I’ve been invited to give a talk at the Robotics and the Living Conference conference that will be held on December 9th, 2013 at the university of Cergy-Pontoise.
About the conference
Intelligent systems require a body to interact with the environment depending on the context. This apparently simple proposal has strong implications for the understanding of living organisms in their complexity, but also for anyone who wants to design robots closer to living organisms in the way they operate.
In biology, sensory and motor information is integrated with both the morphology of the sensors and the physiology of the musculo-skeletal system, this integration is called “Simplexity” by Alain Berthoz. Simplexity also corresponds to the idea of reentry that perception is based on the act, rather than the action.
In robotics and AI, the idea that the physical structure (physical) and the function that underlies it (the computational processes) are connected is formulated by Rolf Pfeifer with the concept of “morphological computation”. Recent developments in new materials, sensors and actuators closest to human tissue, human sensory receptors and to human physiology will make possible a new bio- inspired robotics.
In cognitive sciences, Kevin O’Regan suggest that it is the sensorimotor contingencies and our action in the world that can permit us to perceive things through our body. The incarnation of perceptual activity means, at the same time, the structure and form of the living body plays a key role in the dynamics of interpersonal interaction. Foe Charles Lenay, our consideration of the body of others (the body image that everyone gives to the perception of its partners) can propose ways of original explanation of the recognition of others or of imitative expressions.
The counterpart of this research is the work of Nicolas Rougier in computational neuroscience, studying in particular the interaction between the brain, body and environment and offers an epistemological break in the concept model. Similarly, Philippe Gaussier offers computational neuroscience models for social and developmental robotics, or how neuronal dynamics and behaviors emerge from interaction of the robot in its physical and social environment.