Search Results (1,474)

Value factorization methods have become a standard tool for cooperative multi-agent reinforcement learning (MARL) in the centralized-training, decentralized-execution (CTDE) setting. QMIX (a monotonic mixing network for value factorization), in particular, constrains the joint action–value function to be a monotonic mixing of per-agent utilities, which guarantees consistency with individual greedy policies but can severely limit expressiveness on tasks with non-monotonic agent interactions. This work revisits this design choice and proposes Relaxed Monotonic QMIX (R-QMIX), a simple regularized variant of QMIX that encourages but does not strictly enforce the monotonicity constraint. R-QMIX removes the sign constraints on the mixing network weights and introduces a differentiable penalty on negative partial derivatives of the joint value with respect to each agent’s utility. This preserves the computational benefits of value factorization while allowing the joint value to deviate from strict monotonicity when beneficial. R-QMIX is implemented in a standard PyMARL (an open-source MARL codebase) and evaluated on the StarCraft Multi-Agent Challenge (SMAC). On a simple map (3m), R-QMIX matches the asymptotic performance of QMIX while learning substantially faster. On more challenging maps (MMM2, 6h vs. 8z, and 27m vs. 30m), R-QMIX significantly improves both sample efficiency and final win rate (WR), for example increasing the final-quarter mean win rate from $42.3\%$ to $97.1\%$ on MMM2, from $0.0\%$ to $57.5\%$ on 6h vs. 8z, and from $58.0\%$ to $96.6\%$ on 27m vs. 30m. These results suggest that soft monotonicity regularization is a practical way to bridge the gap between strictly monotonic value factorization and fully unconstrained joint value functions. A further comparison against QTRAN (Q-value transformation), a more expressive value factorization method, shows that R-QMIX achieves higher and more reliably convergent win rates on the challenging SMAC maps considered. Full article

Transcranial Magnetic Stimulation (TMS) is a non-invasive technique for neurological research and therapy, but its effectiveness depends on accurate and stable coil placement. Manual localization based on anatomical landmarks is time-consuming and operator-dependent, while state-of-the-art robotic and neuronavigation systems achieve high accuracy using optical tracking with head-mounted markers and infrared cameras, at the cost of increased system complexity and setup burden. This study presents a cost-effective, markerless robotic-assisted TMS system that combines a 3D depth camera and textile capacitive sensors to assist coil localization and contact control. Facial landmarks detected by the depth camera are used to estimate the motor cortex (C3) location without external tracking markers, while a dual textile-sensor suspension provides compliant “soft-landing” behavior, contact confirmation, and coil-tilt estimation. Experimental evaluation with five participants showed reliable C3 targeting with valid motor evoked potentials (MEPs) obtained in most trials after initial calibration, and tilt-verification experiments revealed that peak MEP amplitudes occurred near balanced sensor readings in 12 of 15 trials (80%). The system employs a collaborative robot designed in accordance with international human–robot interaction safety standards, including force-limited actuation and monitored stopping. These results suggest that the proposed approach can improve the accessibility, safety, and consistency of TMS procedures while avoiding the complexity of conventional optical tracking systems. Full article

Pneumatic soft robotic devices are emerging as promising tools for assisting hand rehabilitation in individuals with post-stroke motor impairment. However, evidence regarding their immediate functional impact remains limited, particularly across different impairment levels. This study presents a pilot validation of the YAD_V2 pneumatic finger rehabilitation robot and evaluates acute changes in finger range of motion (ROM) and task performance during a single intervention session. Twenty stroke participants were categorized into two groups based on the Fugl-Mayer Hand sub score: severe impairment (FMA-Hand < 10) and mild-to-moderate impairment (FMA-Hand ≥ 10). ROM was measured using integrated bending sensors during voluntary flexion–extension before, during, and after a 10-min pneumatic actuation session. A mixed 2 × 3 repeated-measure ANOVA revealed a significant Group × Time interaction (F(2, 36) = 4.628, p = 0.016, partial η² = 0.205). In the severe group, ROM increased from 8.53° to 28.46° during actuation (p = 0.002), and partially returned to baseline afterward. In the mild–moderate group, no significant ROM changes were observed; however, cube-transfer time improved significantly (mean improvement: 0.88 s, p = 0.039). These findings indicate that pneumatic assistance induces distinct acute effects depending on impairment severity. This study provides preliminary evidence supporting the feasibility of the YAD_V2 robotic system and highlights the need for multi-session clinical trials to determine therapeutic efficacy. Full article

