Fuzzy Model Predictive Control of a Quadrotor Unmanned Aerial Vehicle

This paper proposes a new design of a Fuzzy Model Predictive Control (FMPC) algorithm for controlling a quadrotor unmanned aerial vehicle. The quadrotor's nonlinear model is represented by a Takagi-Sugeno (T-S) fuzzy model, where the nonlinear model is approximated by multiple linear models, and the fuzzy combination of these linear models constructs the overall fuzzy model of the nonlinear system. The performance of the proposed FMPC algorithm is compared with Linear Model Predictive Control (LMPC), and Nonlinear Model Predictive Control (NMPC) applied on the nonlinear model of the quadrotor system. Simulation results show that FMPC yields better performance and a wider flight range than LMPC. FMPC also proved to have almost similar results to NMPC in trajectory tracking but with less computational time. Finally, the control invariant sets are calculated and the results are analyzed.