Evolution of neural controllers for robot formation

In this paper, we present a new method for multiple robots formation, which means certain geometrical constrains on the relative positions and orientations of the robots throughout their travel. In our method, we apply multiobjective evolutionary computation to generate the neural networks that control the robots to get to the target position relative to the leader robot. The advantage of the proposed algorithm is that in a single run of multiobjective evolution are generated multiple neural controllers. We can select neural networks that control each robot to get to the target position relative to the leader robot. In addition, the robots can switch between neural controllers, therefore creating different geometrical formations. The simulation and experimental results show that the multiobjective-based evolutionary method can be applied effectively for generating neural networks which enable the robots to perform formation tasks.