Legged robots, especially quadruped robots, are widely used in various environments due to their advantage in overcoming rough terrains. However, falling is inevitable. Therefore, the ability to overcome a falling state is an essential ability for legged robots. In this paper, we propose a method to fully recover a quadruped robot from a fall using a single-neural network model. The neural network model is trained in two steps in simulations using reinforcement learning, and then directly applied to AiDIN-VIII, a quadruped robot with 12 degrees of freedom. Experimental results using the proposed method show that the robot can successfully recover from a fall within 5 s in various postures, even when the robot is completely turned over. In addition, we can see that the robot successfully recovers from a fall caused by a disturbance.

Path planning in obstacle-dense environments is a challenging problem, particularly for robots with asymmetric rectangular footprints. To address this problem, we propose a novel collision-checking approach, called a Rotatable Area, which represents a range of heading angles where the robot can rotate without colliding with obstacles. Based on the relationship between two rotatable areas, we define safe local motion and extend this concept to the RoA-Planner, a path planning framework in SE(2) dense space. We validate our planner through extensive simulations and real-world experiments in complex and narrow environments. The results demonstrate that our method achieves fast planning speed while ensuring safety and robustness, making it suitable for practical applications.

A manipulator mounted on a mobility platform requires a design that simultaneously meets the demands for lightweight, compactness, and energy efficiency. A gravity compensator is an effective solution that mechanically offsets the gravity torque caused by the weight of the manipulator links, enabling the use of low-capacity motors while maintaining performance comparable to conventional systems. However, existing gravity compensators face notable challenges in terms of spatial efficiency and durability. To address these challenges, this study developed the two-DOF simultaneous gravity compensator based on a spring and slider-crank mechanism to compensate for the gravity torque generated by two consecutive pitch joints. Experimental results demonstrated a high torque compensation rate of 71.58% and a power reduction rate of 32.23% across the full range of motion of the two joints, further confirming consistency with theoretical predictions.

Quadruped robots exhibit superior terrain adaptability and mobility, enabling autonomous navigation in complex environments. However, path planning for quadruped robots is inherently challenging due to the non-circular and high-dimensional nature of their collision model, which introduces a trade-off between collision detection accuracy and planning robustness. To overcome this challenge, we propose the concept of a 3D Rotatable Area (3D RoA). The 3D RoA represents the range of collision-free and stable yaw angles at a given position, explicitly capturing local terrain geometry and collision constraints. Based on the 3D RoA, we further present a path planning framework that efficiently recognizes and navigates highly cluttered environments. Our approach generates traversable and collision-free SE(3) paths through 2D search, reducing the search space while accurately handling the collision geometry of the quadruped robot. The effectiveness of the proposed framework is validated through extensive simulations and real-world experiments using multiple platforms (Video: https://youtu.be/v-w3oquQ16A).

Reinforcement learning (RL) has become a powerful tool for quadrupedal locomotion, and a sim-to-real approach is widely adopted to avoid hardware damage during training. However, the “sim-to-real gap” remains a critical challenge, particularly for robots driven by high-gear-ratio actuators, in which nonlinear friction effects are strongly amplified. Conventional methods, such as actuator networks or heuristic domain randomization, often require specialized sensors or extensive trial-and-error to tune appropriate randomization ranges. Building on a recent system-identification framework for actuator dynamics, we extend it with an augmented friction model that incorporates the Stribeck effect to capture the low-velocity nonlinearities characteristic of high-gear-ratio actuators. The physical parameters are identified from real-robot trajectory data using an evolutionary algorithm, and the resulting simulation is used to train a locomotion policy that is transferred zero-shot to a 55 kg quadruped without additional fine-tuning or base- or controller-level dynamics randomization. On our platform, adding the Stribeck term lowers the actuator identification error by 16% relative to a Coulomb–Viscous model on the trajectory used for identification, and this advantage generalizes to an unseen trajectory not used for identification. It also lowers the simulation-to-reality mean-velocity degradation from 40.2% and 26.8% for the Coulomb–Viscous model to 27.2% and 21.6% for our method at the 0.3 and 1.0 m/s commands, respectively. The trained policy achieves stable locomotion on flat ground as well as rough terrain including steps and stairs. These results indicate that explicitly modeling low-velocity friction is beneficial for high-fidelity sim-to-real transfer in high-reduction systems.