Adaptive neural network-based robust <i>H</i><SUB align="right">∞ tracking control of a quadrotor UAV under wind disturbances

Abstract

The paper deals with the stabilisation and trajectory tracking control of an autonomous quadrotor helicopter system in the presence of wind disturbances. The proposed adaptive tracking controller uses radial basis function neural networks (RBF NNs) to approximate unknown nonlinear functions in the system. Two controllers are proposed in this paper to handle the modelling errors and external disturbances: H∞ adaptive neural controller H∞-ANC and H∞-based adaptive neural sliding mode controller H∞-ANSMC. The design approach combines the robustness of sliding mode control (SMC) with the ability of H∞ to deal with parameter uncertainties and bounded disturbances. Furthermore, the RBF models are derived using Lyapunov stability analysis. The simulation results show that H∞-ANSMC is able to eliminate the chattering phenomenon, reject perturbation mismatch and leads to a better performance than H∞-ANC. A comparative simulation study between the proposed controllers is presented and the results are discussed.

Research on key technologies of fire fighting UAV

“…The experimental test results show that the algorithm has fully improved the path tracking ability of UAV and has stable and accurate performance when it follows 3D curve flight. Zakaria Bellahcene et al [3] proposed to use the adaptive neural network to calculate the nonlinear function in the UAV system in the study of UAV stability and trajectory tracking control. The final experimental results show that the performance and stability of UAV in the case of wind disturbance are significantly improved.…”

A review of quadrotor control methods

“…The authors in [20] structure ,this method is qualified to deal with the influences of perturbations and measurement errors. In the [21], the authors have been presented adaptive neural network based H∞ control to approximate nonlinear systems, and validate the controller is able to eliminate perturbance.…”

Time-varying formation dynamics modeling and constrained trajectory optimization of multi-quadrotor UAVs

“…A QUAV is a kind of aircraft with four rotors, having two types of X-type distribution and crosstype distribution [21,22]. In this paper, the X-type is used, and the schematic diagram of the QUAV is shown in Fig.…”