Robust Adaptive Single Neural Control for Yaw Angle with Input Nonlinearity on Helicopter Testbed

In this paper, we deal with the yaw control problem of a small-scale helicopter mounted on an experimental platform. The yaw dynamics of helicopter involve input nonlinearity, time-varying parameters and the couplings between main and tail rotor. An attractive control strategy that combines neural networks with traditional adaptive controls has been successfully used for yaw control with input nonlinearities. In contrast to conventional adaptation law, the sliding condition is taken as the objective function instead of the error function used in MIT rule. From the concept of the sliding mode control, the adaptive controller guarantees the stability of the closed-loop system and convergence of the output tracking error to a desired bound, even if the model parameters are unknown or in the presence of disturbance. The simulation results are further compared with those obtained by normal PID control to demonstrate the improvements of the proposed algorithm