Self-regulating Membership Function Of The Fuzzy Controller Design Based On Particle Swarm Optimization

Particle swarm algorithm has good global optimization ability, local optimizationability, and robustness. The algorithm is simple, easy to implement and a little parametersneed to adjust. Because the algorithm searches the optimal value through the randomnumber, so it can provide a good solution to multiple optimal problems. In view of theadvantages mentioned above, particle swam algorithm has become an importantoptimization method, and it has attracted much attention in academic research field.Fuzzy controller is simple, robust, and suitable for nonlinear system control. It is akind of nonlinear controller based on expert experience, and there is no determinativemethod to unity the membership function and fuzzy control rules. In view of thetraditional fuzzy controller is difficult to determine the membership functions, the particleswarm algorithm (PSO) which combines the traditional fuzzy control method has beenpresented, and use this improved method, a fuzzy controller based on the PSO will bedesigned.Using the MATLAB tool to simulate the fuzzy PID control based on PSO and fuzzyPID control, the results can proof the feasibility of this new method. And then, usingFPGA and MCU to design a fuzzy controller which can regulate the membershipfunction by itself. In the design, PSO is applied by the MCU, and the fuzzy controller isapplied by the FPGA. The controller was used in motion control of a car model. The testresults show that, this method can ensure the stability in running process, realize thesmooth change for the control speed and control force, and it achieves the expectedcontrol target.