Development and applications of a vision-based unmanned helicopter

Abstract

The vision-based unmanned aerial vehicles (UAVs) have aroused great interest recently. This calls for the in-depth research work on the vision-based UAV, and its applications, which are presented in this thesis. A systematic and effective design methodology has been presented to construct the UAV with minimum complexity and time cost. An efficient software system for the UAV has been developed. Next, an application of the vision-based ground target following is presented. An advanced vision algorithm is proposed to detect the target, and integrated with the flight control system to guide the UAV to follow the target in-flight. Moreover, another application is presented on the vision-based motion estimation of the UAV in GPS-denied conditions. A vision augmented approach is proposed to realize the vision-based motion estimation and flight control based on the Kalman filtering. Both applications have been tested in the flight to demonstrate the capabilities of the proposed vision-based UAV.