Abstract: PID and CMAC compound controller was developed about thirty years ago but yet it lacks effective multi-parameter self-tuning method, including parameters of CMAC learning rate and inertia weight. Without proper control parameters, the control system based on PID and CMAC controller will converge very slowly, even becoming unstable after the simulation for a period of time. A new kind of multi-parameter self-adaptive optimization scheme based on adaptive genetic algorithm (AGA) was proposed for PID and CMAC controller. The central idea of self-adaptive multi-parameter optimization based on AGA is that PID parameters Kp, Ki, Kd and CMAC parameters η, α are regarded as a group of gene in GA. The PID and CMAC parameters are adjusted adaptively by AGA according to a certain objective function. Finally, the simulations show that the proposed control method can improve the performance effectively. It is a simple convenient and effective control scheme for the nonlinear and uncertain control system.