Improving the fuzzy logic controller of a car racing competition with adjusted fuzzy sets

This paper describes the improvement of the Fuzzy Logic controller that won the car racing competition 2007 in the IEEE Congress on Evolutionary Computation (CEC2007). In particular, the use of the fuzzy sets within the original fuzzy controller of Ho and Garibaldi was improved. Several simulation experiments were conducted to optimize the membership functions of the fuzzy sets. The experiments are described and results discussed. The sum of different strategies qualify for an improvement of the original controller: Rearrangement of influence parameters, symmetric membership functions, and adjusted fuzzy rules. An overall improvement of 0.5 % in the competition score compared to the original winning controller is retrieved.

[1]  Jonathan M. Garibaldi,et al.  A novel fuzzy inferencing methodology for simulated car racing , 2008, 2008 IEEE International Conference on Fuzzy Systems (IEEE World Congress on Computational Intelligence).

[2]  Jesús Alcalá-Fdez,et al.  jFuzzyLogic: a Java Library to Design Fuzzy Logic Controllers According to the Standard for Fuzzy Control Programming , 2013, Int. J. Comput. Intell. Syst..

[3]  Jonathan M. Garibaldi,et al.  A fuzzy approach for the 2007 CIG simulated car racing competition , 2008, 2008 IEEE Symposium On Computational Intelligence and Games.

[4]  Tzuu-Hseng S. Li,et al.  Implementation of human-like driving skills by autonomous fuzzy behavior control on an FPGA-based car-like mobile robot , 2003, IEEE Trans. Ind. Electron..

[5]  Vicente Milanés Montero,et al.  Autonomous car fuzzy control modeled by iterative genetic algorithms , 2009, 2009 IEEE International Conference on Fuzzy Systems.

[6]  Julian Togelius,et al.  The 2007 IEEE CEC simulated car racing competition , 2008, Genetic Programming and Evolvable Machines.