Robust Robot Walker: Learning Agile Locomotion over Tiny Traps

TL;DR

This paper presents a novel proprioception-based learning framework enabling quadruped robots to robustly navigate tiny traps, improving practical agility without relying on unreliable external sensors.

Contribution

We introduce a two-stage training framework with a contact encoder and classification head for trap detection, along with a new benchmark for tiny trap navigation.

Findings

Effective in simulation and real-world tests

Improves robustness over existing methods

Facilitates practical quadruped deployment

Abstract

Quadruped robots must exhibit robust walking capabilities in practical applications. In this work, we propose a novel approach that enables quadruped robots to pass various small obstacles, or "tiny traps". Existing methods often rely on exteroceptive sensors, which can be unreliable for detecting such tiny traps. To overcome this limitation, our approach focuses solely on proprioceptive inputs. We introduce a two-stage training framework incorporating a contact encoder and a classification head to learn implicit representations of different traps. Additionally, we design a set of tailored reward functions to improve both the stability of training and the ease of deployment for goal-tracking tasks. To benefit further research, we design a new benchmark for tiny trap task. Extensive experiments in both simulation and real-world settings demonstrate the effectiveness and robustness of our method. Project Page: https://robust-robot-walker.github.io/