Path planning for cooperating robots using a GA-fuzzy approach

Abstract

We consider the difficult path planning problem for a team of robots working in the same workspace. Their navigation movements are determined by the fuzzy logic controllers (FLCs) having a common knowledge base which consists of membership function distributions and fuzzy rules. Such an FLC requires the design of an appropriate knowledge base. We propose, in this paper, to automate this design task by use of a genetic algorithm (GA) which selects some good rules from a large rule base using the information of membership function distributions of the variables. Results of computer simulations are given which demonstrate the feasibility of this approach.