Enhancing Tactile Sensor Technology with Neuromorphic Models and Machine Learning
Komali Lenka, Pulaparthi Lakshmi Asha Jyothi, Kallakuri N V P S Brahma Ramesh, Dasaradha Ramayya Lanka, M. Srikanth, Jonnapalli Tulasi Rajesh
Abstract
This study presents a novel approach to enhancing touch sensor technology by utilizing machine learning techniques and neuromorphic models. Tactile sensing has struggled to accurately understand complex physical interactions, despite its significance for artificial touch and robotic dexterity. Our approach improves the classification of tactile data by using neuromorphic computing-models inspired by brain processing-to identify changes in the environment, variations in force, and textures. We evaluate our model on benchmark tactile datasets and demonstrate its superior sensory resolution and increased adaptability. For robotics, prosthetics, and human-machine interfaces, the results are very important because they show that neuromorphic tactile sensors improve the accuracy and speed of data classification. This study showcases the potential of neuromorphic models to improve tactile perception, paving the way for more advanced and adaptable sensory systems