MPC-based Visual Servo Control for UAVs

Abstract: Vision information is essential for planning and control of autonomous systems.Vision-based control systems leverage rich visual input for motion planningand manipulation tasks. This thesis studies the problem of Image-BasedVisual Servo (IBVS) control for quadrotor UAVs. Despite the effectiveness ofvision-based systems, the control of quadrotors with IBVS presents the nontrivialchallenge of matching the 6 DoF control output obtained by the IBVSwith the 4DoF of the quadrotor. The novelty of this work lies in addressing theunder-actuation problem of quadrotors using linear Model Predictive Control(MPC). MPC is a well-known optimization control technique that leverages amodel of the system to predict its future behaviour as a function of the inputsignal. We extensively evaluate the performance of the designed solution onboth simulated environment and real-world experiments.