Learning Free Gait Transition for Quadruped Robots via Phase-Guided Controller
This addresses the problem of versatile locomotion for quadruped robots, representing an incremental improvement in multi-task learning for legged robotics.
The paper tackles the challenge of enabling quadruped robots to perform multiple gaits and transitions by introducing a phase-guided controller, resulting in the Black Panther robot smoothly and robustly executing learned gaits like walk, trot, pacing, bounding, and complex movements in natural environments.
Gaits and transitions are key components in legged locomotion. For legged robots, describing and reproducing gaits as well as transitions remain longstanding challenges. Reinforcement learning has become a powerful tool to formulate controllers for legged robots. Learning multiple gaits and transitions, nevertheless, is related to the multi-task learning problems. In this work, we present a novel framework for training a simple control policy for a quadruped robot to locomote in various gaits. Four independent phases are used as the interface between the gait generator and the control policy, which characterizes the movement of four feet. Guided by the phases, the quadruped robot is able to locomote according to the generated gaits, such as walk, trot, pacing and bounding, and to make transitions among those gaits. More general phases can be used to generate complex gaits, such as mixed rhythmic dancing. With the control policy, the Black Panther robot, a medium-dog-sized quadruped robot, can perform all learned motor skills while following the velocity commands smoothly and robustly in natural environment.