Mixture-of-Experts RL for Fault-Tolerant Legged Locomotion
For legged robots in remote environments like planetary exploration, this work provides a modular approach to maintain locomotion under actuator faults, offering improved robustness and efficiency.
The paper proposes a fault-aware modular control architecture for legged robots that uses fault-diagnosis information to activate specialized experts for different actuator failures, outperforming monolithic policies in locomotion performance across failure scenarios, especially under reduced network capacity.
Legged robots deployed in planetary exploration and other remote environments must maintain reliable locomotion despite actuator failures and challenging terrain conditions. Although reinforcement learning has achieved strong results in legged locomotion, monolithic policies can struggle to efficiently represent the diverse control strategies required to compensate for different fault conditions. In this work, we propose a fault-aware modular control architecture that explicitly leverages fault-diagnosis information to activate specialized control experts associated with distinct actuator failure modes. Experimental results show that explicit fault-conditioned modular policies consistently outperform monolithic policies of comparable size, achieving higher locomotion performance across failure scenarios. Moreover, the proposed modular architecture retains competitive performance even under significantly reduced network capacity, highlighting its suitability for compute-constrained robotic platforms, such as those typically employed in space applications. The code associated with this work is available at: https://github.com/iit-DLSLab/fault-locomotion-isaaclab.