Pile foundations are critical load-bearing components in bridge structures, particularly in soft, high-moisture soils susceptible to external disturbances. This study investigated the impact of large-scale soil excavation on the stability of adjacent pile foundations through comprehensive field monitoring of a newly constructed bridge during both the bridge construction and channel excavation phases. The close proximity of the excavation site to the pile caps facilitated a detailed assessment of soil–structure interaction. The results indicate that the pile axial force peaked at the pile head and decreased progressively with depth, consistent with the load transfer mechanism of friction piles. Notably, a distinct variation in axial force was observed at the bedrock interface, attributed to reduced relative displacement between the pile and the surrounding soil. Furthermore, channel water filling raised the local groundwater table, which increased the buoyancy and reduced negative skin friction, thereby decreasing the pile axial force. The study also highlighted the sensitivity of pile deformation in soft soil to unbalanced earth pressure. Asymmetric excavation and surface surcharge loading were identified as critical factors compromising pile stability and overall structural safety. These findings provide valuable insights for construction practices and offer effective strategies to mitigate adverse excavation effects, ensuring long-term structural stability. Full article

In order to solve the dynamic analysis and interactive imaging control problems in the deformation process of bionic soft lenses, dielectric elastomer (DE) actuators are separated from a convex lens, and data-driven eye-controlled motion technology is investigated. According to the DE properties, which are consistent with the deformation characteristics of hydrogel electrodes, the motion and deformation effect of eye-controlled lenses under film prestretching, lens size, and driving voltage, is studied. The results show that when the driving voltage increases to 7.8 kV, the focal length of the lens, whose prestretching λ is 4, and the diameter d is 1 cm, varies in the range of 49.7 mm and 112.5 mm. And the maximum focal-length change could reach 58.9%. In the process of eye controlling design and experimental verification, a high DC voltage supply was programmed, and eye movement signals for controlling the lens were analyzed by MATLAB software (R2023b). Eye-controlled interactive real-time motion and tunable imaging of the lens were realized. The response efficiency of soft lenses could reach over 93%. The adaptive lens system developed in this research has the potential to be applied to medical rehabilitation, exploration, augmented reality (AR), and virtual reality (VR) in the future. Full article

Agricultural robotics and artificial intelligence (AI) are becoming essential to building more sustainable, efficient, and resilient food systems. As climate change, food security pressures, and labour shortages intensify, the integration of intelligent technologies in agriculture has gained strategic importance. This systematic review provides a consolidated assessment of AI and robotics research in agriculture from 2000 to 2025, identifying major trends, methodological trajectories, and underexplored domains. A structured search was conducted in the Scopus database—which was selected for its broad coverage of engineering, computer science, and agricultural technology—and records were screened using predefined inclusion and exclusion criteria across title, abstract, keywords, and eligibility levels. The final dataset was analysed through descriptive statistics and science-mapping techniques (VOSviewer, SciMAT). Out of 4894 retrieved records, 3673 studies met the eligibility criteria and were included. As with all bibliometric reviews, the synthesis reflects the scope of indexed publications and available metadata, and potential selection bias was mitigated through a multi-stage screening workflow. The analysis revealed four dominant research themes: deep-learning-based perception, UAV-enabled remote sensing, data-driven decision systems, and precision agriculture. Several strategically relevant but underdeveloped areas also emerged, including soft manipulation, multimodal sensing, sim-to-real transfer, and adaptive autonomy. Geographical patterns highlight a strong concentration of research in China and India, reflecting agricultural scale and investment dynamics. Overall, the field appears technologically mature in perception and aerial sensing but remains limited in physical interaction, uncertainty-aware control, and long-term autonomous operation. These gaps indicate concrete opportunities for advancing next-generation AI-driven robotic systems in agriculture. Funding sources are reported in the full manuscript. Full article

