A neural network-based controller for homogeneous and heterogeneous distributed robotic systems

A neural network-based controller for robot members of a distributed robotic system is presented. It is shown that a four-layer network with robot perception as input and robot action as output can produce appropriate behaviors for members of a robot team, performing specific tasks. Also included in this paper are the results of experiments, comparing the performance of a homogeneous robot team with that of a team composed of heterogeneous members.