A Data-driven Contact Estimation Method for Wheeled-Biped Robots

TL;DR

This paper introduces a data-driven contact estimation method for wheeled-biped robots that improves accuracy and efficiency by learning from real-robot torque data, eliminating the need for specialized sensors.

Contribution

The paper presents a novel Bayes filter-based contact estimator tailored for wheeled-biped robots, leveraging real-robot torque measurements for updates.

Findings

Outperforms deep-learning baselines in accuracy

More sample-efficient than existing methods

Effective in real-robot and simulation experiments

Abstract

Contact estimation is a key ability for limbed robots, where making and breaking contacts has a direct impact on state estimation and balance control. Existing approaches typically rely on gate-cycle priors or designated contact sensors. We design a contact estimator that is suitable for the emerging wheeled-biped robot types that do not have these features. To this end, we propose a Bayes filter in which update steps are learned from real-robot torque measurements while prediction steps rely on inertial measurements. We evaluate this approach in extensive real-robot and simulation experiments. Our method achieves better performance while being considerably more sample efficient than a comparable deep-learning baseline.