Deep Multi-Modal Network Based Data-Driven Haptic Textures Modeling

Abstract

This paper presents a novel data-driven approach for haptic texture modeling using a deep multi-modal network. The network is trained using contact acceleration data that are collected when a stylus is scanned on a textured surface with diverse scanning velocities, directions, and forces, which used for recreating the acceleration profile in real-time. We present some preliminary results to demonstrate the effectiveness of the proposed approach.