The aim of this report is to implement a Multi-Input Multi-Output controller to a plant model. The plant model used is a Quadrotor, the parameters of which has been derived from DJI Phantom 2. The goal of the quadrotor is to align with the reference trajectory input by the user. A non-linear model predictive control has been implemented on the quadrotor using MATLAB Model Predictive Control Toolbox. Wind disturbance is introduced and evaluated in various cases. Parameter tuning is done using a grid search method. The paper also evaluated the effectiveness of MPC as a controller on a Quadrotor
Open file NLMPC_demo.m and run it using MATLAB
Change Vx and Vy for different use cases