The Research On Vision System In Aerial Robots

Aerial robots, which is widely applied both in military and civilian areas, has become one of the most popular research subjects in many institutions and organizations for a long time all over the world. Vision system plays one of the most important parts in the aerial robots for locating and tracking mission.The varied background and the moving of the camera make the object detection difficult. To solve this problem, a vision system of aerial robots is presented based on machine learning in the thesis. Firstly the history, sorts and actuality of the aerial robots are introduced. Then in chapter 2 we can know some basic knowledge of image process and pattern recognition. The structure of this vision system is shown in chapter 3. In the following, the object detection algorithm based on PCA and SVM is emphasized in chapter 4. In the chapter 6, the experimental results show the system effectiveness. The future work of the vision system for Aerial robots is discussed and prospected in the final part of this thesis.