IIHS eyes higher-speed test for automatic emergency braking

Now that nearly every new vehicle comes with automatic emergency braking (AEB), the Insurance Institute for Highway Safety is looking for ways to encourage even better systems that can prevent more severe front-to-rear crashes that occur at higher speeds.

Through its ratings of front crash prevention systems and an industry commitment it helped facilitate, IIHS sought to make AEB systems virtually universal. This goal has been achieved. Under the voluntary commitment brokered by IIHS and the National Highway Traffic Safety Administration, 12 out of 20 major automobile manufacturers equipped nearly all the light vehicles they produce for the U.S. market with AEB last year — well ahead of the September 2022 target.

The technology is already slashing crash rates by as much as half for vehicles equipped with it. However, the test that IIHS uses to evaluate AEB systems only represents a slim fraction of the rear-end crashes AEB is designed to mitigate, a new IIHS study shows.

“Thankfully, in the real world, AEB systems are preventing crashes at higher speeds than the maximum 25 mph our test program uses,” says IIHS Senior Research Scientist David Kidd, the author of the new paper. “The problem is that our current evaluation doesn’t tell us how well specific systems perform at those speeds.”

IIHS introduced the vehicle-to-vehicle front crash prevention test in 2013 and made a basic, advanced or superior rating a requirement for a 2014 Top Safety Pick+ award. Since 2017, an advanced or superior rating — which is also the level of performance specified under the manufacturers’ commitment — has been necessary for the lower-tier Top Safety Pick.

But the success of the program means that the test no longer effectively differentiates among systems: About 85 percent of the 2022 model year vehicles IIHS has evaluated so far earn a superior rating. For this reason, the Institute is dropping the vehicle-to-vehicle front crash prevention from the award criteria next year, though vehicle-to-pedestrian tests will still be required. Kidd’s study is the first step in determining whether the vehicle-to-vehicle test should be replaced and, if so, with what.

The soon-to-be-obsolete evaluation simulates front-to-rear crashes in which a vehicle approaches another vehicle stopped in the road. The test is conducted at both 12 and 25 mph. When the test program was being developed, the goal was to promote the adoption of functional front crash prevention systems, and research tests showed that those that performed best at 12 and 25 mph also did best at higher speeds.

In the new study, Kidd found that only 3 percent of police-reported rear-end crashes happen at such low speed limits. Increasing the speed of the IIHS test to 35-45 mph would make it relevant to more than 10 times as many police-reported rear-end crashes, he found. Expanding it to gauge a system’s ability to mitigate crashes with motorcycles and large trucks would also greatly increase the evaluation’s relevance to fatal crashes.

Rear-end crashes accounted for about a third of police-reported crashes in 2019 as well as nearly the same proportion of injury crashes and 7 percent of fatal ones. Previous IIHS research has shown that front crash prevention systems — including both AEB systems and those that only warn of an impending collision — could address as many as 70 percent of such crashes. On equipped vehicles, forward collision warning and AEB are already reducing police-reported rear-end crashes by 27 and 50 percent, respectively, another IIHS study has found.

On the other hand, research has also shown that the safety benefits of AEB are not as dramatic when roads are wet or covered with snow; when one of the vehicles involved in the crash is turning or changing lanes; and on roadways where the speed limit is 70 mph or higher.

To lay the groundwork for possible improvements to the Institute’s vehicle-to-vehicle AEB evaluation, Kidd used the federal databases of fatal and police-reported crashes to identify fatal, injury and noninjury rear-end crashes during 2016-19 in which the striking vehicle was a passenger vehicle. Then he calculated the proportion that occurred in scenarios like those covered by the IIHS testing program as well as those that happened at higher speeds and under different conditions.

The data showed that roughly 80 percent of police-reported rear-end crashes and about 40 percent of those that cause fatalities occur when a vehicle traveling straight strikes a stopped or slower-moving vehicle, as in the IIHS test. However, very few happen at speeds as slow as the IIHS evaluation.

The federal databases don’t track crash speeds, so Kidd used the posted speed limit as an indicator of the speed the striking vehicle was traveling before the crash. Only about 3 percent of all rear-end crashes, 3 percent of nonfatal injury crashes and 1 percent of fatal crashes happened on roads where the speed limit was 25 mph or less in scenarios similar to the one simulated by the IIHS test.

Not surprisingly, since speed is a major factor in crash severity, nearly 70 percent of fatal rear-end crashes happened in areas where the speed limit was 55 mph or higher. But about half of more common nonfatal crashes occurred on roads with speed limits from 35 to 45 mph — including 36 percent in a situation covered by the IIHS evaluation.

Separately, Kidd confirmed that speed limit is a reasonable proxy for travel speed at the time of the crash by using information from the event data recorders (EDRs) of vehicles involved in police-reported rear-end crashes in which at least one vehicle was towed from the scene. EDRs, which are installed in most U.S. vehicles built since 2005, record the speed of the vehicle and other information in the seconds before, during and after a crash.

Using that data, Kidd found that across all speed limits, the travel speed of striking vehicles involved in rear-end crashes averaged 2 mph lower than the posted limit.

In addition to speed limits, Kidd found another key difference between fatal and nonfatal crashes. A motorcycle, medium-sized truck or heavy truck was the struck vehicle in a small portion of nonfatal crashes but more than 40 percent of fatal ones. (In fatal rear-end crashes with motorcycles, it’s usually the motorcyclist who is killed, while in crashes with heavy trucks it is more often the occupants of the striking passenger vehicle.) Research tests with targets that emulate tractor-trailers and motorcycles are therefore needed to determine how well the results of the present IIHS evaluation apply to these vehicle types.

Based on Kidd’s findings, IIHS plans to conduct research tests on six vehicles equipped with different front crash prevention systems at speeds up to 45 mph. Tests will also be conducted using different types of passenger vehicles and other vehicles like a motorcycle and various sizes of trucks as the stationary vehicle.

At 45 mph, the evaluation would be relevant to 43 percent of police-reported rear-end crashes and 12 percent of fatal rear-end crashes, which occur at or below that travel speed.