Learning for the Control of Dynamical Motion Systems

Pierre-François Marteau (1) , Sylvie Gibet (1)

1 VALORIA - Laboratoire de Recherche en Informatique et ses Applications de Vannes et Lorient

This paper addresses the dynamic control of multijoint systems based on learning of sensory-motor transformations. To avoid the dependency of the controllers to the analytical knowledge of the multijoint system, a non parametric learning approach is developed which identifies non linear mappings between sensory signals and motor commands involved in control motor systems. The learning phase is handled through a General Regression Neural Network (GRNN) that implements a non parametric Nadarayan-Watson regression scheme and a set of local PIDs. The resulting dynamic sensory-motor controller (DSMC) is intensively tested within the scope of hand-arm reaching and tracking movements in a dynamical simulation environment. (DSMC) proves to be very effective and robust. Moreover, it reproduces kinematics behaviors close to captured hand-arm movements.

Mots clés: Generalized Regression Neural Network Senori-motor mapping Hand-arm movement

Domaines: Recherche d'information [cs.IR]

Pierre-François Marteau, Sylvie Gibet. Learning for the Control of Dynamical Motion Systems. Seventh International Conference on Intelligent Systems Design and Applications (ISDA 2007), Oct 2007, Rio, Brazil. pp.454-459, ⟨10.1109/ISDA.2007.27⟩. ⟨hal-00492829⟩