Robust Contact State Estimation in Humanoid Walking Gaits

TL;DR

This paper introduces a deep learning-based framework for robustly detecting leg contact states in humanoid robots, generalizing across surfaces and platforms, and validated both in simulation and real-world tests.

Contribution

It presents a novel, simulation-trained deep learning approach for contact detection that generalizes well and is easily transferable to real robots.

Findings

Accurately detects contact states in various conditions

Generalizes across different surfaces and robot platforms

Effective in real-world humanoid robot experiments

Abstract

In this article, we propose a deep learning framework that provides a unified approach to the problem of leg contact detection in humanoid robot walking gaits. Our formulation accomplishes to accurately and robustly estimate the contact state probability for each leg (i.e., stable or slip/no contact). The proposed framework employs solely proprioceptive sensing and although it relies on simulated ground-truth contact data for the classification process, we demonstrate that it generalizes across varying friction surfaces and different legged robotic platforms and, at the same time, is readily transferred from simulation to practice. The framework is quantitatively and qualitatively assessed in simulation via the use of ground-truth contact data and is contrasted against state of-the-art methods with an ATLAS, a NAO, and a TALOS humanoid robot. Furthermore, its efficacy is demonstrated in base estimation with a real TALOS humanoid. To reinforce further research endeavors, our implementation is offered as an open-source ROS/Python package, coined Legged Contact Detection (LCD).