Andrew J. Leslie , Raymond J. Kiefer , Susan H. Owen , Carol A. Flannagan
{"title":"Examination of General Motors Super Cruise system field effects using state police report crash data","authors":"Andrew J. Leslie , Raymond J. Kiefer , Susan H. Owen , Carol A. Flannagan","doi":"10.1016/j.jsr.2024.11.013","DOIUrl":null,"url":null,"abstract":"<div><div><strong>Introduction:</strong> While SAE Level 2 systems combining Adaptive Cruise Control (ACC) and lane centering functionality can conveniently reduce the need for the driver to frequently brake, accelerate, and steer the vehicle, the potential safety benefit of such systems merit further exploration. <strong>Method:</strong> This study attempted to isolate the field effect of the General Motors (GM) Super Cruise Level 2 system on lane departure and rear-end striking crashes. This system allows hands-free driving on GPS-mapped system-compatible roads and employs a camera-based driver monitoring system and a series of escalating alerts to prompt the driver to pay close attention to the road ahead and take steering control when takeover requests are issued. Police-report data were used to identify system-relevant and corresponding control crashes, and historic telematics-based data were used to understand system use surrounding crashes. A total of 131,757 Model Year 2017–2021 vehicles were matched to police-reported crashes from 12 states, which identified 415 analysis-relevant crashes on Super Cruise-compatible roads (certain limited-access freeways and trunk roads). Multiple statistical approaches were used to address challenges inherent to evaluating Level 2 systems, including small crash samples and system availability/use. <strong>Results:</strong> Results indicated there was no evidence for a difference in system-relevant crash risk for Super Cruise-equipped vehicles compared to matched highly-ADAS equipped vehicles without Super Cruise. The trend toward Super Cruise, as well as other L2 systems, being offered on an increasing number of vehicles with fewer Operational Design Domain (ODD) constraints should in the coming years notably increase the statistical power of the testing approach employed here. These results should not be interpreted as generalizing to other emerging L2 systems, including those that fundamentally differ with respect to Operational Design Domain limitations (e.g., lower speed operation only), requiring hands-on operation for system use, or the exclusive use of a steering-wheel based Driver Monitoring System (DMS).</div></div>","PeriodicalId":48224,"journal":{"name":"Journal of Safety Research","volume":"92 ","pages":"Pages 207-216"},"PeriodicalIF":3.9000,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Safety Research","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0022437524001622","RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ERGONOMICS","Score":null,"Total":0}
引用次数: 0
Abstract
Introduction: While SAE Level 2 systems combining Adaptive Cruise Control (ACC) and lane centering functionality can conveniently reduce the need for the driver to frequently brake, accelerate, and steer the vehicle, the potential safety benefit of such systems merit further exploration. Method: This study attempted to isolate the field effect of the General Motors (GM) Super Cruise Level 2 system on lane departure and rear-end striking crashes. This system allows hands-free driving on GPS-mapped system-compatible roads and employs a camera-based driver monitoring system and a series of escalating alerts to prompt the driver to pay close attention to the road ahead and take steering control when takeover requests are issued. Police-report data were used to identify system-relevant and corresponding control crashes, and historic telematics-based data were used to understand system use surrounding crashes. A total of 131,757 Model Year 2017–2021 vehicles were matched to police-reported crashes from 12 states, which identified 415 analysis-relevant crashes on Super Cruise-compatible roads (certain limited-access freeways and trunk roads). Multiple statistical approaches were used to address challenges inherent to evaluating Level 2 systems, including small crash samples and system availability/use. Results: Results indicated there was no evidence for a difference in system-relevant crash risk for Super Cruise-equipped vehicles compared to matched highly-ADAS equipped vehicles without Super Cruise. The trend toward Super Cruise, as well as other L2 systems, being offered on an increasing number of vehicles with fewer Operational Design Domain (ODD) constraints should in the coming years notably increase the statistical power of the testing approach employed here. These results should not be interpreted as generalizing to other emerging L2 systems, including those that fundamentally differ with respect to Operational Design Domain limitations (e.g., lower speed operation only), requiring hands-on operation for system use, or the exclusive use of a steering-wheel based Driver Monitoring System (DMS).
期刊介绍:
Journal of Safety Research is an interdisciplinary publication that provides for the exchange of ideas and scientific evidence capturing studies through research in all areas of safety and health, including traffic, workplace, home, and community. This forum invites research using rigorous methodologies, encourages translational research, and engages the global scientific community through various partnerships (e.g., this outreach includes highlighting some of the latest findings from the U.S. Centers for Disease Control and Prevention).