Vehicle shape influence in car crash simulations

The progressive spread of active safety systems in the road vehicle fleet requires precise and accurate tools for assessing the performance of Advanced Driver Assistance Systems (ADAS) aimed at collision mitigation. These systems can be evaluated with both a retrospective and a prospective approach: virtual forward simulation (VFS) plays a major role in prospective evaluations. Sound software routines for VFS need valid models to simulate driving dynamics and the behaviour of active safety systems, just as they need models capable of correctly solving collision mechanics to deal with cases where ADAS intervention is unable to avoid the collision. The reliability of the output parameters of VFS, such as post-crash velocities, delta-v, force directions, etc., is a function of the employed driving dynamics and collision models; also, it depends on the accuracy of the input parameters as the shapes of the involved vehicles. Specifically, the possibility to use simplified shapes enables technicians to simulate all possible impacts, independently of the availability of the actual vehicle shape that is often limited in software packages. This study investigates the influence of vehicle shape on the quality of 2D impact simulations between passenger cars. Several impact configurations are simulated using a validated VFS tool based on a reduced order impact model. The degree of detail of the vehicle shape is provided as input to each simulated configuration and varied from a simple rectangle to the real profile of the car. The influence of vehicle shape on the simulation outputs is analysed by comparing them with reference values. In addition to the main outputs such as delta-v, characteristic parameters of the impact phase such as the orientation of the impact plane and the position of the point of impact are also assessed. Trends are also obtained regarding the single variables to generalize the results as much as possible. The outcomes of the study represent a relevant basis to determine the contexts where approximations of the vehicle shapes can still provide sound information in VFS applications.