Generating a Terrain-Robustness Benchmark for Legged Locomotion: A Prototype via Terrain Authoring and Active Learning

TL;DR

This paper introduces a method to generate diverse, realistic terrains for legged robot locomotion testing using terrain authoring and active learning, aiming to improve evaluation benchmarks.

Contribution

It presents a novel terrain generation approach that produces high-quality, challenging terrains for benchmarking legged robot locomotion in simulation.

Findings

Generated terrains are diverse and realistic.

The dataset improves robustness testing of locomotion policies.

Code and datasets are publicly available.

Abstract

Terrain-aware locomotion has become an emerging topic in legged robotics. However, it is hard to generate diverse, challenging, and realistic unstructured terrains in simulation, which limits the way researchers evaluate their locomotion policies. In this paper, we prototype the generation of a terrain dataset via terrain authoring and active learning, and the learned samplers can stably generate diverse high-quality terrains. We expect the generated dataset to make a terrain-robustness benchmark for legged locomotion. The dataset, the code implementation, and some policy evaluations are released at https://bit.ly/3bn4j7f.