Iterative Learning Control for Quadrotor Pose Tracking

Abstract

This work proposes a dual-layer control strategy for managing the pose of a quadrotor unmanned aerial vehicles in repetitive tasks. The first layer uses iterative learning control to reduce the error between the desired reference trajectory signal and the system output. This layer generates the desired flight trajectories and transmits them to the second control layer. The second control layer uses a dual-feedback proportional derivative strategy to achieve trajectory tracking accuracy. We conducted experiments using a lemniscate trajectory mission to verify the efficiency of the proposed method.