Pedestrian accident simulation and protection technology evaluation

Abstract

Pedestrian safety is an important societal issue and as one of the stakeholders, vehicle manufacturers are attempting to improve pedestrian protection by enhancing vehicle design. In order to enhance vehicle design, first it is necessary to gain an improved understanding of the interactions between a pedestrian and a vehicle during an accident. Secondly, this knowledge needs to be transformed into vehicle design and technology changes. This thesis focuses on the construction of new models and methodologies to provide an improved understanding and the application of this understanding to design, develop and evaluate a number of pedestrian protection technologies. A review of the pedestrian safety issue and different approaches to pedestrian protection research provide the background to the chosen approach. This is described in terms of an overall methodology for any pedestrian protection technology that also provides a framework for this research. The construction and evaluation of pedestrian accident simulations with a reference C class vehicle are described in detail. The influence of accident conditions and the expected ranges of various quantitative pedestrian injury and motion measures are identified. Vehicle impact velocity, pedestrian size and stance have significant influences on these measures. Therefore it is not possible to state, for instance, that under all accident conditions, one vehicle impact location is likely to cause lower injury measures than another is. There is a clear increase in pedestrian measures (e.g. head velocity, HIC, tibia acceleration, knee bending) with a large increase in impact velocity (i.e. 25 to 40 km/h). However, some measures (e.g. HIC) do not necessarily increase with a small increase in impact velocity (e.g. 25 to 30 km/h) because of the new pedestrian motion (e.g. a new head impact location). Large differences exist between the 6 year old pedestrian and adult pedestrian model measures (e.g. larger post head impact motion but smaller HIC and tibia acceleration) and pedestrian stance has a complex influence on all measures with few overall trends. Pedestrian protection headlamp, bumper system and hood system concepts are developed in biomechanical, analytical and numerical component models. These concepts are used to construct and subsequently benchmark, with pedestrian accident simulations, two modified vehicle models that incorporate different combinations of the technologies. Both the absolute measures and ranges of the measures from the reference vehicle simulations are compared. There are large differences between the pedestrian measures from the reference and modified vehicles but much smaller differences between the modified vehicles. Impacts with the modified vehicles cause the largest differences in pedestrian motion at 40 km/h, for the 6 year old pedestrian, in stance TV, in the early (up to 20 ms) and late (after 140 ms) stages of the accident simulations. Although the modified vehicles reduce pedestrian injury measures for some of the accident conditions, neither of them reduce all measures for all of the conditions. However, significant improvements in experimental sub system measures [EEVC 1998] are achieved with a prototype modified vehicle that incorporates some of the technologies. Benchmarking is hindered by complex injury measure trends and by pedestrian and vehicle model limitations. Recommendations are made with respect to all of these factors. Further recommendations include the need for optimisation of the modified vehicle technologies in accident simulations, a more complete investigation of other technology functional requirements (e.g. low speed damageability) and accident reconstruction as a means to achieve improved model validation.