Generalized Locomotion in Out-of-distribution Conditions with Robust Transformer

TL;DR

This paper introduces ROLT, a transformer-based approach that enhances legged robot locomotion robustness in out-of-distribution conditions through body tokenization and a novel dropout strategy, enabling better generalization and noise resilience.

Contribution

The paper presents ROLT, a new transformer-based model with body tokenization and consistent dropout, improving robustness and generalization in unseen dynamic and perceptual conditions for legged robots.

Findings

ROLT outperforms existing methods in robustness.

The policy generalizes to unseen dynamic conditions.

The model handles challenging perceptual noise during testing.

Abstract

To succeed in the real world, robots must deal with situations that differ from those seen during training. Those out-of-distribution situations for legged robot mainly include challenging dynamic gaps and perceptual gaps. Here we study the problem of robust locomotion in such novel situations. While previous methods usually rely on designing elaborate training and adaptation techniques, we approach the problem from a network model perspective. Our approach, RObust Locomotion Transformer(ROLT),a variation of transformer,could achieve robustness in a variety of unseen conditions. ROLT introduces two key designs: body tokenization and consistent dropout. Body tokenization supports knowledge share across different limbs, which boosts generalization ability of the network. Meanwhile, a novel dropout strategy enhances the policy's robustness to unseen perceptual noise. We conduct extensive experiments both on quadruped and hexapod robots. Results demonstrate that ROLT is more robust than existing methods. Although trained in only a few dynamic settings, the learned policy generalizes well to multiple unseen dynamic conditions. Additionally, despite training with clean observations, the model handles challenging corruption noise during testing.