Adaptive neural network-based fault-tolerant control for a three degrees of freedom helicopter

Abstract

In this paper, the tracking control problem of a three degrees of freedom (3-DOF) helicopter with unknown actuator faults, model uncertainties and external time-varying disturbances is solved by using the proposed adaptive neural network-based fault-tolerant control (FTC) strategy. Firstly, the mathematical model of the 3-DOF helicopter is established. Then, neural networks (NNs) are utilised to approximate and compensate the unknown functions in the presence of the controlled system. It should be stressed that hyperbolic tangent functions are introduced to reduce the negative effects caused by NN approximation errors and external time-varying disturbances. Moreover, auxiliary functions are designed to compensate filter errors, which further enhance the tracking performance of the closed-loop system. Finally, simulation results are provided to verify the effectiveness of the proposed FTC scheme.