Netural internal model control of a mobile robot

Abstract EN

The use conventional approaches as Artificial natural network (ANN) in the internal model control (IMC) structure is considered in this paper.

An ANN is used in this model based control both as process model and as controller.

The natural network is trained with observed input-output data from the system, to present its inverse dynamic.

The resulting inverse model natural network can then be used as a controller, typically in a feed-forward fashion.

The proposed procedure is then applied to control a non linear complex system generally subjected to external constraints posed by the environment : the mobile robot.

Simulation results for different approaches show acceptable control performances with good correspondences between the desires and obtained trajectories.