several neural network-based methods for solving combinatorial [...] Read more.
The subset sum problem is a classical NP-hard problem. Various methods have been developed to address this issue, including backtracking techniques, dynamic programming approaches, branch-and-bound strategies, and Monte Carlo methods. In recent years, researchers have proposed several neural network-based methods for solving combinatorial optimization problems, which have shown commendable performance. However, there has been limited research on the performance guarantees of recurrent neural networks (RNNs) when applied to the subset sum problem. In this paper, we conduct a novel investigation into the performance guarantees of RNNs to solve the subset sum problem for the first time. A construction method for RNNs is developed to compute both exact and approximate solutions of subset sum problems, and the mathematical model of each hidden layer in RNNs is rigorously defined. Furthermore, the correctness of the proposed RNNs is strictly proven through mathematical reasoning, and their performance is thoroughly analyzed. In particular, we prove wNN≥wOPT(1−ε) mathematically, i.e., the errors between the approximate solutions obtained by the proposed ASS-NN model and the actual optimal solutions are relatively small and highly consistent with theoretical expectations. Finally, the validity of RNNs is verified through a series of examples, where the actual error value of the approximate solution aligns closely with the theoretical error value. Additionally, our research reveals that recurrence relations in dynamic programming can effectively simulate the process of constructing solutions. Full article
(This article belongs to the Special Issue Bio-Inspired Optimization Algorithms and Designs for Engineering Applications: 3rd Edition)
►▼ Show Figures

Figure 1

24 pages, 3963 KiB  
Article
Development of a Bayesian Network-Based Parallel Mechanism for Lower Limb Gait Rehabilitation
by Huiguo Ma, Yuqi Bao, Chao Jia, Guoqiang Chen, Jingfu Lan, Mingxi Shi, He Li, Qihan Guo, Lei Guan, Shuang Li and Peng Zhang
Biomimetics 2025, 10(4), 230; https://doi.org/10.3390/biomimetics10040230 - 8 Apr 2025
Viewed by 251
Abstract
This study aims to address the clinical needs of hemiplegic and stroke patients with lower limb motor impairments, including gait abnormalities, muscle weakness, and loss of motor coordination during rehabilitation. To achieve this, it proposes an innovative design method for a lower limb [...] Read more.
This study aims to address the clinical needs of hemiplegic and stroke patients with lower limb motor impairments, including gait abnormalities, muscle weakness, and loss of motor coordination during rehabilitation. To achieve this, it proposes an innovative design method for a lower limb rehabilitation training system based on Bayesian networks and parallel mechanisms. A Bayesian network model is constructed based on expert knowledge and structural mechanics analysis, considering key factors such as rehabilitation scenarios, motion trajectory deviations, and rehabilitation goals. By utilizing the motion characteristics of parallel mechanisms, we designed a rehabilitation training device that supports multidimensional gait correction. A three-dimensional digital model is developed, and multi-posture ergonomic simulations are conducted. The study focuses on quantitatively assessing the kinematic characteristics of the hip, knee, and ankle joints while wearing the device, establishing a comprehensive evaluation system that includes range of motion (ROM), dynamic load, and optimization matching of motion trajectories. Kinematic analysis verifies that the structural design of the device is reasonable, aiding in improving patients’ gait, enhancing strength, and restoring flexibility. The Bayesian network model achieves personalized rehabilitation goal optimization through dynamic probability updates. The design of parallel mechanisms significantly expands the range of joint motion, such as enhancing hip sagittal plane mobility and reducing dynamic load, thereby validating the notable optimization effect of parallel mechanisms on gait rehabilitation. Full article
(This article belongs to the Special Issue Advanced Service Robots: Exoskeleton Robots 2025)
►▼ Show Figures

Figure 1

19 pages, 1026 KiB  
Article
Surface EMG Sensing and Granular Gesture Recognition for Rehabilitative Pouring Tasks: A Case Study
by Congyi Zhang, Dalin Zhou, Yinfeng Fang, Naoyuki Kubota and Zhaojie Ju
Biomimetics 2025, 10(4), 229; https://doi.org/10.3390/biomimetics10040229 - 7 Apr 2025
Viewed by 301
Abstract
Surface electromyography (sEMG) non-invasively captures the electrical activity generated by muscle contractions, offering valuable insights into motion intentions. While sEMG has been widely applied to general gesture recognition in rehabilitation, there has been limited exploration of specific, intricate daily tasks, such as the [...] Read more.
Surface electromyography (sEMG) non-invasively captures the electrical activity generated by muscle contractions, offering valuable insights into motion intentions. While sEMG has been widely applied to general gesture recognition in rehabilitation, there has been limited exploration of specific, intricate daily tasks, such as the pouring action. Pouring is a common yet complex movement requiring precise muscle coordination and control, making it an ideal focus for rehabilitation studies. This research proposes a granular computing-based deep learning approach utilizing ConvMixer architecture enhanced with feature fusion and granular computing to improve gesture recognition accuracy. Our findings indicate that the addition of hand-crafted features significantly improves model performance; specifically, the ConvMixer model’s accuracy improved from 0.9512 to 0.9929. These results highlight the potential of our approach in rehabilitation technologies and assistive systems for restoring motor functions in daily activities. Full article
(This article belongs to the Special Issue Artificial Intelligence (AI) in Biomedical Engineering)
►▼ Show Figures

