|
Citation
|
LeBlanc D, Leslie A, Bogard SE, Peterson C, Flannagan CAC, Kiefer RJ, Bohannon L. 27th International Technical Conference on the Enhanced Safety of Vehicles (ESV); April 3-6, 2023; Abstract #: 23-0341, pp. 15p. Washington, DC USA: US National Highway Traffic Safety Administration, 2023 open access. |
|
Abstract
|
27th International Technical Conference on the Enhanced Safety of Vehicles (ESV): Enhanced and Equitable Vehicle Safety for All: Toward the Next 50 Years
https://www-esv.nhtsa.dot.gov/Proceedings/27/27ESV-000341.pdf
Over-the-air data was captured from thousands of vehicles in natural driving across the United States to evaluate the GM production Level 2 (L2) Super Cruise, Front Pedestrian Braking, and camera-based Adaptive Cruise Control (ACC) features. This research mainly focused on how often, when, and where customers used Super Cruise, and examined safety surrogate measures. Super Cruise allows hands-free driving on compatible GPS-mapped roads, which, at the time, included only limited-access freeways and trunk roads. Super Cruise uses a driver monitoring system that includes a face camera and a series of escalating alerts to prompt the driver to pay close attention to the road ahead and to take steering control when takeover requests are issued.
Two Cadillac fleets were used, including 2642 model year 2018/2019 vehicles equipped with Super Cruise, as well as 1196 similar fusion ACC-equipped model year 2017-2019 vehicles. During the 14-month data collection period the Super Cruise fleet accumulated 24M miles and the ACC-only fleet accumulated nearly 14M miles. Telematics data was retrieved each ignition cycle using GM's OnStar system, including Super Cruise state changes, harder braking events, and GPS traces. This research examined contextual influences on Super Cruise use and compared crashes, braking levels, and vehicle speeds across Super Cruise, ACC, and manual driving modes.
Super Cruise engagements occurred on 72% of equipped vehicles, totaling 1.7M miles of engaged driving. Engagements accounted for 18% of the driven distance on system compatible roadways, with a median engagement duration of 2.6 minutes. Relative to manual driving, Super Cruise and ACC were generally used when there is less surrounding traffic (i.e., free-flow, nighttime) and the roadway environment is less complex (e.g., rural, non-rain, non-curved roads). Drivers frequently interacted with Super Cruise via steering overrides (e.g., to change lanes), receiving and responding to driver attention prompts, and responding to takeover requests. Approximately 57% of Super Cruise engagements included driver attention reminder(s), with 91% of such reminders resolved without further escalating alerts. Takeover request results indicated 58% were due to Operational Design Domain limits, and 24% were driver-attention related.
No Advanced Automatic Crash Notification events were observed on Super Cruise compatible roads for either the Super Cruise or the comparison ACC fleet. Modeling analyses under matched driving conditions indicated that harder braking (exceeding 2.6 m/s 2 ) events were 1.7 times more likely during ACC/manual driving than during Super Cruise driving, and that median and top speeds during Super Cruise engagements did not exceed those of ACC/manual driving during free-flow traffic conditions, which was found to be the dominant use case for Super Cruise. Together with the relatively modest median Super Cruise engagement duration of 2.6 minutes, these findings suggest drivers use Super Cruise selectively, have generally fewer harder braking events when Super Cruise is engaged, and appear comfortable with the interactive Super Cruise feature. |