Fuzzy SMC for UAV to Resist Strong Wind with Disturbance Approximation

Jiaqi Lv,Junhui Shen,Hongwei Xia,Guangsheng Ma,Changli Yu,Youtao Shen,Chengyu Zhou
DOI: https://doi.org/10.1109/icicip.2018.8606715
2018-01-01
Abstract:In this paper, a quadrotor UAV dynamic model with the wind speed factor is established and sliding mode controllers are designed for attitude stabilization. Neural network and fuzzy control are proposed to improve the performance of anti-strong wind controller for unmanned aerial vehicle (UAV). Uncertainty of dynamic model can be approximated by radical basis function (RBF) neural network, and fuzzy control can dynamically correct the coefficients of the symbol function during the state convergence process. The simulation results show that the neural network constructed in this paper can track the uncertainty of the model, combining sliding mode with fuzzy control maintains the advantages of fast response and robustness. At the same time, the chattering problem is greatly reduced, which guarantees the anti-strong wind requirements of UAV and improves the stability of the controllers.
What problem does this paper attempt to address?