Push Recovery

Push recovery research focuses on enabling robots, and exoskeletons assisting humans, to regain balance after unexpected disturbances. Current efforts concentrate on developing robust control strategies, often employing model predictive control (MPC) or reinforcement learning (RL) algorithms, sometimes incorporating hierarchical architectures or signal temporal logic (STL) for improved safety and task completion. These advancements are crucial for enhancing the safety and reliability of robots in various applications, from humanoid assistants to lower-limb exoskeletons, and contribute to a deeper understanding of dynamic balance control in both robotic and biological systems.