Research on mobile robot behaviors based on chaotic neural network

In order to improve the adaptability of the mobile robot in complex environment, this paper proposes novel method that chaotic neural network and genetic algorithm are applied to basic behaviors (such as obstacle avoidance, target approach and following wall) and their integrated behaviors. These basic behaviors and behavior switch that can be active sometimes will be controlled by the recurrent diagonal chaotic neural network, which is made up of multi basic module with the standard structure sub-network. Simulation results show that this method can effectively reduce complexity of the network and the code, and increase the evolution speed of mobile robot behaviors.