Sharp declines in temperature pose a significant risk for mass mortality events in the Chinese soft-shelled turtle (Pelodiscus sinensis). To assess the effects of acute cold stress on intestinal health, turtles were exposed to temperatures of 28 °C (control), 14 °C, and 7 °C for 1, 2, 4, 8, and 16 days. The results showed that acute cold stress at 14 °C and 7 °C induced time-dependent alterations in intestinal morphology and histopathology. The damage was more severe at 7 °C, characterized by inflammatory cell infiltration, lymphoid hyperplasia, and extensive detachment and necrosis across the villi, muscle layer, and submucosa. 16S rDNA sequencing revealed significant shifts in intestinal microbiota composition in the 7 °C group, dominated by Helicobacter and Citrobacter. Transcriptomic analysis identified differentially expressed genes (DEGs) that respond to acute cold stress and are involved in the Toll-like receptor signaling pathway (Tlr2, Tlr4, Tlr5, Tlr7, and Tlr8), the NOD-like receptor signaling pathway (Traf6, Traf2, Casr, Rnasel, Pstpip1, Plcb2, Atg5, and Mfn2), apoptosis (Tuba1c, Ctsz, Ctsb, Kras, Hras, Pik3ca, Bcl2l11, Gadd45a, Pmaip1, Ddit3, and Fos), and the p53 signaling pathway (Serpine1, Sesn2, Ccng2, Igf1, Mdm2, Gadd45a, Pmaip1, and Cdkn1a). Metabolomic profiling highlighted differentially expressed metabolites (DEMs) that cope with acute cold stress, such as organic acids (oxoglutaric acid, L-aspartic acid, fumaric acid, DL-malic acid, and citric acid) and amino acids (including L-lysine, L-homoserine, and allysine). The integrated analysis of DEGs and DEMs underscored three key pathways modulated by acute cold stress: linoleic acid metabolism, neuroactive ligand–receptor interaction, and the FoxO signaling pathway. This study provides a comprehensive evaluation of intestinal health in Chinese soft-shelled turtles under acute cold stress and elucidates the underlying mechanisms. Full article

This review presents a focused assessment of the rapidly expanding field of gelatin-based eutectogels and identifies the gaps in current literature that justify this examination. Research on deep eutectic solvents (DESs and NADES) has advanced quickly, yet there is still no integrated view of how these solvent systems influence adhesion in gelatin-based gels. Eutectogels are soft materials formed by gelling DESs or NADES with biopolymers. Gelatin is widely used because it is biocompatible, biodegradable, and readily available. We provide a clear overview of the chemistry of DESs and NADES and describe how gelatin forms networks in these media. The review summarizes established knowledge on adhesion, highlighting the contributions of polymer network density, interfacial hydrogen bonding, and solvent mobility. New perspectives are introduced on how these factors interact to control adhesion strength, toughness, and reversibility. A key topic is the role of hydrogen bond donors (HBDs) and acceptors (HBAs). They define the hydrogen bonding environment of the solvent and represent an underexplored way to tune mechanical and adhesive behavior. Examples such as moisture-resistant adhesion and temperature-responsive bonding show why these systems offer unique and adjustable properties. The review concludes by outlining major challenges, including the lack of standardized adhesion tests and constraints in scalable production, and identifying directions for future work. Full article

