Navigation Vector Map Design And Implementation

Vector map is the main format in modern auto navigation. In auto navigation, road shape and grade, speed limit, landmark, buildings, and the environment around us shall be displayed in digital map. How to organize and store all the above mentioned map data is an important research topic.Our research extracts the characteristics of map elements by way of abstract generalization, including the common features and the relationship between them. Therefore the data structure of CHUNK is designed. We planned a set of coded procedure and resolution protocol, which gives the possibility for coded data to have augmentability, compatibility and reduce redundancy. And it realized the possibility of across platforms of the code. We used the method of semi precision storage in the smallest unit of the stroing CHUNK, such as position and coordinates. We can reduce50%data storage space by using semi-precision method.For polygon objects, we use the method of separating simple and complex polygon to improve resolution efficiency. Simple polygon, starting from one point clockwise, and connecting with two consecutive points to form a triangle, we can complete the triangle segmentation to the polygon. But this method is not applicable to the complex polygon, to which we shall use more complicated calculation to segment the triangle (strip, fan-shaped or simple triangle). There are around80%polygons are simple ones in the map elements, which increase the rendering speed.We use photographs to store the index of map data. Several improvements are made to the calculation method during the process of organizing the index, by comparing the result with calculation method of boost, the former one is obviously better than the latter one. By using road segmentation to solving the covering issue, we optimized the calculation and got positive effect.The above mentioned several aspects consist of the main part of the thesis. The fundamental is chunk. We have different structures of the chunk, code resolution protocol and relation between the chunks, and finally generate the vector map data and represent the rendering result. Comparing with other way (raster map), our method has the characteristics of high data generating speed, smaller storage space and high rendering speed.