Poster 3: Reference Machine Vision for ADAS
(Room Palms Ballroom)
16 Jul 19
5:30 PM
-
7:00 PM
Tracks:
Vehicle Automation Technology
Lane Departure Warning (LDW) and Lane Keep Assist (LKA) systems have the potential to prevent or mitigate 483,000 crashes in the United States every year [1]. This includes 87,000 nonfatal injury crashes and 10,345 fatal crashes every year in the United States. While LDW and LKA technologies are available, there has been low customer acceptance and penetration of these technologies. These deficiencies can be traced to the inability [1,2] of many of the perception systems to consistently recognize lane markings and localize the vehicle with respect to the lane marking in a real-world environment of variable markings, changing weather conditions and occlusion by other vehicles. These challenges translate to (i) inconsistent detection of lane markings; (ii) misidentification of lane markings; and (iii) the inability of the systems to locate lane markings in some conditions. These challenges can be addressed both by improving the consistency and detectability of the lane markings and by improving the perception algorithms currently employed in the sensors. Extensive research has been carried out on lane estimation methods. However, there is no available standard or benchmark to evaluate the quality of either the lane markings or the perception algorithms [2] encompassing all different weather and road conditions. The objective of our project is to develop a reference Lane Detection system that will provide a benchmark for evaluating the effectiveness of different lane markings and perception algorithms to reliably engage LDW and LKA systems.