CONTACT US: Contact info
|
Citation |
Krampe J, Junge M. Accid. Anal. Prev. 2020; 138: e105321. |
|
Affiliation |
Volkswagen AG, Wolfsburg, Germany. Electronic address: mirko.junge@volkswagen.de. |
|
Copyright |
(Copyright © 2020, Elsevier Publishing) |
|
DOI |
|
PMID |
32135304 |
|
Abstract |
Our goal was the development of a robust and data-driven approach to ISO 26262 injury severity (S-parameter) estimation, replacing the current heuristic methods. The situations investigated as part of an ISO 26262 hazard & risk analysis are broken down into crash configurations. These crashes are analyzed from the perspective of the ADAS-equipped vehicle with the failing system as well as from the crash opponent's point of view. Mainly due to sample size limitations, we focus on belted front-row vehicle occupants. We cluster the crash data into traffic domains (TD) based on the speed limit, i.e., residential streets, city roads, arterial thoroughfares, rural roads, and intercity highways and calculate the crash speed distribution for each domain. The injury severity clustering is based on the ISS injury aggregator with cut-offs at 4, 9, and 16 for S1, S2, and S3, respectively. We estimated the 90th-percentiles of the S-parameter cut-offs with a 95% confidence level using the GIDAS accident database. The percentiles were calculated for the ADAS-equipped vehicle as well as for the crash opponent, stratified for crash type (front, oblique, side). The stratification had to be detailed further for side crashes as impact direction (near-side vs. far-side) and availability of a curtain airbag restraint system have a significant impact on injury severity. The application of the results towards the assessment of a crossing scenario is detailed in the discussion. Language: en |
|
Keywords |
ASIL; ISO 26262; Injury severity; S-parameter; crash configuration; injury risk |