Child occupant safety in unconventional seating for vehicles with automated driving systems

Abstract

The objective of this study was to use computational models to study how unconventional seating positions and orientations in vehicles with Automated Driving System (ADS) may affect occupant response metrics of children with various restraint conditions. A literature review was first conducted to frame a simulation plan, including selections of surrogate ADS-equipped vehicles, potential seating arrangements, impact scenarios, anthropomorphic test device (ATD) models, and child restraint system (CRS) models that are relevant to the selected ATD models. Due to the lack of impact tests with child ATDs and CRS in farside, oblique, and rear impacts, 17 sled tests were conducted with CRS harness-restrained ATDs and vehicle belt-restrained ATDs in frontal, farside, oblique, and rear impact conditions. The sled tests were then used to validate a set of MADYMO (MAthematical DYnamic MOdels) v7.7 models. A total of 550 simulations were then conducted with four child ATDs and various CRS conditions across a range of conventional and unconventional seating locations and orientations under five impact directions. We did not find major safety concerns with ATDs restrained by harnessed CRSs based on the nature of ATD contacts. Compared with frontal and rear impacts, CRSs may rotate laterally in farside and oblique impacts, which could result in higher head and chest injury measures than frontal due to inertial loading to the CRS, and the larger lateral rotation of the CRS may lead to a contact between the CRS and vehicle interior. The major safety concern for vehicle belt-restrained ATDs (with and without booster) is that they have the potential to contact the seat next to them or the instrument panel behind them in a farside or oblique impact. Unconventional seating does not necessarily create additional safety concerns beyond what we know with the conventional seating. However, due to the orientation of the unconventional seats, the occupants on those seats may be involved in a higher percentage of oblique and rear-oblique impacts relative to their seating orientations than conventional seats, which may be considered in the future safety design process. This is the first study using different child ATDs and CRSs to investigate child occupant responses in a wide range of impact directions and seating orientations. Results from the sled tests and simulations provide a better understanding of child occupant responses in those crash conditions, but also identified several limitations of using frontal ATDs in other crash directions.

Keywords: Automated vehicles; Child passenger safety; Child restraint systems; Computational modeling; Sled tests; Unconventional seating.