Search Results (557)

This paper investigates robotic surface-to-surface object transfer, a release manipulation task in which a robot transfers an object from an end-effector that functions solely as a large supporting surface to an external surface such as the ground. Such transfers commonly arise in many practical manipulation scenarios. Unlike simple releasing actions, surface-to-surface transfer requires maintaining force equilibrium through controlled rolling and sliding at the contact interfaces. We present a manipulation model that captures the essential contact kinematics and enables force balance throughout the transfer. To assess robustness, we introduce a stability simulation framework that evaluates dynamic stability by monitoring variations in gravitational potential energy across object configurations. The proposed approach is implemented on a quadrupedal robot and validated through a series of experiments with objects of varying geometries. The results demonstrate the effectiveness of the method and underscore the role of stability-aware control in surface-to-surface transfer. Full article

While walking robots possess significantpotential for various real-world applications, the reliance on high-performance sensors and complex control architectures for precise gait control remains a significant barrier to commercialization and lightweight design. To overcome these engineering limitations and lay the groundwork for a sensing paradigm adaptable to complex terrains, this study proposes an AI-based sensorless feedback control framework that incorporates the biological principles of proprioception. To this end, a walking robot leveraging the morphological intelligence of the Klann linkage was developed. We constructed a time-series dataset by defining motor current signals as ‘interoceptive sensing’ information—analogous to biological muscle feedback—and synchronizing them with absolute angular data. This dataset was used to train an Attention-LSTM (A-LSTM) model, which predicts future motor states in real-time by decoding nonlinear physical information embedded within internal current data, independent of external environmental sensors. By integrating the proposed model into a PI controller, a stable biomimetic walking loop was successfully implemented without the need for additional position sensors. Full article

The rehabilitation training of patients with lower limb motor dysfunction highly relies on the precise control of biomimetic knee joint robots. Existing control strategies generally suffer from insufficient control accuracy and weak anti-interference ability, and an optimization plan that balances high precision and strong anti-interference has not yet been formed, which seriously affects the effectiveness of rehabilitation training. In order to improve the control accuracy and anti-interference ability of biomimetic knee joint robots for leg rehabilitation training of patients with lower limb movement disorders, the purpose of this study is to address the performance shortcomings of existing biomimetic knee joint robot control strategies. The goal is to propose a high-precision and strong anti-interference control strategy to provide more reliable rehabilitation support for patients with lower limb movement disorders. Therefore, this article proposes an optimization strategy based on the Spider Bee Algorithm (SWO) combined with fuzzy PID control. Based on a biomimetic knee joint robot model, this study simulates three common pathological states of knee joint ligament injury, meniscus injury, and muscle atrophy in patients, and compares the trajectory tracking and anti-interference performance of PID, fuzzy PID, and SWO fuzzy PID control strategies. The experimental results show that the SWO fuzzy PID control strategy has the best comprehensive performance: the overshoot of knee joint angle control is only 9.7%, and the peak angle error is reduced to 2.1948°; when simulating pathological conditions, the system takes the shortest time to recover stability: 1.068 s for ligament injuries and 0.929 s for meniscus injuries, with maximum response errors below 0.017°. Simulation experiments on healthy subjects showed that the system had a tracking error of ≤5° under two rehabilitation training modes, meeting clinical accuracy requirements, and had good performance in restoring stability under irregular vibration interference. The core contribution of this study is the proposal of the SWO fuzzy PID optimization control strategy, which effectively addresses the shortcomings of existing strategies and significantly improves the control accuracy and anti-interference ability of bionic knee joint robots, providing theoretical support and practical reference for the application of bionic knee joint robots. Full article

