A CPG-Based Agile and Versatile Locomotion Framework Using Proximal Symmetry Loss

TL;DR

This paper presents a novel humanoid robot locomotion framework combining a CPG-based walk engine with reinforcement learning and a proximal symmetry loss to achieve agile, versatile, and human-like walking on complex terrains.

Contribution

It introduces a proximal symmetry loss function to improve sample efficiency and robustness of the reinforcement learning process in humanoid locomotion.

Findings

Robot demonstrated human-like agility and versatility in simulations.

Framework effectively handled external disturbances and noise.

Enhanced sample efficiency due to the symmetry loss function.

Abstract

Humanoid robots are made to resemble humans but their locomotion abilities are far from ours in terms of agility and versatility. When humans walk on complex terrains, or face external disturbances, they combine a set of strategies, unconsciously and efficiently, to regain stability. This paper tackles the problem of developing a robust omnidirectional walking framework, which is able to generate versatile and agile locomotion on complex terrains. The Linear Inverted Pendulum Model and Central Pattern Generator concepts are used to develop a closed-loop walk engine, which is then combined with a reinforcement learning module. This module learns to regulate the walk engine parameters adaptively, and generates residuals to adjust the robot's target joint positions (residual physics). Additionally, we propose a proximal symmetry loss function to increase the sample efficiency of the Proximal Policy Optimization algorithm, by leveraging model symmetries and the trust region concept. The effectiveness of the proposed framework was demonstrated and evaluated across a set of challenging simulation scenarios. The robot was able to generalize what it learned in unforeseen circumstances, displaying human-like locomotion skills, even in the presence of noise and external pushes.