Evolutionary algorithm based neural network controller with selective sensor usage for autonomous mobile robot navigation

This paper deals with designing a neural network based navigator that is optimized in a user-defined sense for a mobile robot using ultrasonic sensors to travel to a goal position safely and efficiently without any prior map of the environment. The neural network has a dynamically reconfigurable structure that not only can optimize the weights but also the input sensory connectivity in order to meet any user-defined objective. Further, in order to enhance generalization capability of a single network, a modular network is used in which each network module is optimized for a specific local environment based on environment classification. After training all the modules, competitive and cooperative module coordination methods are applied and compared. Both computer simulation and real experiments show the effective performance of the algorithm