Stefan Schroedl

Mining GPS Traces for Map Refinement

Despite the increasing popularity of route guidance systems, current digital maps are still inadequate for many advanced applications in automotive safety and convenience. Among the drawbacks are the insufficient accuracy of road geometry and the lack of fine-grained information, such as lane positions and intersection structure. In this paper, we present an approach to induce high-precision maps from traces of vehicles equipped with differential GPS receivers. Since the cost of these systems is rapidly decreasing and wireless technology is advancing to provide the communication infrastructure, we expect that in the next few years large amounts of car data will be available inexpensively. Our approach consists of successive processing steps: individual vehicle trajectories are divided into road segments and intersections; a road centerline is derived for each segment; lane positions are determined by clustering the perpendicular offsets from it; and the transitions of traces between segments are utilized in the generation of intersection models. This paper describes an approach to this complex data-mining task in a contiguous manner. Among the new contributions are a spatial clustering algorithm for inferring the connectivity structure, more powerful lane finding algorithms that are able to handle lane splits and merges, and an approach to inferring detailed intersection models.

Lane keeping based on location technology

Vehicle positioning with an accuracy of 10 cm or less will enable lane-keeping assistance in addition to other safety benefits when an enhanced lane-level digital map is in place. With constantly evolving technology and sensors, a high-precision positioning system that fits into the automotive market can be expected within the next decade. Such a system will incorporate Global Positioning System (GPS) and inertial system (INS) for enhanced positioning performance and availability. In this paper, the technology fields that will have a significant impact on the deployment of a centimeter-level vehicle-positioning system will be discussed. Vision-based lane-recognition (VBLR) systems are relatively mature and have already been introduced to the market for lane-departure warning, etc. However, both systems have some limitations. GPS/INS-based systems may suffer from frequent satellite signal masking or blockage, while vision-based systems do not work well in adverse weather conditions or with poor lane signature. Effectively combining these two technologies can make a robust lane-departure warning system. A precision map was made for the test area near Stuttgart using DaimlerChrysler Research and Technology North America (RTNA)s map-making approach. A Mercedes S-class equipped with both a vision system and a high-precision GPS/INS was used for the test. The positioning map-matching results and the vision offset are compared and the complementary effectiveness is illustrated.