Biarticular actuators can enhance efficiency and stability in legged locomotion by transferring energy between joints. Their effectiveness depends strongly on the lever arm ratio—the ratio of the actuator’s moment arm at one joint to its moment arm at another—which governs how torque is distributed across joints during movement. Inspired by biomechanics, early robotic studies implemented biarticular actuators to improve energy efficiency, joint coordination, and positional control, primarily in planar or single-joint systems, leaving a gap in fully 3D robotic legs. Here, we present a geometry optimization framework for a robotic leg incorporating both biarticular and monoarticular actuators. Using human motion capture and joint torque data, we optimized the linkage mechanisms so that the system can maintain the required joint torques while keeping biarticular actuator moment arm ratios near their optimal values during walking and running. The optimized leg achieved a minimum achievable cost of transport of approximately $0.41 \mathrm{J}/(\mathrm{kg}·\mathrm{m})$ for walking and $0.62 \mathrm{J}/(\mathrm{kg}·\mathrm{m})$ for running. Full article

Quadruped robots possess strong adaptability to rugged terrain, soft ground, and multi-obstacle environments, offering broad application prospects in extraterrestrial planetary exploration. However, large diurnal temperature variations on extraterrestrial bodies exacerbate joint friction nonlinearity, degrading motion control accuracy and stability. To address this, a quadruped robot prototype with hybrid serial–parallel legs is designed for lunar exploration, and an 18-DOF dynamic model is derived using d’Alembert’s principle. Based on the PPO (Proximal Policy Optimization) reinforcement learning algorithm, joint friction parameters are identified using joint velocity and foot–ground contact force. By introducing friction compensation and contact force, an accurate dynamics-based feedback linearization control model is constructed, and a motion impedance control law is designed. Finally, joint friction parameters are identified and validated through both virtual and experimental prototypes, and the proposed control method is tested on flat and sloped terrain. Results show that the method can precisely regulate contact force and foot position, keeping RMSE (Root Mean Square Error) of position within 21.04 mm while preventing slipping and false contact. Full article

This paper proposes a learning-based contact force controller using deep neural networks (DNN) and a PI controller. Stable contact force control between the foot and the ground is essential for humanoid robots to maintain balance during bipedal walking. While admittance controllers have been extensively employed for contact force control in humanoid robots, their performance is limited by the high nonlinearity inherent in robot systems. To overcome these limitations, we propose a deep neural network (DNN)–based inverse model, which leverages input–output data that inherently capture system nonlinearities. The proposed learning-based contact force controller computes the target foot height based on the target force, measured force, and measured foot height, without relying on a dynamic model of the articulated robotic leg. Furthermore, a PI controller is integrated to mitigate steady-state errors. Experimental comparisons between the proposed controller and an admittance controller were conducted using an articulated robotic leg. Compared with an admittance controller, the proposed method reduced overshoot by 96% and settling time by 61% on average in step responses and decreased force-tracking RMSE by 66.3% on average across both step and sinusoidal experiments. Full article

Bipedal wheeled robots combine the advantages of wheeled mobility and legged agility, enabling high-speed locomotion and obstacle negotiation in complex environments. However, their dynamic behavior is inherently unstable and highly coupled, making robust control particularly challenging in the presence of task conflicts, external disturbances, and modeling uncertainties. This paper proposes an RBF–L1–WBC framework that integrates L1 adaptive control to compensate for model inaccuracies and disturbances, radial basis function (RBF) neural networks to approximate nonlinear variations in linear quadratic regulator (LQR) gains, and whole-body control (WBC) to coordinate multiple tasks while mitigating control conflicts. Experimental findings confirm that the proposed methodology yields statistically significant improvements in both attitude regulation precision and velocity tracking accuracy, surpassing the performance of benchmark controllers including classical LQR, adaptive LQR, and classical Virtual Model Control (VMC). Full article

