Sensorless Estimation of Contact Using Deep-Learning for Human-Robot Interaction

TL;DR

This paper presents a deep learning approach for sensorless contact estimation in human-robot interaction, focusing on accurately modeling friction hysteresis to improve joint torque sensing without external sensors.

Contribution

It introduces a novel long-term memory deep learning scheme and modifications to Residual Learning architecture for better dynamics identification and reduced inference time.

Findings

Accurately approximates static friction hysteresis.

Demonstrates robustness in joint and task compliance experiments.

Reduces inference time while maintaining high accuracy.

Abstract

Physical human-robot interaction has been an area of interest for decades. Collaborative tasks, such as joint compliance, demand high-quality joint torque sensing. While external torque sensors are reliable, they come with the drawbacks of being expensive and vulnerable to impacts. To address these issues, studies have been conducted to estimate external torques using only internal signals, such as joint states and current measurements. However, insufficient attention has been given to friction hysteresis approximation, which is crucial for tasks involving extensive dynamic to static state transitions. In this paper, we propose a deep-learning-based method that leverages a novel long-term memory scheme to achieve dynamics identification, accurately approximating the static hysteresis. We also introduce modifications to the well-known Residual Learning architecture, retaining high accuracy while reducing inference time. The robustness of the proposed method is illustrated through a joint compliance and task compliance experiment.