Learning Locomotion on Complex Terrain for Quadrupedal Robots with Foot Position Maps and Stability Rewards

TL;DR

This paper presents a reinforcement learning approach for quadrupedal robots that uses foot position maps and stability rewards to improve locomotion over complex terrains, ensuring better precision and stability.

Contribution

It introduces a novel integration of foot position maps and a stability reward within an attention-based framework for enhanced terrain locomotion.

Findings

Improved success rates on in-domain terrains.

Effective generalization to out-of-domain terrains.

Enhanced stability and foot placement accuracy.

Abstract

Quadrupedal locomotion over complex terrain has been a long-standing research topic in robotics. While recent reinforcement learning-based locomotion methods improve generalizability and foot-placement precision, they rely on implicit inference of foot positions from joint angles, lacking the explicit precision and stability guarantees of optimization-based approaches. To address this, we introduce a foot position map integrated into the heightmap, and a dynamic locomotion-stability reward within an attention-based framework to achieve locomotion on complex terrain. We validate our method extensively on terrains seen during training as well as out-of-domain (OOD) terrains. Our results demonstrate that the proposed method enables precise and stable movement, resulting in improved locomotion success rates on both in-domain and OOD terrains.