Terrain-adaptive Central Pattern Generators with Reinforcement Learning for Hexapod Locomotion

TL;DR

This paper introduces a novel terrain-adaptive control method for hexapod robots that combines central pattern generators with deep reinforcement learning, improving adaptability and stability on complex terrains.

Contribution

It integrates DRL with CPGs to enhance terrain adaptability in legged robots, a novel approach compared to traditional fixed-pattern control methods.

Findings

Improved terrain adaptability in hexapod locomotion

Faster convergence rate in learning process

Simplified reward design for reinforcement learning

Abstract

Inspired by biological motion generation, central pattern generators (CPGs) is frequently employed in legged robot locomotion control to produce natural gait pattern with low-dimensional control signals. However, the limited adaptability and stability over complex terrains hinder its application. To address this issue, this paper proposes a terrain-adaptive locomotion control method that incorporates deep reinforcement learning (DRL) framework into CPG, where the CPG model is responsible for the generation of synchronized signals, providing basic locomotion gait, while DRL is integrated to enhance the adaptability of robot towards uneven terrains by adjusting the parameters of CPG mapping functions. The experiments conducted on the hexapod robot in Isaac Gym simulation environment demonstrated the superiority of the proposed method in terrain-adaptability, convergence rate and reward design complexity.