This paper describes the work done in the implementation of an onboard 2D map stitching algorithm realized on a fully autonomous quadrotor platform during the International Micro Air Vehicle Competition (IMAV) 2014, held in Delft, Holland. The unmanned aerial vehicle (UAV) platform uses a homography-based model to calculate the image transformations used to achieve onboard 2D map stitching in real-time while the UAV performs a planned flight path over the area of interest. We describe the structure of our hardware and software system as well as the techniques we used to have a fast and efficient system which runs an onboard stitching algorithm while completing autonomous flight. The algorithm described includes a homography failure checking system that could intelligently detect failures in homography and provide an alternative solution that could be used for map stitching.
展开▼