How Autonomous Mapping Can Help a Road Lane Detection System ?

In this paper we first present an autonomous mapping algorithm based on a road lane detector system. Then we focus on the benefits one can except by using the generated map to help the detection of the road lane in the context of advance driving assistance systems (ADAS). In our study we exploit a single vision-based road lane detection system. Two modes are investigated: in the ´map generation´ mode, the lane detection system is used to generate a map under good weather, lighting, and traffic conditions; in the ´map use´ mode, it is used as a driving assistance system and can rely on the previously generated map to improve the results of detection. The lane detection system is used along with a low cost GPS in both modes. Experimental results show that using the map can improve the detection of the road lane under adverse conditions (occlusions generated by traffic, degraded lane markings, bad weather like fog). Moreover the use of the map can help avoiding false detections and increasing the detection range of the system