AGV System Design using Competitive and Cooperative Co-evolution

Design process of robust flow-path network and transporter routing for AGV systems is proposed in this paper. With robust network and routing, an effectiveness of a system does not sink in any task. However, for this robust system, the number of possible tasks is very large in AGV systems, therefore we cannot test the promising system with all of possible tasks. The problem is solved by the method of difficult task detection with genetic algorithm (GA). The effective system which has very large searching space is designed with cooperative co-evolution simultaneously, because the difficult tasks depend on the systems. We apply competitive co-evolution to the simultaneous design