Inadequacies in current systems highlight challenges to achieving full autonomy
We’re just two or three years away — maybe less — from having fully autonomous vehicles in our showrooms and on our roads, if you believe the most optimistic hype about the pace of such technology development. Don’t believe it.
Recent testing by the highly-regarded Insurance Institute for Highway Safety (IIHS) not only casts doubt on the feasibility of such positive projections, it raises questions about the efficacy of widely-accepted Advanced Driver Assist Systems (ADAS) already on the market.
Adaptive cruise control (ACC) and active lane-keeping are two of the most basic ADAS features on the ladder of progression leading to full autonomy. They’ve been on the market for several years now and are offered by multiple automakers so their level of technical maturity should be well-established.
But road and track tests conducted on vehicles with these systems by the IIHS, as part of a program to develop a consumer ratings protocol for them, led to some surprising and disturbing conclusions.
“The early results underscore the fact that today’s systems aren’t robust substitutes for human drivers,” the IIHS said. “The outcomes ranged from the irksome, such as too-cautious braking, to the dangerous; for example, veering toward the shoulder if sensors couldn’t detect lane lines.”
In essence, adaptive cruise control, which is marketed under a variety of brand names, maintains a preset speed and/or following distance from the vehicle in front. It is designed to adjust vehicle speed to match that of a slower vehicle ahead, in some cases including coming to a full stop.
Active lane-keeping, also marketed under a variety of brand names, provides automatic steering inputs to keep the vehicle within its lane. Typically, such systems also require drivers to maintain a hold on the steering wheel.
These systems are considered to be Level 2 on the SAE International scale defining levels of autonomy, where fully self-driving capability equates to Level 5. They can assist with steering, speed control and following distance, but the human driver is still in charge and responsible for control.
“The (IIHS) tests are an outgrowth of our research on Level 2 autonomy,” says Jessica Jermakian, IIHS senior research engineer. “We zeroed in on situations our staff have identified as areas of concern during test drives with Level 2 systems, then used that feedback to develop road and track scenarios to compare vehicles.”
Five representative vehicles were used in the test program, although several other models also offer the technologies. These five also were equipped with automatic emergency braking systems rated superior by IIHS.
One of the questions researchers looked to answer is, “do the systems handle driving tasks as humans would?” Not always, was the resounding answer.
Engineers evaluated the ACC systems in four different sets of track tests to see how they reacted to vehicles stopping ahead and lead vehicles exiting the lane, as well as how the systems accelerate and decelerate. All performed satisfactorily, to varying degrees, in those controlled tests but some proved less satisfactory in real-world traffic situations.
Out on the real road, engineers noted instances in which some vehicles failed to respond to stopped vehicles ahead. In addition, unnecessary or overly cautious braking occurred for incidents such as tree shadows on the road and oncoming vehicles in another lane.
“The braking events we observed didn’t create unsafe conditions… however, unnecessary braking could pose crash risks in heavy traffic, especially if it’s more forceful,” Jermakian says.
To evaluate active lane-keeping systems, IIHS engineers focused on two typically challenging situations — curves and hills. They also observed how the systems performed in traffic.
Most of the systems tested failed to consistently keep the vehicles in their lanes on curves, often requiring the driver to provide additional steering to successfully navigate them.
The IIHS engineers had noted early on that systems that rely on seeing road markings to keep vehicles in their lanes were sometimes flummoxed by hills. As a vehicle crests a hill, the lane markers on the road beyond visibly disappear.
All the vehicles failed to maintain their lanes at some point in a series of hill tests, in some case frequently and alarmingly. One vehicle, when cresting a particular hill, swerved left and right until it determined the correct place in the lane, jolting the test drivers.
The bottom line from all this testing is clear. “A production autonomous vehicle that can go anywhere, anytime isn’t available at your local car dealer and won’t be for quite some time,” says David Zuby, IIHS chief research officer. “We aren’t there yet!”