Neuromimetic sensorimotor processing

ESR5

Objectives

The tactile features that generate the space of tactile inputs correspond to specific contact conditions governed by contact mechanics encoded by the brain. At a fundamental level these features correspond to mathematical invariants which are hypothesized to be learned by the neural circuits of the brain, and for which we recently provided experimental evidence. In this project, we will use a (artificial) deep neural network to learn these invariants from neural recordings and data obtained from stimulating artificial skin.

Expected Results

Identification of neural coding invariants of haptic sensing.

Placement

Host institution: Actronika

Enrolments (in Doctoral degree): Lunds Universitet

Supervisors

Vincent Hayward, Etienne Burdet

Presentation of ESR5

PhD defence: May 7th 2024

My name is Aruna Ramasamy. I am 23 years old. I pursued my undergraduate studies in Electronics and Communication Engineering in Anna University in my home country India. My inquisitiveness in Machine learning and Robots landed me in a master’s titled Computer vision, Robotics and Machine learning at University of Surrey in United Kingdom. I am always gravitated towards research as it provides a platform to explore and enrich one’s knowledge. Eventually, I found my way as a research student in the INTUITIVE project. 

PhD goals

This thesis ventures into the domain of Neuromimetic Sensorimotor Processing – a field dedicated to bridging the divide between human sensory experience and artificial replication. At the core of this investigation is the endeavor to decode the language of touch, aiming to identify the so-called invariants that reveal the intricacies of texture felt by our fingertips. These invariants are stable features of tactile data that persist through various exploration conditions of touch, that can help us to reveal the brain’s method of recognizing and distinguishing between tactile sensations.

Results

Deliverable 3.1 Learning classifier system for high-dimensional sensorimotor data
Characterization of neuromimetic sensorimotor processing. Identification of tactile interaction invariants. Implement learning classifier system for high-dimensional data

Conference Article
Ramasamy, A.; Faux, D.; Hayward, V.; Auvray, M.; Job, X.; Kirsch, L. 
Human Self-touch vs Other-Touch Resolved by Machine Learning
EuroHaptics, 2022
DOI: 10.1007/978-3-031-06249-0_25