Requirements on Delay of VRU Context Detection for Cooperative Collision Avoidance
Michel Morold, Quang-Huy Nguyen, Marek Bachmann, Klaus David, Falko Dressler · 2020
Every year, approximately 350 000 vulnerable road users (VRUs) still lose their lives in road traffic accidents world-wide. To reduce this number, cooperative VRU collision avoidance systems aim at complementing current vehicle or infrastructure-based systems for VRU protection. The cooperative approach assumes user equipment (UE) like a smartphone or a smartwatch on the VRU side, which allows to obtain and exchange movement and contextual information with other road users for determining the current risk level for a collision. In recent publications, the usage of additional contextual information, such as the pedestrian's current activity (e.g., standing, walking or crossing a curb), has been shown to improve the collision detection accuracy. However, those approaches focused on the detection accuracy, but did not investigate how time delays for both detection and communication of VRU contexts affect the ability to detect collisions. We fill this gap by investigating the influence of both activity recognition and communication delays on the collision detection performance. As a baseline, we use the standardized Euro NCAP test protocol. For the evaluation, we exemplarily consider the usage of a curb detection module for position correction, which improves the collision detection accuracy at the moment the pedestrian crosses the curb.