Multi-Class Human/Object Detection on Robot Manipulators using Proprioceptive Sensing

TL;DR

This paper develops and evaluates multi-class human/object detection models using proprioceptive sensing on robot manipulators, achieving over 91% accuracy and demonstrating the effectiveness of sliding window preprocessing for real-time contact analysis.

Contribution

It introduces a three-class detection framework with various neural network models and identifies optimal preprocessing strategies, advancing beyond binary classification in robot-human interaction safety.

Findings

Best model achieved 91.11% accuracy in real-time detection.

Sliding window preprocessing outperformed other strategies.

Multi-class detection provides more detailed contact information.

Abstract

In physical human-robot collaboration (pHRC) settings, humans and robots collaborate directly in shared environments. Robots must analyze interactions with objects to ensure safety and facilitate meaningful workflows. One critical aspect is human/object detection, where the contacted object is identified. Past research introduced binary machine learning classifiers to distinguish between soft and hard objects. This study improves upon those results by evaluating three-class human/object detection models, offering more detailed contact analysis. A dataset was collected using the Franka Emika Panda robot manipulator, exploring preprocessing strategies for time-series analysis. Models including LSTM, GRU, and Transformers were trained on these datasets. The best-performing model achieved 91.11\% accuracy during real-time testing, demonstrating the feasibility of multi-class detection models. Additionally, a comparison of preprocessing strategies suggests a sliding window approach is optimal for this task.