Research On Vision Navigation System For Intelligent Vehicles On Highway

Intelligent vehicle (IV), which is on the top level in the research field of vehicle engineering, is the hotspot around the world nowadays. Intelligent vehicle is a comprehensive intelligent system which integrates the function of environment perception, planning and decision-making, driving assistance, etc. Intelligent vehicle, which represents the direction toward the development of future vehicle, is also a product that contains multi-technical subjects including computer vision, artificial intelligence, auto-control theory, and electronic technology. Therefore, intelligent vehicle has a highly expansive application prospect.As vision-based intelligent vehicle gradually is developed to be the mainstream of various intelligent vehicles,and the research on key techniques for intelligent vehicle navigation system does great contribution to safe driving and even.Because the vision guidance system can get wide and integrate vision, costs fewer, and can fit the perceiving way of the human, this paper applies monocular vision method into the vehicle assistant driving and the automatic guidance. At present, although there are many vision guidance systems, none of them can meet our requirement in real-time, robustness, and practice. The main contents of my dissertation are as follows:Firstly,the destination of this part is to analyze the mechanism of the vision system performance, obtain the imaging characteristics of the vision system in specific environment and configure the relevant parameters rationally, which have practical significance, and provide reasonable preconditions for further investigation.Secondly,this paper analyzes the structure of the vehicle assistant driving and the automatic guidance system, and the emphasis is road detection and tracking. Compared with reappearing three-dimensional scene by stereo vision, it demonstrates the merit and practical significance of the monocular vision method.Lastly, In the lane detecting, we propose a method which fuses the way of lane edges detection and the method of color-based segmentation. The method reduces the shadow and noise points influence on lane detection and is convenient in the applications in the vision navigation of the Automatic Highway Vehicle.