Towards Human Haptic Gesture Interpretation for Robotic Systems

TL;DR

This paper advances robotic touch gesture interpretation by defining new gesture classes, collecting a force dataset from a robotic arm, and demonstrating that neural networks on raw data outperform other methods in classifying human touch gestures.

Contribution

It introduces four new touch gesture classes, provides an extensive force dataset from a robotic arm, and compares various classification methods, highlighting neural networks' superior performance.

Findings

Neural network classifiers on raw data achieve highest accuracy.

Four new gesture classes effectively cover key gesture characteristics.

The dataset enables standardized evaluation of touch gesture recognition methods.

Abstract

Physical human-robot interactions (pHRI) are less efficient and communicative than human-human interactions, and a key reason is a lack of informative sense of touch in robotic systems. Interpreting human touch gestures is a nuanced, challenging task with extreme gaps between human and robot capability. Among prior works that demonstrate human touch recognition capability, differences in sensors, gesture classes, feature sets, and classification algorithms yield a conglomerate of non-transferable results and a glaring lack of a standard. To address this gap, this work presents 1) four proposed touch gesture classes that cover an important subset of the gesture characteristics identified in the literature, 2) the collection of an extensive force dataset on a common pHRI robotic arm with only its internal wrist force-torque sensor, and 3) an exhaustive performance comparison of combinations of feature sets and classification algorithms on this dataset. We demonstrate high classification accuracies among our proposed gesture definitions on a test set, emphasizing that neural net-work classifiers on the raw data outperform other combinations of feature sets and algorithms. The accompanying video is here: https://youtu.be/gJPVImNKU68