NEUROMORPHIC TACTILE PERCEPTION FOR PROSTHETIC AND ROBOTIC APPLICATIONS

Abstract

In this dissertation, we explore the possibility of utilizing biologically-inspired approaches for tactile perception. The objective is to develop an intelligent tactile sensory system for robotic and prosthetic applications. To accomplish this goal, we develop a neuromorphic tactile module by integrating sensory and processing circuits. Our system comprises a piezoresistive fabric material as the sensor that emulates the skin, an interface that produces spike patterns that emulate the neural signals from the mechanoreceptors in the skin, and an extreme learning Machine (ELM) chip for computing and pattern recognition. We demonstrate learning capability of this module in several tactile pattern recognition tasks, including object recognition and texture classification. We further explore the possibility of employing brain-inspired learning mechanisms to overcome some of limitations of current tactile sensing systems, most importantly lack of adaptability and life-long learning.