Vulnerable Road Users Safety in Infrastructure Assisted Intelligent Transportation System

About 20-50 million people are involved in road accidents every year caused by human errors, with more than 500,000 of the victims being low-mobility, high-physical-impact-prone users of the road such as pedestrians, cyclists, motorbikes, scooter, and the likes. Such road users are called vulnerable road users (VRUs). Taking VRU awareness message (VAM) for assessment of the potential collision risk, we argue that the state-of-the-art metric, time-to-collision, is insufficient for accurate collision risk analysis and collision avoidance. A new metric, minimum safe distance, is proposed for realistic collision risk analysis based on VAM exchange to enhance VRU safety. Results show feasibility of pre-emptive triggering of collision warning at approaching vehicle thus enhancing VRU safety. Furthermore, VRU cluster formation and management schemes are proposed for high density VRUs to achieve at least one order of magnitude reduction in VAM messaging overhead.