Learning Robust, Agile, Natural Legged Locomotion Skills in the Wild

TL;DR

This paper introduces a novel reinforcement learning framework that produces robust, agile, and natural legged robot gaits for challenging terrains by integrating adversarial training with real animal data, improving sim-to-real transfer.

Contribution

It presents a new adversarial training approach combined with a teacher-student pipeline for more natural and robust locomotion skills in legged robots.

Findings

Robust traversal of stairs, rocky, and slippery terrains achieved.

Gaits are more natural, agile, and energy-efficient than baseline methods.

Effective sim-to-real transfer demonstrated on a quadruped robot.

Abstract

Recently, reinforcement learning has become a promising and polular solution for robot legged locomotion. Compared to model-based control, reinforcement learning based controllers can achieve better robustness against uncertainties of environments through sim-to-real learning. However, the corresponding learned gaits are in general overly conservative and unatural. In this paper, we propose a new framework for learning robust, agile and natural legged locomotion skills over challenging terrain. We incorporate an adversarial training branch based on real animal locomotion data upon a teacher-student training pipeline for robust sim-to-real transfer. Empirical results on both simulation and real world of a quadruped robot demonstrate that our proposed algorithm enables robustly traversing challenging terrains such as stairs, rocky ground and slippery floor with only proprioceptive perception. Meanwhile, the gaits are more agile, natural, and energy efficient compared to the baselines. Both qualitative and quantitative results are presented in this paper.