A Genetic Algorithm Based Architecture for Evolving Type-2 Fuzzy Logic Controllers for Real World Autonomous Mobile Robots

The type-2 Fuzzy Logic Controller (FLC) has started to emerge as a promising control mechanism for autonomous mobile robots navigating in real world environments. This is because such robots need control mechanisms such as type-2 FLCs which can handle the large amounts of uncertainties present in real world environments. However, manually designing and tuning the type-2 Membership Functions (MFs) for an interval type-2 FLC to give a good response is a difficult task. This paper will present a Genetic Algorithm (GA) based architecture to evolve the type-2 MFs of interval type-2 FLCs for mobile robots that will navigate in real world environments. The GA based system converges after a small number of iterations to type-2 MFs which give a very good performance. We have performed a series of real world experiments in which the evolved type-2 FLCs controlled a real robot in an outdoor arena. The evolved type-2 FLCs dealt with the uncertainties present in the real world to give a very good performance that has outperformed their type-1 counterparts as well as the manually designed type-2 FLCs.