Study: Collision Avoidance Technology Continues to Struggle

New research from the Insurance Institute for Highway Safety (IIHS) is claiming that forward-collision warning and automatic emergency braking are less successful at identifying trucks and motorcycles. In a study that comprised more than 160,000 accidents, the IIHS asserted that these systems prevented accidents with regular passenger vehicles 53 percent of the time. However, motorcycles only benefited 41 percent of the time and trucks 38 percent.

“These reductions are impressive for all vehicle types, but the safety benefits could be even larger if front crash prevention systems were as good at mitigating and preventing crashes with big trucks and motorcycles as they are with cars,” Jessica Cicchino, IIHS vice president of research, explained.

From the IIHS:

Specifically, such systems could prevent an additional 5,500 crashes a year with medium or heavy trucks and another 500 crashes with motorcycles.

The effect on fatalities could be substantial too. A medium or heavy truck or a motorcycle is struck by a passenger vehicle in about 43 percent of fatal rear-end crashes, even though another passenger vehicle is the struck vehicle in 97 percent of rear-end crashes overall, an earlier IIHS study showed.

“Motorcycles and large trucks present unique risks,” Cicchino said. “Along with being hard for other drivers to see, motorcycles don’t have a steel frame surrounding and protecting the rider the way cars do. At the other end of the spectrum, large trucks are so massive that when a passenger vehicle hits one, it’s more likely to be fatal to the people inside the passenger vehicle. The height of large trucks can also result in dangerous underride crashes.”

To determine how well today’s front crash prevention systems are addressing these crashes in the real world, Cicchino and IIHS Senior Research Scientist David Kidd compared police-reported rear-end crash rates for model year 2016-20 passenger vehicles with and without AEB and forward collision warning when the struck vehicle was another passenger vehicle, a medium or heavy truck or a motorcycle. The crash data used were police-reported crashes during 2017-21 in 18 U.S. states.

There’s a lot of pressure to mandate specific advanced driving aids in the United States. Europe has started doing so and American regulatory policies have started to mimic what the EU has planned ever since the Biden administration came into power. Though the industry has been moving in the direction of standardizing frontal crash avoidance and similar technologies for nearly a decade and insurance agencies seem eager to see them broadly implemented.

The report itself actually praises the industry for entering into a “voluntary manufacturer commitment brokered by IIHS and the National Highway Traffic Safety Administration.” These kinds of public-private partnerships would have been frowned upon in the past, likely considered as borderline fascistic. But they’ve become increasingly commonplace among regulators, with corporate lobbying groups often in perfect sync with government policies.

That said, public concerns have lingered about the true efficacy of these systems. Federal regulators constantly harp on Tesla’s Autopilot and somehow manage to ignore nearly identical systems installed by other manufacturers. We’ve likewise reported on numerous studies showing that these types of features dramatically reduce driver response times and boast failure rates higher than we’d like to see.

Meanwhile, the hardware necessary for the systems to function properly are adding to vehicle cost. As MSRPs have risen in recent years, due to rampant inflation and corporate greed. But the relevant devices have likewise played a role in padding transaction fees and can also be quite expensive to replace when damaged. Additionally, consumer privacy advocates are worried these systems are invasive and set the stage for regulators to normalize driver monitoring — as many new driving aids come with in-cabin cameras that keep tabs on vehicle occupants.

The European Union has actually mandated driver monitoring by 2026, with the United States having done similarly under the guise of combating drunk driving. However, everyone watching the industry is under the opinion that the systems will function identically and will almost assuredly be cameras based.

Some of the above applies to your author. Though my biggest concern has been just how difficult it is to get good data on the topic. Interestingly, the IIHS already conducted a study pertaining to the effectiveness of forward-collision warning and automatic emergency braking in 2017 (with Jessica Cicchino at the helm). The research covered older vehicles, using more rudimentary systems. But the results were more-or-less the same — with automatic emergency braking allegedly helping drivers avoid rear-end collisions roughly 50 percent of the time.

Previous studies have suggested automatic emergency braking could reduce accidents by as much as 40 percent. But these, again, come from government-auto-industry-related partnerships and insurance agencies. The Partnership for Analytics Research in Traffic Safety claimed front-to-rear crashes were reduced by 49 percent when the striking vehicle had forward collision alerts and automatic emergency braking.

Studies from independent outlets have been less kind. The American Automobile Association (AAA) conducted some testing that basically showed most automatic braking systems struggled to perform in everything but the most ideal environments. Pedestrian detection was particularly disappointing. Though the AAA doesn’t seem to mind seeing the above installed into vehicles, even if it has claimed these systems are prone to fail above 40 mph or subjected to lackluster visibility.

The truth of the matter is that these systems are now so commonplace that it’s hard to get a bead on exactly how effective they are. Most independent studies are limited to testing a handful of relevant models, potentially giving less than the full picture. Meanwhile, larger papers typically focus on aggregate data sourced from police reports, government agencies, and insurance firms. But these rarely include any testing of the actual systems and typically end with generalized assumptions about how much good they’d theoretically do with more widespread implementation.

The IIHS occupies an interesting middle ground here. While it does seem to acknowledge some of the shortcomings associated with advanced driving aids and conducts routine safety testing on as many vehicles as it can, its association with the insurance industry and clear support of advanced driving aids make it difficult to endorse without caveats. Like the government regulators at the NHTSA, the IIHS likewise seems to ignore the proliferation of touchscreens and lousy infotainment designs posing legitimate safety hazards.

To its credit, the IIHS said it has plans to launch new vehicle-to-vehicle front crash prevention evaluation that takes some of these novel safety systems into account. The group believes those updated testing protocols will start impacting its overall vehicle safety ratings early in 2024.

The IIHS has also allied itself with researchers from government regulator Transport Canada to assess how frontal crash avoidance systems react to an array of obstacles. Thus far, the two organizations have examined five 2021-22 models from Acura, Subaru, Toyota, Volkswagen and Volvo confronted by twelve individual “nonpassenger vehicles or surrogate targets” and seven “different passenger vehicles or surrogate targets” at speeds above 30 mph.

“We’re pleased to be participating in this unique collaboration with IIHS, a leader in vehicle safety testing,” said Dr. Peter Burns, head of vehicle crash avoidance and human factors research at Transport Canada. “Transport Canada is committed to advancing road and vehicle safety in Canada and worldwide.”

The best we can say is the above is mixed news. With a few years of data to parse through, these kinds of safety systems clearly aren’t bulletproof and we’d advise all drivers against relying on them to save their bacon. But they can also intervene during a worst-case scenario to help avoid tragedy, assuming their very existence hasn’t simply made everyone a worse driver.

Maybe future testing conducted by the IIHS and its many partners will shed some light on the topic. But protocols will need to be focused on determining the true effectiveness of individual systems to be of any real use to consumers. We have heard enough expert opinions about how these systems “could” save a hypothetical number of lives by X date. Despite so much optimism, fatal accident rates have risen rather dramatically in the United States since 2015. That might be something the IIHS and government regulators might want to consider.

[Image: IIHS]

Become a TTAC insider. Get the latest news, features, TTAC takes, and everything else that gets to the truth about cars first by  subscribing to our newsletter.