Neural sliding mode control of low-altitude flying UAV considering wave effect

This paper investigates the adaptive sliding mode control (SMC) for fixed-wing UAV using neural networks (NNs). Considering the wave effect during low-altitude flying, the wave height is identified based on autoregressive model while the model parameters are handled by recursive least square method. Considering the aerodynamic uncertainties caused by unknown sea environment, NNs are employed to deal with the system nonlinearities. For the update of neural weights, the prediction error that indicates the learning performance is constructed. Based on the information of neural approximation and wave height identification, the neural learning control is finally developed for the altitude subsystem. The neural SMC is accordingly constructed for the velocity subsystem. Under the proposed method, the uniformly ultimately bounded stability is achieved. Simulation is presented to show that the proposed method can achieve good tracking performance.