Fault-Tolerant Neural Network Flight Control for Helicopter Altitude and Attitude Systems Under Output Constraints

In this paper, an adaptive radial basis function neural network flight control scheme is proposed for the unmanned helicopter altitude and attitude systems. For the flying helicopters, the issues of modeling error, output constraints and actuator failure cannot be ignored when constructing the flight controller, which have great impact on the control performance of unmanned helicopters. To solve these problems, the adaptive radial basis function neural network method is utilized to approximate the missing parts in the input channel due to modeling errors and actuator failures. Then, the barrier Lyapunov function technology is used to constrain the control output, and the flight controller is designed via uniting the backstepping control scheme. In which, the virtual control law is adjusted according to the tracking errors, and the online weight is updated by using the adaptive radial basis function neural network. The tracking errors are proved to converge to a reasonable scope via the Lyapunov stability theory. Finally, a numerical experiment is used to verify the effectiveness of the designed flight controller.