Decentralized PID neural network control for a quadrotor helicopter

A decentralized PID neural network (PIDNN) control scheme is proposed to solve the nonlinear control problems in quadrotor helicopter. First, the dynamic model is established via Newton-Euler formalism. Then, a nested loop control approach is proposed to solve the stabilization and navigation problems in the quadrotor. A decentralized PIDNN controller is designed for the inner loop to stabilize the attitude angle. A conventional PID controller is used for the outer loop in order to generate the reference path for the inner loop. Moreover, the connective weights of the PIDNN are trained on-line by error back-propagation method. The experiment is made to study the performance of controller based on the independently developed quadrotor helicopter system, which shows that the controller has good stability, maneuverability and robustness.