Robust and Versatile Bipedal Jumping Control through Reinforcement Learning

Zhongyu Li
University of California, Berkeley
Xue Bin Peng
Simon Fraser University
Pieter Abbeel
University of California, Berkeley
Sergey Levine
University of California, Berkeley
Glen Berseth
Université de Montréal, Mila
Koushil Sreenath
University of California, Berkeley
Paper Website

Paper ID 52

Session 7. Mobile Manipulation and Locomotion

Poster Session Wednesday, July 12

Poster 20

Abstract: This work aims to push the limits of agility for bipedal robots by enabling a torque-controlled bipedal robot to perform robust and versatile dynamic jumps in the real world. We present a reinforcement learning framework for training a robot to accomplish a large variety of jumping tasks, such as jumping to different locations and directions. To improve performance on these challenging tasks, we develop a new policy structure that encodes the robot’s long-term input/output (I/O) history while also providing direct access to a short-term I/O history. In order to train a versatile jumping policy, we utilize a multi-stage training scheme that includes different training stages for different objectives. After multi-stage training, the policy can be directly transferred to a real bipedal Cassie robot. Training on different tasks and exploring more diverse scenarios lead to highly robust policies that can exploit the diverse set of learned maneuvers to recover from perturbations or poor landings during real-world deployment. Such robustness in the proposed policy enables Cassie to succeed in completing a variety of challenging jump tasks in the real world, such as standing long jumps, jumping onto elevated platforms, and multi-axes jumps.