Off-road vehicle control designs often neglect the complex tire–soil interactions inherent to soft terrain. This paper proposes a Variable Impedance Control (VIC) strategy integrated with a high-fidelity terramechanics model. First, a real-time sinkage estimation algorithm is derived using experimentally identified Bekker parameters and the quasi-rigid wheel assumption to capture the nonlinear feedback between soil deformation and vehicle dynamics. Building on this, the VIC strategy adaptively regulates virtual stiffness, damping, and inertia parameters based on real-time suspension states. Comparative simulations on an ISO Class-C soft soil profile demonstrate that this framework effectively balances ride comfort and safety constraints. Specifically, the VIC strategy reduces the root-mean-square of vertical body acceleration by 46.9% compared to the passive baseline, significantly outperforming the Linear Quadratic Regulator (LQR). Furthermore, it achieves a 48.6% reduction in average power relative to LQR while maintaining suspension deflection strictly within the safe range. Moreover, unlike LQR, the VIC strategy improves tire deflection performance, ensuring superior ground adhesion. These results validate the method’s robustness and energy efficiency for off-road applications. Full article

Vehicle repeated passes over soft terrain alter the soil’s bearing and shear behavior, thereby affecting vehicle mobility and energy consumption. To address this issue, this study conducted cyclic compression and shear tests on beach sand with moisture contents of 5%, 15%, and 25%. A constitutive model incorporating the coupling effects of loading cycles (N) and moisture content (ω) was developed based on the Bekker and Janosi model framework. The model expresses compression parameters as functions of N and ω, and describes shear behavior through the strength evolution function k(N,ω) and deformation modulus function h(N,ω). Results show excellent agreement between the model predictions and experimental data (R² > 0.92). Furthermore, a vehicle–soil coupled dynamics model was established based on the proposed constitutive model, forming a comprehensive analytical framework that integrates soil meso-mechanics with full vehicle–terrain interaction. This work provides valuable theoretical and technical support for predicting vehicle trafficability on coastal soft soils and optimizing vehicle suspension systems. Full article

This study explores how classical painting techniques, particularly those rooted in the Renaissance tradition, can produce illusions of depth that vary with the viewer’s position. Focusing on a work rich in soft shading and subtle tonal transitions, we investigate how movement across the frontal plane influences the perception of spatial structure. A sequence of high-resolution photographs was taken from slightly offset viewpoints, simulating natural viewer motion. Using image alignment and pixel-wise difference mapping, we reveal perceptual shifts that suggest the presence of latent three-dimensional cues embedded within the painted surface. The findings offer visual and empirical support for concepts such as and dynamic engagement, where depth is constructed not solely by the image, but by the interaction between the artwork and the observer. Our approach demonstrates how digital analysis can enrich art historical interpretation, offering new insight into how still images can evoke the illusion of spatial presence. Full article

Deep foundation excavation in water-rich sand strata presents complex deformation characteristics driven by fluid–solid interaction, which distinguishes it from excavations in cohesive soft clay. This study investigates the spatiotemporal evolution and deformation mechanisms of retaining structures through a comparative analysis of field monitoring data and 3D numerical simulation, based on a subway station project in Xi’an. While the numerical simulation predicted a continuous “bulging” deformation mode, field monitoring revealed a distinct transition from a “bulging” profile to a “step-like” deformation pattern as the excavation deepened. Quantitatively, while the simulation captured the spatial trend, the measured maximum surface settlement (7.8 mm) exceeded the simulated value (1.2 mm), highlighting the dominant role of seepage consolidation. Detailed analysis indicates that this discrepancy—and the unique step-like evolution—is primarily driven by two mechanisms: the rapid stress relaxation of cohesionless sand during the time lag of support installation, and the superimposed seepage forces induced by continuous dewatering, which are often simplified in standard elastoplastic models. The study further identifies that the vertical displacement of the pile top is governed by the combined effects of basal heave and the “kick-out” deformation at the pile toe. These findings demonstrate that in high-permeability water-rich sand, deformation control depends critically on minimizing the unsupported exposure time of the excavation face. This research provides a theoretical basis for optimizing the spatiotemporal sequencing of excavation in similar geological conditions. Full article

