Analysis of the Field Effectiveness of General Motors Model Year 2019-2023 Advanced Driver Assistance System (ADAS) Features

Abstract

This police-report based study examined the extent to which General Motors advanced driver assistance (ADAS) features are reducing system-relevant crashes, both overall and for crashes with reported injuries. A total of 565,936 crashing Model Year 2019–2023 vehicles were identified by matching VIN-linked content data to police-reported crashes from 15 states. The quasi-induced exposure method was used to compare system-relevant and system-irrelevant (control) crash counts for equipped and unequipped vehicles. Logistic regression was used to adjust for 13 covariates. Results indicated that fusion Automatic Emergency Braking, camera Automatic Emergency Braking, and camera Forward Collision Alert features reduced rear-end striking crashes by 54%, 39%, and 17%, respectively. Corresponding 52%, 47%, and 28% effectiveness values were observed in the injury-focused analysis, providing evidence of additional crash mitigation benefits. For Front Pedestrian Braking, a 25% reduction in front pedestrian injury crashes was observed. The Lane Keep Assist with Lane Departure Warning feature provided 9% and 6% reductions, respectively, in roadway departure crashes and same-direction sideswipe crashes. The Lane Change Alert with Side Blind Zone Alert system reduced lane change crashes by 11%. Reverse Automatic Braking, Rear Cross Traffic Alert, and Rear Park Assist (where each of these systems generally included all of the preceding systems) produced, respectively, an 84%, 57%, and 51% reduction in backing crashes. These results continue to provide widespread evidence of the substantial crash avoidance and injury reduction (crash mitigation) opportunities afforded by ADAS features and support identifying opportunities for accelerating progress toward a zero crashes vision.