D Shaikh

Neurorobotics: an embodied approach to understanding the neural basis of adaptive behavior

Neurorobotics is the study of neural systems embodied in robots.

The goal behind building any neurorobotic system is to test biologically inspired hypotheses regarding the functioning of the chosen neural system. This is done by building a computational model of a neural system of choice, embodying it in a (neuro)robot and then verifying the ability of the system to generate desired robot behaviour in a relevant task in a real environment.

Neurorobotics addresses questions in neuroethology regarding the neural bases of sensory perception, legged locomotion, learning and memory. Neurorobotics is based on the paradigm of “embodiment” – the functioning of the brain is shaped by the interaction of the body with the environment it is situated in as well as the properties of the specific ecological niche that the body inhabits.

In this talk I will first present the neurorobotics approach in general. I will then present examples from my work within neurorobotics, addressing fundamental topics in neuroscience such as sensorimotor integration, multisensory integration, crossmodal learning, neural spatial maps and working memory. I will ground the examples in neuroethologically relevant behaviours such as phonotaxis, target tracking and navigation.

 

Invited by David Robbe

 

Monday 9th April – 11 a.m.

Inmed conference room

Partager l'article