Fuzzy radial basis function neural network PID control system for a quadrotor UAV based on particle swarm optimization

For quadrotor UAV is a kind of under-actuated system with multiple inputs and strong coupling, manual optimization of PID control parameter for the quadrotor aircraft is time-consuming, and it is difficult to achieve good control effect. Therefore, the fuzzy radial basis function (RBF) neural network PID control system for a quadrotor UAV based on particle swarm optimization (PSO) is designed to realize steady control of the system. In this control system, the control parameters of PID controller were adaptively adjusted by fuzzy RBF neural network, the control parameters of system were optimized by the hybrid learning methods integrating the offline particle swarm optimization (PSO) algorithm with the online Gradient Descent algorithm of local searching ability. The Matlab/Simulink simulation results show that the control system solve the problems of complex object control like the quadrotor UAV effectively, its parameter optimization algorithms can quickly converged and cannot be easily fallen to partial minimum, the control method has fast tracking ability, small overshoot, good decoupling ability, strong robust and adaptive ability, and it has better quality than the traditional PID coupling control method.