English

Zero-Shot Terrain Generalization for Visual Locomotion Policies

Robotics 2020-11-12 v1

Abstract

Legged robots have unparalleled mobility on unstructured terrains. However, it remains an open challenge to design locomotion controllers that can operate in a large variety of environments. In this paper, we address this challenge of automatically learning locomotion controllers that can generalize to a diverse collection of terrains often encountered in the real world. We frame this challenge as a multi-task reinforcement learning problem and define each task as a type of terrain that the robot needs to traverse. We propose an end-to-end learning approach that makes direct use of the raw exteroceptive inputs gathered from a simulated 3D LiDAR sensor, thus circumventing the need for ground-truth heightmaps or preprocessing of perception information. As a result, the learned controller demonstrates excellent zero-shot generalization capabilities and can navigate 13 different environments, including stairs, rugged land, cluttered offices, and indoor spaces with humans.

Keywords

Cite

@article{arxiv.2011.05513,
  title  = {Zero-Shot Terrain Generalization for Visual Locomotion Policies},
  author = {Alejandro Escontrela and George Yu and Peng Xu and Atil Iscen and Jie Tan},
  journal= {arXiv preprint arXiv:2011.05513},
  year   = {2020}
}
R2 v1 2026-06-23T20:04:07.545Z