Online situation and risk assessment: Improving cyclists’ safety in intersections?

Abstract

Over the past years, cycling has reached increased popularity in urban areas. However, cycling is also associated with severe injuries or even fatalities when involved in crashes with motorists. In urban areas, one of the riskiest situations is when cyclists cross an intersection while motorists turn right. As a technical solution an infrastructural cooperative intelligent transport system (C-ITS) was implemented. Based on the analysis results of trajectory and video data collected at the AIM Research Intersection in Braunschweig, Germany, in September 2016, an online situation and risk assessment algorithm was developed. This algorithm determines the probability of a collision between right turning motorists and crossing cyclists and translates the results into one of five risk levels. Based on the predicted level of risk, warnings are sent to the installed C-ITS, the so-called Amber Light. If a first degree conflict is predicted, the warning system lights up for 1.5 seconds. If a second degree conflict is calculated, the Amber Light flashes for a second. This technology is evaluated in a field test at the AIM Research Intersection. Over a period of four weeks, trajectory and video data will be collected and log files of the real-time risk assessment generated. In parallel, a questionnaire is provided online. Goal is to gather opinions and experiences of motorists and cyclists. The effects of the warning system on road traffic safety will be analyzed in terms of objectives measures, such as number and severity of conflicts and approach speed, and subjective measures, such as motorists’ trust in and acceptance of the warning system. The implications of the results will be discussed in terms of safety and further improvements.