A Study On FPGA-Based Anti-collison And Obstacle Avoidance Technology

With the rapid development of Chinese economy, family cars become popular and available for ordinary families. However, the accident rate increases gradually due to the increasing number of vehicles in cities, which leads to the enormously potential danger to people lives and property security. Therefore, it becomes more and more important to research the anti-collision and obstacle avoidance technology nowadays. In order to provide smarter traffic information to car drivers, this paper conducts a research on anti-collision and obstacle avoidance technology with the assistance of camera and radar.Firstly, this thesis summarizes the current development on the field of anti-collision and obstacle avoidance technology mastered by auto-manufacturers both domestic and abroad, and compares the different technical solutions in terms of anti-collision and obstacle avoidance technology between the main auto-manufactures. Along with the technical development, nowadays the function of FPGA becomes stronger, which powerful operation ability and big data storage capability highlights the advantage in favor of video picture processing. The thesis mainly focuses on the research of anti-collision and obstacle avoidance system based on FPGA video picture processing and traditional millimeter wave radar. This system uses camera module and millimeter wave radar as data input module and EP4CE6E22C8 N as data processor, combines the picture processing technology with radar ranging, so as to judge the preceding vehicle's state in advance by designing the whole system including safe distance module, image acquisition and storage, region matching and trigger condition of warning information. After accomplishing the realization of image acquisition, Sobel algorithm for edge detection and mathematical morphology are adopted in order to detect and identify the target vehicle. According to the vehicle's braking system and driver's habit, the vehicle collision warning system enables to provide reliable reference information to the diver in a more intelligent way through the design of fuzzy rule database and trigger condition.In this graduation design, the critical security model is built and emulated by MATLAB. And the image processing is realized through experimental board and displayed through VGA.