Mobile Robot Behavior Controller Based on Genetic Diagonal Recurrent Neural Network

It is crucial that a robot should have both learning and evolutionary ability to adapt to dynamic environments. This paper proposes a new mobile robot behavior controller based on genetic algorithm (GA) and diagonal recurrent neural network (DRNN). The DRNN has the advantages of time series prediction capability because of its memory nodes, as well as local recurrent and self- feedback connections. Genetic algorithm is introduced to optimize the learning rate and the structure of DRNN in order to achieve better performance. Finally, the GA-DRNN is applied to the mobile robot behavior controller. Simulation results show that the controller based on GA-DRNN possesses higher precision, compared with controller based on DRNN.