Applying evolution strategies to neural networks robot controller

Abstract

In this paper an evolution strategy (ES) is introduced, to learn weights of a neural network controller in autonomous robots. An ES is used to learn high-performance reactive behavior for navigation and collisions avoidance. The learned behavior is able to solve the problem in different environments; so, the learning process has proven the ability to obtain a specialized behavior. All the behaviors obtained have been tested in a set of environment and the capability of generalization is showed for each learned behavior. No subjective information about “how to accomplish the task” has been included in the fitness function. A simulator based on mini-robot Khepera has been used to learn each behavior.