Automatic generation of Swarm Robotic behaviors using multi-objective evolution

Abstract

This paper investigates the utilization of a multiobjective approach for evolving artificial neural networks (ANNs) that act as controllers for a collective box-pushing task based on radio frequency (RF)-localization of a group of virtual E-puck robots simulated in a 3D, physics-based environment. The elitist Pareto-frontier Differential Evolution (PDE) algorithm is used to generate the Pareto optimal sets of ANN that optimize the conflicting objectives of maximizing the virtual Epuck robots’ behaviors for pushing a box towards a wall based on RF-localization as well as minimizing the number of hidden neurons used in its feed-forward ANN controller. A new fitness function which combines two different behaviors (1) RFlocalization behavior and (2) box-pushing behavior is also proposed. The experimentation results showed that the virtual Epuck robots were capable of moving towards to the target and thereafter push the box towards the target wall with very small neural network architecture. Hence, the results demonstrated that the utilization of the PDE approach in evolutionary robotics can be practically used to generate neural-based controllers that display intelligent collective behaviors in swarming autonomous mobile robots.