In 2010 Volvo cars introduced advanced driver assistance systems (ADAS) designed to detect vulnerable road users (VRUs) in specific conflict situations. The aim of this study was to evaluate the first generation of the optionally mounted Pedestrian ADAS, which covers car-to-pedestrian collisions, and Cyclist ADAS, which covers car-to- cyclist collisions.

Data from collisions in Sweden between passenger cars and pedestrians or cyclists were collected from 2015-2017. Crashes involving Volvo cars with third-party liability insurance at If P&C Insurance/Volvia were included in the dataset, and cars with these ADAS were compared to crashes involving cars without the systems. A total exposure of 490,000 insured vehicle years was used in the evaluation.

Overall, the number of collisions for cars with the Pedestrian ADAS system was 21% less than the number for cars without it. When studying straight crossing path crashes only, which accounted for more than half of all car-to- pedestrian collisions in Sweden, these were reduced by 36%. However, the results are not statistically significant due to the low number of crashes. For the ADAS, which covers car-to-cyclist collisions, an overview of data available for retrospective performance evaluation is discussed.

One clear restriction in the evaluation of VRU ADAS at this point in time is the relatively low number of cars equipped with the system together with the low rates of car-to-cyclist collisions (≈ 0.0002 per insured vehicle year) and car-to-pedestrian collisions (≈ 0.0001 per insured vehicle year).

This study is the first real-world evaluation of the initial generation of VRU ADAS targeting traffic situations relevant for these technologies. ADAS for avoiding collisions with pedestrians and cyclists have a high traffic-safety potential;​ recent and future generations of these systems, cover more conflict situations and are thus expected to provide increasing safety benefits.