Figure 1

41 pages, 10335 KiB  
Article
Electrical Equivalent Circuit Parameter Estimation of Commercial Induction Machines Using an Enhanced Grey Wolf Optimization Algorithm
by Premkumar Manoharan, Sowmya Ravichandran, Jagarapu S. V. Siva Kumar, Mustafa Abdullah, Tan Ching Sin and Tengku Juhana Tengku Hashim
Biomimetics 2025, 10(4), 228; https://doi.org/10.3390/biomimetics10040228 - 6 Apr 2025
Viewed by 203
Abstract
This paper addresses the critical challenge of optimizing the energy efficiency of induction motors, which are pivotal components across diverse industrial sectors due to their substantial energy consumption. Given the non-measurable internal parameters of induction motors, parameter identification becomes a complex, multidimensional optimization [...] Read more.
This paper addresses the critical challenge of optimizing the energy efficiency of induction motors, which are pivotal components across diverse industrial sectors due to their substantial energy consumption. Given the non-measurable internal parameters of induction motors, parameter identification becomes a complex, multidimensional optimization problem characterized by highly nonlinear and multimodal error surfaces. Traditional optimization algorithms often weaken, yielding suboptimal results due to an inadequate balance between the exploration and exploitation phases. To overcome these limitations, this study introduces an Adaptive Weight Grey Wolf Optimizer (AWGWO) to enhance the accuracy and reliability of induction motor parameter estimation. The AWGWO incorporates an adaptive weight mechanism that dynamically adjusts the exploration and exploitation balance, effectively mitigating issues such as premature convergence to local optima. Extensive simulation validation was conducted across various induction motor models, including eight commercial motors, and demonstrated that AWGWO consistently outperforms state-of-the-art algorithms in terms of convergence speed, solution accuracy, and robustness in multimodal optimization landscapes. The AWGWO consistently exhibited faster convergence, significantly reducing premature convergence. Moreover, the adaptive weight mechanism enabled a more effective balance between exploration and exploitation, leading to higher accuracy in parameter estimation. Comparative analyses reveal that AWGWO outperforms existing algorithms not only in achieving lower error rates, but also in maintaining stability. This study significantly contributes to progress in the field by providing an effective tool for induction motor parameterization, thereby offering potential improvements in energy efficiency. Full article
►▼ Show Figures

Figure 1

16 pages, 3807 KiB  
Article
Development of Structurally Graded Alumina–Polymer Composites as Potential Orthodontic Bracket Materials
by Yin Mun Wong, Anthony J. Ireland and Bo Su
Biomimetics 2025, 10(4), 227; https://doi.org/10.3390/biomimetics10040227 - 5 Apr 2025
Viewed by 218
Abstract
To create an orthodontic bracket material combining the favourable properties of ceramic and polymer while minimising their limitations, graded porous ceramic scaffolds were created using unidirectional gelation-freeze casting, following which the pores were infiltrated with polymer. Two processing parameters were investigated: (1) sedimentation [...] Read more.
To create an orthodontic bracket material combining the favourable properties of ceramic and polymer while minimising their limitations, graded porous ceramic scaffolds were created using unidirectional gelation-freeze casting, following which the pores were infiltrated with polymer. Two processing parameters were investigated: (1) sedimentation times of 0, 8, and 24 h, with ceramic solid loading of 20 vol.% and 2.5 wt.% gelatine concentration, and (2) ceramic solid loadings of 15, 20, and 25 vol.% with a fixed 2.5 wt.% gelatine concentration and an 8 h sedimentation time. The graded ceramic structures demonstrated porosity gradients ranging from 9.86 to 63.84 vol.%, except those with 25 vol.% ceramic solid loading at 8 h sedimentation. The Al2O3-UDMA/TEGDMA composites had compressive strengths of 60.25 to 120.92 MPa, modulus of elasticity of 19.84 to 35.29 GPa, and fracture toughness of 0.78 to 1.78 MPa·m1/2. The values observed were between those of dense ceramic and pure polymer. Statistical analysis was conducted using Excel® 2019 (Microsoft®, Washington, DC, USA). Means, standard deviations, and 95% confidence intervals (CI) were calculated at a significance level of α = 0.05, alongside polynomial regression to evaluate relationships between variables. Composites with 20 vol.% ceramic solid loading at 8 h sedimentation displayed promising potential for further clinical validation. Full article
►▼ Show Figures

