Concurrent Training of a Control Policy and a State Estimator for Dynamic and Robust Legged Locomotion
This addresses robust locomotion for legged robots in challenging environments, though it appears incremental as it combines existing simulation-to-real and concurrent training approaches.
The paper tackles the problem of robust legged locomotion by concurrently training a control policy and state estimator in simulation, then transferring them to a real robot. The result is a system capable of traversing diverse terrains like hills and slippery plates, achieving speeds up to 3.75 m/s on flat ground and 3.54 m/s on low-friction surfaces.
In this paper, we propose a locomotion training framework where a control policy and a state estimator are trained concurrently. The framework consists of a policy network which outputs the desired joint positions and a state estimation network which outputs estimates of the robot's states such as the base linear velocity, foot height, and contact probability. We exploit a fast simulation environment to train the networks and the trained networks are transferred to the real robot. The trained policy and state estimator are capable of traversing diverse terrains such as a hill, slippery plate, and bumpy road. We also demonstrate that the learned policy can run at up to 3.75 m/s on normal flat ground and 3.54 m/s on a slippery plate with the coefficient of friction of 0.22.