Black-box Identification and Iterative Learning Control for Quadcopter

Abstract

This paper describes the black-box system identification and iterative learning control algorithm which apply to a quadcopter. At first, two feedback control loops from angles and angular velocities are implemented to stabilize the system, then the experimental data and the models of roll and pitch are identified via a black-box routine. For more accuracy of angles measurement, the complementary filter is used to fuse the data of the inertial measurement unit (IMU). An iterative learning control (ILC) method is then applied to the closed loop system with the aim of following trajectories of the roll and the pitch angles. The designed controller was first applied to the roll and pitch model separately and then applied to both angles simultaneously. The experimental result is presented and discussed.