MUJICA: Multi-skill Unified Joint Integration of Control Architecture for Wheeled-Legged Robots
For wheeled-legged robot locomotion, MUJICA addresses the challenge of balancing wheeled and legged control with a unified framework that enhances sim-to-real robustness and enables autonomous skill selection, though the improvements are incremental over existing methods.
MUJICA is a unified proprioceptive control framework for wheeled-legged robots that integrates multiple low-level skills (omnidirectional moving, high platform climbing, fall recovery) into a single policy, trained with accurate DC-motor constraint modeling and a high-level skill selector. In real-world experiments on the Unitree Go2-W robot, it demonstrated significant improvements in adaptability and task success in unstructured environments.
Wheeled-legged robots hold promise for traversing complex terrains and offer superior mobility compared to legged robots. However, wheeled-legged robots must effectively balance both wheeled driving and legged control. Furthermore, due to noisy proprioceptive sensing and real-world motor constraints, realizing robust and adaptive locomotion at peak performance of motors remains challenging. We propose the Multi-skill Unified Joint Integration of Control Architecture (MUJICA), a unified, fully proprioceptive control framework for wheeled-legged robots that integrates diverse low-level skills-including omnidirectional moving, high platform climbing, and fall recovery-within a single policy. All skills, distinguished by unique indicator variables, are trained jointly with accurate DC-motor constraint modeling. Additionally, a high-level skill selector is learned to dynamically choose the optimal skill based solely on proprioceptions, enabling adaptive responses to the surrounding environment. Therefore, MUJICA enhances sim-to-real robustness and enables seamless transitions across diverse locomotion modes, facilitating autonomous adjustment to the environment. We validate our framework in both simulation and real-world experiments on the Unitree Go2-W robot, demonstrating significant improvements in adaptability and task success in unstructured environments.