Learning Terrain-Adaptive Locomotion with Agile Behaviors by Imitating Animals

TL;DR

This paper introduces a learning framework enabling quadruped robots to imitate animal behaviors and adapt to challenging terrains, achieving natural locomotion and effective terrain traversal through imitation learning and terrain adaptation.

Contribution

It presents a novel two-step learning approach combining imitation from animals and terrain adaptation for robust quadruped locomotion.

Findings

Successfully traverses various terrains including stairs

Achieves 1.1 m/s speed on stairs in real robot

Demonstrates natural and adaptive behaviors

Abstract

In this paper, we present a general learning framework for controlling a quadruped robot that can mimic the behavior of real animals and traverse challenging terrains. Our method consists of two steps: an imitation learning step to learn from motions of real animals, and a terrain adaptation step to enable generalization to unseen terrains. We capture motions from a Labrador on various terrains to facilitate terrain adaptive locomotion. Our experiments demonstrate that our policy can traverse various terrains and produce a natural-looking behavior. We deployed our method on the real quadruped robot Max via zero-shot simulation-to-reality transfer, achieving a speed of 1.1 m/s on stairs climbing.