Based On Ultrasonic Distance Measurement And Integration Of Image Information Rotor UAV Obstacle Avoidance Algorithm

Autonomous obstacle avoidance flight has a very important meaning for UAV,which can guarantee the UAV to perform complex, multi-function flight action. At present domestic and international research on UAVs avoidance aspect have had a great breakthrough, typically dependenting on three-dimensional elevation maps, binocular camera or other precision equipment and other technologies, but for low-cost,high-precision, reliable, mobile flexible autonomous obstacle avoidance methods remain to be studied.The greatest degree of access to data is the main source of data for UAV obstacle avoidance,the high-precision rotor UAV autonomous obstacle avoidance is more difficult to achieve to some extent, which is based on the actual need. we urgently need to develop a low-cost, high precision rotor UAV autonomous obstacle avoidance algorithm. This paper uses a CCD image acquisition and Canny operator edge detection to process image, so you can ensure to obtain a clear image contours,using the latter means of non-maximum suppression and hysteresis threshold is processing to filter out some minor factors, making the boundary more clear.The issue of border outline obstacles have higher requirements, in order to direct according to a clear obstacle avoidance algorithm outline of the design. This paper based on image acquisition and processing technology on the environment first obstacle image processing for obstacle avoidance algorithm is designed to provide data sources to achieve a low-cost, high-precision rotor UAV autonomous obstacle avoidance algorithm. The research process is firstly with the ultrasonic ranging obstacle detection distance from the rotor UAV when the distance is smaller than the set distance it will produce a trigger signal to trigger image acquisition sensor acquisition obstacle image;secondly the collected images edge detection algorithm will give a clear obstacle boundary contour; Finally, according to a clear outline execute obstacle avoidance algorithm the UAV can flight autonomously.Finally, to improve the design algorithm based on the analysis of the actual parameters of the sample data, completing a data processing DSP, FPGA image processing platform system design and related hardware equipment selection. This topic will image processing and DSP technology organically combine together, improvingreal-time obstacle model process. Compared to the traditional image of the remote data transmission to the host computer for processing the way, the method greatly reduces the image noise in the transmission process suffered while shortening the time obstacle avoidance process, greatly improving the real-time flight.