To achieve precise in-hand manipulation and feedback control using soft robotic fingers, it is essential to accurately measure their deformable structures. In particular, estimating the three-dimensional shape of a soft finger under contact conditions is a critical challenge, as the deformation state directly affects manipulation reliability. However, nonlinear deformations and occlusions arising from interactions with external objects make the estimation difficult. To address these issues, we propose a soft finger structure that integrates small magnets and magnetic sensors inside the body, enabling the acquisition of rich deformation information in both contact and non-contact states. The design provides a 15-dimensional time-series signal composed of motor angles, motor currents, and magnetic sensor outputs as inputs for shape estimation. Built on the sensing signals, we propose a mode-selection-based learning approach that outputs multiple candidate shapes and selects the correct one. The proposed network predicts the three-dimensional positions of four external markers attached to the finger, which serve as a proxy representation of the finger’s shape. The network is trained in a supervised manner using ground-truth marker positions measured by a motion capture system. The experimental results under both contact and non-contact conditions demonstrate that the proposed method achieves an average estimation error of approximately 4 mm, outperforming conventional one-shot regression models that output coordinates directly. The integration of magnetic sensing is demonstrated to be able to enable accurate recognition of contact states and significantly improve stability in shape estimation. Full article

The performance of New Zealand’s construction companies depends on the adaptability and skills of their workforce. The soft skills of the company’s building professionals are thought to contribute to the delivery of successful construction projects. This pilot study captures the perceptions of the importance of critical soft skills in semi-structured interviews with thirteen Quantity Surveyors (QSs) and fourteen Project Managers (PMs) working in New Zealand. For both cohorts the most important skill is communication, followed by workplace ethics. An exploratory Mann–Whitney U comparison suggests a difference in their ranking of emotional intelligence in interactions with other stakeholders, with PM deeming it more important than QS. Within-cohort Spearman rank correlation shows different patterns of association among soft-skill clusters for QS and PM, offering contextual insight rather than confirmatory inference. After communication and ethics, QS prioritise dispute resolution while PM value project reasoning. A combination of individual traits and practical experience influences the successful transition from a QS role to the broader PM role. The findings are limited by the small sample size but may be useful in professional development courses and recruitment efforts, contributing to a more adaptable and flexible construction workforce. Full article

Subsurface deposition determines whether soils, aquifers, or ocean sediment represent a sink or temporary reservoir for microplastics. Deposition is generally studied by applying the Smoluchowski–Levich equation to determine a particle’s sticking efficiency, which relates the number of particles filtered by sediment to the probability of attachment occurring from an interaction between particles and sediment. Sticking efficiency is typically measured using column experiments or estimated from theory using the Interaction Force Boundary Layer (IFBL) model. However, there is generally a large discrepancy (orders of magnitude) between the values predicted from IFBL theory and the experimental column measurements. One way to bridge this gap is to directly measure a microparticle’s interaction forces using Atomic Force Microscopy (AFM). Herein, an AFM method is presented to measure sticking efficiency for a model polystyrene microparticle (2 μm) on a model geomaterial surface (glass or quartz) in environmentally relevant, synthetic freshwaters of varying ionic strength (de-ionized water, soft water, hard water). These data, collected over nanometer length scales, are compared to sticking efficiencies determined through traditional approaches. Force measurement results show that AFM can detect extremely low sticking efficiencies, surpassing the sensitivity of column studies. These data also demonstrate that the 75th to 95th percentile, rather than the mean or median force values, provides a better approximation to values measured in model column experiments or field settings. This variability of the methods provides insight into the fundamental mechanics of microplastic deposition and suggests AFM is isolating the physicochemical interactions, while column experiments also include physical interactions like straining. Advantages of AFM over traditional column/field experiments include high throughput, small volumes, and speed of data collection. For example, at a ramp rate of 1 Hz, 60 sticking efficiency measurements could be made in only a minute. Compared to column or field experiments, the AFM requires much less liquid (μL volume) making it effortless to examine the impact of solution chemistry (temperature, pH, ionic strength, valency of dissolved ions, presence of organics, etc.). Potential limitations of this AFM approach are presented alongside possible solutions (e.g., baseline correction, numerical integration). If these challenges are successfully addressed, then AFM would provide a completely new approach to help elucidate which subsurface minerals represent a sink or temporary storage site for microparticles on their journey from terrestrial to oceanic environments. Full article