Answers are grounded in this paper and cite their sources.
Contrary to previously held notions, recent psychological experiments suggest that the human tactile system is both a fast and accurate recognition device for real objects. It seems reasonable, then, to build a robotic tactile perception system for the purpose of object identification. Toward this end, we present a set of low level tactile primitives and the exploratory and analytic procedures used to identify and extract them. We then discuss how these primitives may be combined into tactile features and, finally, how these features might be utilized by the robot perceptual system as a whole.
No takes yet. Share an insight, caveat, or question.
Sharon Stansfield (1986) studied this question.
Synapse has enriched one closely related paper. Consider it for comparative context: