GA-Aided Elman Neural Network Controller For Behavior-Based Robot

Multi-robot systems differ from single robot systems mostly in that the environments can be affected by other robots. So we can consider every robot in dynamic environments. Therefore it is crucial that each robot should have both learning and evolutionary ability to adapt to dynamic environments. This paper proposes a new robot behavior decision controller using Elman neural network (Elman NN) and genetic algorithm (GA).The Elman NN has the advantages of time series prediction capability because of its memory nodes, as well as local recurrent connections. Genetic algorithm (GA) is introduced to determine the connection weight values of Elman NN in order to achieve better behavior performance. The computer simulation is given to show the validity of the method