Figure 1

22 pages, 8575 KiB  
Article
Improved Black-Winged Kite Algorithm with Multi-Strategy Optimization for Identifying Dendrobium huoshanense
by Chaochuan Jia, Ting Yang, Maosheng Fu, Yu Liu, Xiancun Zhou, Zhendong Huang, Fang Wang and Wenxia Li
Biomimetics 2025, 10(4), 226; https://doi.org/10.3390/biomimetics10040226 - 4 Apr 2025
Viewed by 301
Abstract
An improved black-winged kite algorithm with multiple strategies (BKAIM) is proposed in this paper to address two critical limitations in the original black-winged kite optimization algorithm (BKA): the restricted search capability caused by the low-quality initial population and the reduced population diversity resulting [...] Read more.
An improved black-winged kite algorithm with multiple strategies (BKAIM) is proposed in this paper to address two critical limitations in the original black-winged kite optimization algorithm (BKA): the restricted search capability caused by the low-quality initial population and the reduced population diversity resulting from blind following behavior during the migration phase. Our enhancement implements three strategic modifications across different algorithm stages. During initialization, an opposition-based learning strategy was incorporated to generate a higher-quality initial population. For the migration phase, a differential mutation strategy was integrated to facilitate information exchange among population members, mitigate the tendency of blind leader-following behavior, enhance convergence precision, and achieve an optimal balance between exploration and exploitation capabilities. Regarding boundary handling, the conventional absorption boundary method was replaced with a random boundary approach to increase population diversity and subsequently improve the algorithm’s search capabilities. Comprehensive testing was conducted on four benchmark function sets (CEC2017, CEC2019, CEC2021, and CEC2022) to validate the effectiveness of the improved algorithm. Detailed convergence analysis and Wilcoxon rank-sum test comparisons with other algorithms demonstrated BKAIM’s superior convergence performance and robustness. Furthermore, the support vector machine (SVM) model was optimized by BKAIM for grade identification of Dendrobium huoshanense based on near-infrared spectral data, thereby confirming its effectiveness in practical applications. Full article
►▼ Show Figures

Figure 1

18 pages, 4988 KiB  
Article
A Capsule Decision Neural Network Based on Transfer Learning for EEG Signal Classification
by Wei Zhang, Xianlun Tang, Xiaoyuan Dang and Mengzhou Wang
Biomimetics 2025, 10(4), 225; https://doi.org/10.3390/biomimetics10040225 - 4 Apr 2025
Viewed by 254
Abstract
Transfer learning is the act of using the data or knowledge in a problem to help solve different but related problems. In a brain computer interface (BCI), it is important to deal with individual differences between topics and/or tasks. A kind of capsule [...] Read more.
Transfer learning is the act of using the data or knowledge in a problem to help solve different but related problems. In a brain computer interface (BCI), it is important to deal with individual differences between topics and/or tasks. A kind of capsule decision neural network (CDNN) based on transfer learning is proposed. In order to solve the problem of feature distortion caused by EEG feature extraction algorithm, a deep capsule decision network was constructed. The architecture includes multiple primary capsules to form a hidden layer, and the connection between the advanced capsule and the primary capsule is determined by the neural decision routing algorithm. Unlike the dynamic routing algorithm that iteratively calculates the similarity between primary capsules and advanced capsules, the neural decision network computes the relationship between each capsule in the deep and shallow hidden layers in a probabilistic manner. At the same time, the distribution of the EEG covariance matrix is aligned in Riemann space, and the regional adaptive method is further introduced to improve the independent decoding ability of the capsule decision neural network for the subject’s EEG signals. Experiments on two motor imagery EEG datasets show that CDNN outperforms several of the most advanced transfer learning methods. Full article
(This article belongs to the Special Issue Advances in Brain–Computer Interfaces (BCI): Challenges and Opportunities)
►▼ Show Figures

Figure 1

Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-63
Previous Issue
Next Issue
Back to TopTop