Research of control method based on 3-DOF twin rotor MIMO system

The paper discusses the designing of neural network controller for a twin rotor MIMO (multi-input multi-output) system. The controller is designed such that change in one degree should have minimum effect on other and overall the system should remain stable. For this system commonly known as the helicopter problem, many traditional controllers exist. RBF (radial basis function) neural network and CMAC (cerebella model articulation controller) controllers are designed separately in this paper. The two controllers had better or at least equal performance compared with traditional controllers. We realize the real-time control of helicopter system based on SIMULINK and RTW (real-time workshop). Simulation results show that the system performance with neural network controller is superior to the system performance with traditional PID or LQR controller.