Learning Humanoid Locomotion over Challenging Terrain

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Radosavovic, Ilija, Kamat, Sarthak, Darrell, Trevor, Malik, Jitendra
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914964803944448
author Radosavovic, Ilija
Kamat, Sarthak
Darrell, Trevor
Malik, Jitendra
author_facet Radosavovic, Ilija
Kamat, Sarthak
Darrell, Trevor
Malik, Jitendra
contents Humanoid robots can, in principle, use their legs to go almost anywhere. Developing controllers capable of traversing diverse terrains, however, remains a considerable challenge. Classical controllers are hard to generalize broadly while the learning-based methods have primarily focused on gentle terrains. Here, we present a learning-based approach for blind humanoid locomotion capable of traversing challenging natural and man-made terrain. Our method uses a transformer model to predict the next action based on the history of proprioceptive observations and actions. The model is first pre-trained on a dataset of flat-ground trajectories with sequence modeling, and then fine-tuned on uneven terrain using reinforcement learning. We evaluate our model on a real humanoid robot across a variety of terrains, including rough, deformable, and sloped surfaces. The model demonstrates robust performance, in-context adaptation, and emergent terrain representations. In real-world case studies, our humanoid robot successfully traversed over 4 miles of hiking trails in Berkeley and climbed some of the steepest streets in San Francisco.
format Preprint
id arxiv_https___arxiv_org_abs_2410_03654
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Learning Humanoid Locomotion over Challenging Terrain
Radosavovic, Ilija
Kamat, Sarthak
Darrell, Trevor
Malik, Jitendra
Robotics
Machine Learning
Humanoid robots can, in principle, use their legs to go almost anywhere. Developing controllers capable of traversing diverse terrains, however, remains a considerable challenge. Classical controllers are hard to generalize broadly while the learning-based methods have primarily focused on gentle terrains. Here, we present a learning-based approach for blind humanoid locomotion capable of traversing challenging natural and man-made terrain. Our method uses a transformer model to predict the next action based on the history of proprioceptive observations and actions. The model is first pre-trained on a dataset of flat-ground trajectories with sequence modeling, and then fine-tuned on uneven terrain using reinforcement learning. We evaluate our model on a real humanoid robot across a variety of terrains, including rough, deformable, and sloped surfaces. The model demonstrates robust performance, in-context adaptation, and emergent terrain representations. In real-world case studies, our humanoid robot successfully traversed over 4 miles of hiking trails in Berkeley and climbed some of the steepest streets in San Francisco.
title Learning Humanoid Locomotion over Challenging Terrain
topic Robotics
Machine Learning
url https://arxiv.org/abs/2410.03654