Exploring the lunar complex and extreme terrain presents formidable challenges for conventional lunar rovers. To address these limitations, this study proposes a novel hexapod jumping hybrid robot that incorporates a “figure-of-eight” (butterfly-shaped) six-branched wheel-legged mechanism and a jumping system that stores elastic energy via deformation of its elastic body. Inspired by the multimodal locomotion of grasshoppers, the robot dynamically switches between two operational modes: high-efficiency wheeled locomotion on relatively flat surfaces and agile jumping to traverse steep slopes and surmount large obstacles. A bio-inspired gait, inspired by the crawling patterns of a hexapod insect, is implemented using a Central Pattern Generator (CPG)-based controller to produce coordinated, rhythmic limb movements. Dynamic simulations of the jumping mechanism were conducted to optimize the critical parameters of the elastic structure and its associated control strategy. Experiments on a physical prototype were conducted to validate the robot’s wheeled mobility and jumping performance. The results demonstrate that the robot exhibits excellent adaptability to rugged terrains and obstacle-dense environments. The integration of multimodal locomotion and adaptive gait control significantly enhances the robot’s operational robustness and survivability in the harsh lunar environment, opening new possibilities for future lunar exploration missions. Full article

Hexapod walking robots are a subject of intense research in the existing literature. To move effectively in natural terrain, these robots must be able to adapt to surface irregularities. While most existing designs employ sophisticated technical solutions for the leg mechanisms, none of these projects allow for combined roll and pitch movements of the body segments. This paper addresses this gap, presenting the concept of a hexapod robot with a body formed of three segments connected by two active universal joints. This unique architecture allows the robot to locomote on both sides and autonomously recover from a rollover event. The robot’s legs are underactuated, utilizing a passive spring element to simplify the mechanical design and control system while maintaining effective terrain adaptation capabilities. Experimental results are presented and discussed, validating the theoretical model and demonstrating the effectiveness of the proposed solution on varied terrains. Full article

Legged robots inspired by animal locomotion require actuators with high power density, fast response, and robust force control, yet traditional valve-controlled hydraulic systems suffer from substantial energy losses and weak regeneration performance. Motivated by role allocation across gait phases in animal legs, where in-air positioning requires far less actuation effort than ground contact support and force modulation, this work proposes a novel gas–oil hybrid servo actuator, denoted GOhsa, for quadruped knee joints. GOhsa utilizes pre-charged high-pressure gas to pressurize hydraulic oil, converting the conventional dual-chamber pressure servo control into a single-chamber configuration while preserving the original piston stroke. This architecture enables bidirectional position–force control, enhances energy regeneration applicability, and improves operational efficiency. Theoretical modeling is conducted to analyze hydraulic stiffness and frequency-response characteristics, and a linearization-based force controller with dynamic compensation is developed to handle system nonlinearities. Experimental validation on a single-leg platform demonstrates significant energy-saving performance: under no-load conditions (simulating the swing phase), GOhsa achieves a maximum power reduction of 79.1%, with average reductions of 15.2% and 11.5% at inflation pressures of 3 MPa and 4 MPa, respectively. Under loaded conditions (simulating the stance phase), the maximum reduction reaches 28.0%, with average savings of 10.0% and 9.8%. Tracking accuracy is comparable to traditional actuators, with reduced maximum errors (13.7 mm/16.5 mm at 3 MPa; 15.0 mm/17.8 mm at 4 MPa) relative to the 16.6 mm and 18.1 mm errors of the conventional system, confirming improved motion stability under load. These results verify that GOhsa provides high control performance with markedly enhanced energy efficiency. Full article

Walking on compliant terrain presents a substantial challenge for individuals with lower-limb amputation, further elevating their already high risk of falling. While powered ankle–foot prostheses have demonstrated adaptability across speeds and rigid terrains, control strategies optimized for soft or compliant surfaces remain underexplored. This work experimentally validates an admittance-based control strategy that dynamically adjusts the quasi-stiffness of powered prostheses to enhance gait stability on compliant ground. Human subject experiments were conducted with three healthy individuals walking on two bilaterally compliant surfaces with ground stiffness values of 63 and $25{\displaystyle \frac{\mathrm{kN}}{\mathrm{m}}}$, representative of real-world soft environments. Controller performance was quantified using phase portraits and two walking stability metrics, offering a direct assessment of fall risk. Compared to a standard phase-variable controller developed for rigid terrain, the proposed admittance controller reduced short-term maximum Lyapunov exponents by an average of 7%, indicating improved local dynamic stability. These results support the potential of adaptive prostheses control to enhance gait stability on compliant surfaces, contributing to the development of more robust human–prosthesis interaction. Full article

