A Real-time Integrated Navigation System for UAV Autonomous Landing Using a Single Airborne Camera
We introduce the design of an integrated navigation system for autonomous landing of unmanned aerial vehicle (UAV) based on computer vision. Our task is emphasized in improving the practicability, robustness, real-time and accuracy of integrated system: We use an adaptive threshold selection method to make our algorithm more robust; An effective object recognition algorithm based on template matching is implemented to improve the recognition rate; We propose an improved image segmentation method which optimize execution efficiency of the algorithm; The experiment results prove that our algorithm is fast (average processing time below 60ms), accurate (recognition rate over 94.7％, the angle error below 4° , the position error below 5cm), and robust (can distinguish several suspected targets). In addition, based on this system we won the champion of the aerial robotics group for 2009 Chinese Robotics Competition.
unmanned aerial vehicle autonomous landing vision-based navigation object recognition template matching
Yang Chengxu Wang Hong
State key laboratory of Intelligent Technology and Systems Tsinghua National laboratory for Information Science and Technology Department of Computer Science and Technology, Tsinghua University, Beijing 100084, China
国际会议
重庆
英文
748-752
2011-08-20（万方平台首次上网日期，不代表论文的发表时间）