Autonomous mini-UAV for indoor flight with embedded on-board vision processing as navigation system

Abstract—This paper describes the development of an Un-manned Aerial Vehicle (UAV) with the aid of a vision processing system for indoor navigation. A co-axial radio-controlled (RC) helicopter is upgraded with a customized on-board avionics system which include two Gumstix Linux computer systems as the on-board processors. A camera module is used to capture real time video during the flight to perform real time image processing. High level control commands are sent to the avionics system such that the UAV could perform simple indoor navigation by tracking the colored tracks on the ground. Important data such as attitude, velocity and acceleration of the UAV, together with the real time video will be feedback to the ground station via communication links for giving commands and monitoring purposes. Flight tests have been carried out to verify the results of the vision processing system and to ensure the robustness of the controller in the UAV. I