This presentation addresses key challenges in driver assistance systems, with a focus on self-parking mechanisms and safety. We explore the limitations and challenges of current sensors, particularly in accurately detecting small objects such as road debris and hazards in different road conditions. Our analysis employs a comparative approach, assessing technologies like LiDAR and stereo vision in self-parking scenarios, including operations in nighttime and adverse weather conditions.
The presentation will specifically highlight the precision with which these technologies facilitate navigation in tight spaces, especially at night. Furthermore, we will present experimental results that showcase stereo vision’s capability to detect small objects at night, like tires at distances of 80 meters without headlights – cameras must be able to detect obstacles in low light conditions to be considered for self-parking applications