BarlowWalk: Self-supervised Representation Learning for Legged Robot Terrain-adaptive Locomotion

TL;DR

BarlowWalk introduces a self-supervised learning approach integrated with PPO for legged robot terrain traversal, reducing training time and dependence on external perception while improving performance in complex terrains.

Contribution

It combines self-supervised representation learning with PPO for efficient, terrain-adaptive legged robot control, a novel integration in this context.

Findings

Significantly reduces training time compared to traditional RL methods.

Achieves better terrain adaptation in complex scenarios.

Verifies effectiveness through comparative simulation experiments.

Abstract

Reinforcement learning (RL), driven by data-driven methods, has become an effective solution for robot leg motion control problems. However, the mainstream RL methods for bipedal robot terrain traversal, such as teacher-student policy knowledge distillation, suffer from long training times, which limit development efficiency. To address this issue, this paper proposes BarlowWalk, an improved Proximal Policy Optimization (PPO) method integrated with self-supervised representation learning. This method employs the Barlow Twins algorithm to construct a decoupled latent space, mapping historical observation sequences into low-dimensional representations and implementing self-supervision. Meanwhile, the actor requires only proprioceptive information to achieve self-supervised learning over continuous time steps, significantly reducing the dependence on external terrain perception. Simulation experiments demonstrate that this method has significant advantages in complex terrain scenarios. To enhance the credibility of the evaluation, this study compares BarlowWalk with advanced algorithms through comparative tests, and the experimental results verify the effectiveness of the proposed method.