This paper aims to provide an approach of modeling haptic impressions of surfaces over a wide range of applications by using multiple sensor sources.
A multisensory measurement experiment was conducted using various leather and artificial leather surfaces. After processing of measurement data and feature extraction, different learning algorithms were applied to the measurement data and a corresponding set of data from a sensory study. The study contained evaluations of the same surfaces regarding descriptors of haptic quality (e.g. roughness) by human subjects and was conducted in a former research project.
The research revealed that it is possible to model and project haptic impressions by using multiple sensor sources in combination with data fusion. The presented method possesses the potential for an industrial application.
This paper provides a new approach to predict haptic impressions of surfaces by using multiple sensor sources.
