Understanding the Stability of Deep Control Policies for Biped Locomotion
This work addresses stability challenges in biped locomotion for robotics, but it is incremental as it builds on existing DRL methods.
The study evaluated the push-recovery stability of deep reinforcement learning policies for biped locomotion, comparing them to human subjects and a previous feedback controller, and found that deep policies achieved improved gait stability with specific performance gains.
Achieving stability and robustness is the primary goal of biped locomotion control. Recently, deep reinforce learning (DRL) has attracted great attention as a general methodology for constructing biped control policies and demonstrated significant improvements over the previous state-of-the-art. Although deep control policies have advantages over previous controller design approaches, many questions remain unanswered. Are deep control policies as robust as human walking? Does simulated walking use similar strategies as human walking to maintain balance? Does a particular gait pattern similarly affect human and simulated walking? What do deep policies learn to achieve improved gait stability? The goal of this study is to answer these questions by evaluating the push-recovery stability of deep policies compared to human subjects and a previous feedback controller. We also conducted experiments to evaluate the effectiveness of variants of DRL algorithms.