Improving Dependability of Vision-Based Advanced Driver Assistance Systems Using Navigation Data and Checkpoint Recognition

Advanced Driver Assistance Systems (ADAS), like adaptive cruise control, collision avoidance systems, and, ultimately, piloted and autonomous driving are increasingly evolving into safety-critical systems. These ADAS to a large degree rely on proper function of in-vehicle Computer-Vision Systems (CVS), which is hard to assess in a timely manner, due to their high sensitivity to the variety of illumination conditions (e.g. different sun positions, weather conditions, light reflections and glares, artificial light). On the other hand a diverse set of selfawareness information is commonly available in the vehicle, such as maps and localization data (e.g. GPS). This paper, therefore, studies how the combination of diverse environmental information can contribute to improving the overall visionbased ADAS reliability. To this extent we present a novel concept of a Computer-Vision Monitor (CVM) that regularly identifies checkpoints (predefined landmarks) in the vehicles surrounding, base...

• Type: Journal
• Year: 2015
• Where: SAFECOMP
• Authors: Ayhan Mehmed, Sasikumar Punnekkat, Wilfried Steiner, Giacomo Spampinato, Martin Lettner
• Keywords: SAFECOMP 2015, Security Privacy