This paper presents genetic algorithms (GAs) to perform the optimal design of an auto-tuning fuzzy proportional-integral-derivative (PID) controller and to determine the minimal number of fuzzy rules si- multaneously. Different from PID controllers with fixed gains, the fuzzy PID controller is expressed in terms of fuzzy rules, in which the input variables are the error signals and their derivatives, while the out- put variables are the PID gains. Based on the pro-posed GAs, the centers and the widths of the Gaus-sian membership functions, the fuzzy control rules corresponding to every possible combination of input linguistic variables, and the PID gains are chosen as parameters to be determined. When defining the fit-ness function of the GA, the concept of mul-ti-objective optimization is used such that the fitness function can be defined in a systematic way. To show the effectiveness and validity of the designed fuzzy PID controller, a typical benchmark, a multivariable seesaw system, is used for illustration.