• Leo, Christoph

Abstract

n 2021, pedestrians accounted for 19% of all road deaths in Europe, which means a total of 3,669 fatalities in one year. As part of the EU's Vision Zero plan, new safety measures are needed to reduce the number of road deaths as close to zero as possible by 2050. It is expected that the introduction and market penetration of new active safety systems, such as the autonomous emergency braking system (AEB), will have a significant impact on reducing the number of road accidents. Therefore, a holistic assessment of active and passive safety measures for pedestrian protection in case of accidents involving passenger cars, is essential.<br/><br/>To develop such a holistic assessment method, it is important to identify what needs to be assessed to achieve real-world benefit. Therefore, accident databases were analysed to identify the most common types of injuries and differences between different groups of the population, to ensure that they are included in the assessment procedure. These issues were analysed with the help of real-world accident data. Next, the effect of a generic AEB systems on head impact conditions was investigated. This research demonstrates the need for a holistic safety assessment as the impact conditions were highly affected by the changes in the collision scenarios caused by the AEB. Hence, an integrated pedestrian safety assessment framework was developed as the next step. The method developed enables consideration of both active and passive safety measures, as well as distributions of real-world crash scenario parameters. The probability of a specific virtual test scenario occurring in real life was derived from accident databases scaled to the European level. Pre-crash simulations were used to assess the likelihood that a specific AEB system would be able to avoid the virtual testing scenario. Probabilities of specific collision scenarios were determined for unavoidable virtual test scenarios. The injury risk of these collision scenarios was evaluated through in-crash simulations using the VIVA+ Human Body Models, for both the average male and female. To be efficient in terms of computational efforts for the in-crash simulation, a metamodel has been trained with the HBM simulations to predict injury risks for the whole catalogue of remaining collision scenarios. The application of the integrated assessment framework was demonstrated using a generic vehicle model.<br/><br/>The accident data indicate that severe injuries (AIS3+) in case of pedestrian passenger car collisions were mainly to the head, thorax and lower extremities for both women and men. The accident database and results with the generic vehicle model indicate that females have significantly higher odds of sustaining skeletal injuries to the lower extremities (including the pelvis). A total reduction in crash risk of 81.70% was achieved, based on 61,914 virtual testing scenarios, by implementing a generic AEB. The analysis of the injury risks shows a drastic reduction in injury risk for all analysed injury types for both the average female and male, due to the AEB. It was found that the risk of brain injuries and femoral shaft fractures is equal for both males and females. However, males have a higher risk of skull fractures and fractures involving more than three ribs, while females have a higher risk of proximal femoral fractures, which can also be seen in the accident data.<br/><br/>To facilitate bridging the gap between assessment and real-world, a new methodology has been developed to move away from relying solely on standard load case assessments for pedestrian protection of passenger cars.

Topics

• impedance spectroscopy
• simulation