Positioning Technology Based On Machine Vision Uav Down

Based on the background of the "Ten-Five" Technology Project, We present the design and implementation of a real-time computer vision system for a rotor-craft unmanned aerial vehicle to land onto a known landing target.This vision system consists of customized software and off-the-shelf hardware that perform image processing, segmentation, feature point extraction, target recognition, camera pan/tilt control, and motion estimation. We introduce the design of two landing targets that significantly simplify the computer vision tasks such as corner detection and correspondence matching. Customized algorithms are developed to allow for real-time computation at a frame rate of 3Hz. Such algorithms include certain linear and nonlinear optimization schemes for model-based camera pose estimation. We present results from an flight test of simulation that show the vision-based state estimates are accurate to within 1.5cm in each axis of translation and 2.5 degrees in each axis of rotation, making vision a viable sensor to be placed in the control loop of a hierarchical flight management system.In brief, this paper put forward a set of project that aid the Unmanned Aerial Vehicle to land onto a known target and that is low-cost and is easy to realize. The flight test of simulation shows that the project's precision is higher and the real-time is better. It provide the theories foundation and solutions for research and manufacture of actual aid-landing system of UAV. The project offers the important referenced value to develop the UAV.