Modular reconfigurable robots exhibit significant potential in adapting to complex terrains through cooperative multi-robot formations. However, current control systems often struggle to maintain consistent performance when the number of modules varies due to a lack of unified and adaptive control frameworks. Existing Virtual Model Control (VMC) methods, while effective for fixed-configuration legged robots, are limited in their ability to dynamically adjust control parameters in reconfigurable multi-legged systems. To address this gap, this study proposes a parallel multi-legged control system that integrates a Backpropagation Neural Network (BPNN) with a decoupled VMC framework. The BPNN enables adaptive tuning of motion parameters under varying modular configurations, while the decoupled VMC ensures stable gait control under force feedback. Simulation and physical experiments demonstrate that the proposed system achieves a unified control architecture across quadrupedal and multi-legged configurations, with improved tracking accuracy, stability, and adaptability compared to traditional VMC methods. Full article

Legged robots exhibit high mobility on uneven terrain but face challenges in stability, complex control systems, and energy efficiency. This study proposes a leg mechanism that significantly alters its stiffness by inducing jamming in a chainmail structure through only gravity-induced compression. To evaluate the fundamental characteristics of the proposed mechanism, experiments were conducted to identify the jamming point and to assess stiffness in the jammed state. The results confirmed that the force required to trigger jamming increases proportionally with the mass applied from above, which demonstrates properties similar to friction between solid materials. Furthermore, the stiffness in the jammed state is strongly correlated with the contact points within the structure. These results prove the effectiveness of the proposed passive leg mechanism for stiffness switching. In a case study assuming landing on uneven terrain, the mechanism could be fixed in any orientation based on the designed compressive force. Full article

Legged robot locomotion remains a critical challenge in robotics, demanding control strategies that are not only dynamically stable and robust but also capable of adapting to complex and changing environments. deep reinforcement learning (DRL) has recently emerged as a powerful approach to automatically generate motion control policies by learning from interactions with simulated or real environments. This study provides a systematic overview of DRL applications in legged robot control, emphasizing experimental platforms, measurement techniques, and benchmarking practices. Following PRISMA guidelines, 27 peer-reviewed studies published between 2018 and 2025 were analyzed, covering model-free, model-based, hierarchical, and hybrid DRL frameworks. Our findings reveal that reward shaping, policy representation, and training stability significantly influence control performance, while domain randomization and dynamic adaptation methods are essential for bridging the simulation-to-real-world gap. In addition, this review highlights instrumentation approaches for evaluating algorithm effectiveness, offering insights into sample efficiency, energy management, and safe deployment. The results aim to guide the development of reproducible and experimentally validated DRL-based control systems for legged robots. Full article

Accurate gait-phase detection is essential for rehabilitation monitoring, prosthetic control, and human–robot interaction. Artificial intelligence supports continuous, personalized mobility assessment by extracting clinically meaningful patterns from wearable sensors. A richer view of gait dynamics can be achieved by integrating additional signals, including inertial, plantar flex, footswitch, and EMG data, leading to more accurate and informative gait analysis. Motivated by these needs, this study investigates discrete gait-phase recognition for the right leg using a multi-subject IMU dataset collected from lower-limb sensors. IMU recordings were segmented into 128-sample windows across 23 channels, and each window was flattened into a 2944-dimensional feature vector. To ensure reliable ground-truth labels, we developed an automatic relabeling pipeline incorporating heel-strike and toe-off detection, adaptive threshold tuning, and sensor fusion across sensor modalities. These windowed vectors were then used to train a comprehensive suite of machine learning models, including Random Forests, Extra Trees, k-Nearest Neighbors, XGBoost, and LightGBM. All models underwent systematic hyperparameter tuning, and their performance was assessed through k-fold cross-validation. The results demonstrate that tree-based ensemble models provide accurate and stable gait-phase classification with accuracy exceeding 97% across both test sets, underscoring their potential for future real-time gait analysis and lower-limb assistive technologies. Full article