January 26, 2004 -- USA Today -- Cars

Ford tests ways to keep drivers from dozing off

DETROIT -- Within 45 minutes of taking a seat in Ford Motor's simulator for drowsy driving -- after being awake for 24 hours -- I had driven my virtual car off the road and into a ditch.

Ford (F) and its Volvo subsidiary are using the simulator to study how sleepy drivers react and to try out technology that one day could help keep them from going off the road or slamming into another vehicle. I agreed to be a guinea pig and do a stint in the machine. Clearly, I was proof that sleep deprivation is a risk on the road.

Drowsy driving accounts for 1,500 deaths and 100,000 crashes a year, says the National Highway Traffic Safety Administration. Statistically, it is less of a problem than drunken driving, but many incidents of sleepy driving are not reported by police because it can be hard to detect.

"Driver fatigue and falling asleep while driving is a deadly problem," says Jeff Greenberg, staff technical specialist for safety research at Ford. Greenberg manages the Virtual Test Track Experiment -- VIRTTEX -- at the simulator lab, the only one of its kind in North America.

That's why Ford -- and especially its Volvo unit, whose reputation is built on safety -- is studying sleepy drivers in VIRTTEX, which consists of a Volvo S80 inside a large box with hydraulic arms that create the motion. Ford plans to eventually test 24 or so people, with more than 20 tests completed so far.

The full-motion simulator creates a virtual-reality driving experience not unlike some theme park rides, such as Universal Studios' Back to the Future: The Ride, where riders get the sensation of flying without leaving their seats. Ford has also used the simulator to study driver distraction by having test subjects drive while fiddling with navigation systems, cell phones and radios.

To prepare for the test, I couldn't sleep for 24 hours and wasn't allowed caffeine for 14 hours. A wristwatch-style bracelet from Ford monitored my pulse and kept me from sneaking a few winks or an espresso. Drivers who defy the clock and refuse sleep before taking to the road would likely be drinking coffee or pulling over for a catnap. But Ford wanted me at my worst to check my reflexes, eye movement and heart rate. Keeping people awake when all bodily signals beg for sleep is a form of torture in some parts of the world. Now I know why.

The folks who run VIRTTEX say keeping their subjects up for 24 hours enables them to monitor all the physical signals of how people fall asleep. According to drivers organization AAA, 24 hours without sleep has an effect similar to a blood-alcohol level of 0.10%, the legal threshold for drunken driving in most states.

But it "varies all over the place for different people," says Greenberg.

NHTSA says a drowsy-driving incident typically occurs when the driver has had less than six hours of sleep in 24 hours and has been driving for just under three hours.

The agency began studying the effects of fatigue and drowsiness on drivers in 1996 and has since developed educational material for shift workers -- those who work other than daytime hours -- high schools and colleges.

Two of 12 people tested so far had to stop the test before completion. This happened to me, too. At one point, the tester who was monitoring my eye movement said I appeared to be asleep with my eyes open. He called that "interesting and unusual." One subject sailed through three hours of driving just fine.

According to NHTSA:

37% of drivers it surveyed in 2002 said they had nodded off behind the wheel at least once, and 8% admitted to dozing while driving in the previous six months.

1,500 deaths and 40,000 injuries a year are attributed to dozing. That's a serious number, though far less than the roughly 17,500 deaths and 500,000 serious injuries blamed on drunken driving. But, again, drowsy driving is under-reported.

55% of sleepy-driving accidents involve drivers under age 26.

Drowsy-driving accidents climb from Thanksgiving through New Year's, when late parties, long work hours and family trips are most common and when sleep-deprived executives and exam-addled students are on the road.

"We are so accustomed to being fat and tired and sleepy that it's part of our daily life, and we think nothing of getting behind the wheel and driving despite the horrible ramifications of that act," says Marcia Stein, a spokeswoman for the National Sleep Foundation, a non-profit education and research group.

The foundation says 34% of daytime workers report being close to sleep at work, and 25% of 25 million shift workers report falling asleep on the job.

Part of the problem is that 40 million Americans suffer from sleeping disorders. Among them are 18 million who have an increasingly common condition known as sleep apnea, in which oxygen-deprived, shallow sleep combined with heavy snoring makes sufferers notoriously sleepy during waking hours.

Barely half an hour into "driving" the Volvo S80 sedan simulator, I started careening into the left lane, cutting off virtual SUVs on a make-believe highway. Each time I veered from the right lane, one of several devices Ford and Volvo are testing -- a beeping alarm, a red light projected on the windshield, a vibration in the seat or rumble-strip sensation in the steering wheel -- tried to stir me awake. But I was too sleepy for any of them to do much good. I was jerked awake, but my reflexes were poor. Fifteen minutes later, I was in the ditch.

Because Volvo is taking the lead in the study, devices to shake drowsy drivers likely will wind up in its cars, built and sold in Sweden, before they arrive in the USA.
"There are a number of ways we can go, but we have to make sure the sensors and devices don't give false positives, so people don't want to disconnect them," says Greenberg.

Volvo is testing a lane-drift system that figures into an eventual drowsy-driving solution. The Lane Departure Module uses a camera with image processing software to detect current lane position. A computer linked to the camera constantly measures the distance to the left and right lane markings and detects variations. If it senses an unintended lane departure, it applies torque to the steering wheel, which eases the wheel back toward the intended lane, thus giving a nodding-off driver an extra chance. It's a technological "nudge." It doesn't take control of the steering.

The goal of the sleep-avoidance technology is not to give people a safety net so they are encouraged to keep driving when they should be sleeping, says Greenberg. It's designed to intervene before a vehicle hits a ditch or another car, to prevent an accident until the person can pull over.

Ford hopes that after Volvo introduces the features in a few years, it can make a business case for offering the technology in even its lowest-priced cars, which are bought disproportionately by young drivers. Ford has proved willing to gamble with such potentially intrusive safety technology.

It's the most aggressive automaker in putting "belt-minder" chimes -- alerts that keep going off until both driver and passenger are buckled up, rather than those that cease after initial warnings -- in many of its cars and trucks.

I try all the usual things to stay awake, slapping my own face, singing, all while being monitored. It's a little embarrassing. But I was denied the four things I normally turn to if I get sleepy when I'm driving: coffee, a cigar, an open window or just pulling over. Again, I drift into the left lane and am grazed by a virtual, thank God, tractor-trailer.

New Jersey was the first state to pass a law making it a crime to drive "knowingly fatigued," punishable with jail time. Lawmakers were outraged by a driver who had been awake 30 hours before causing a deadly crash in 1997. New York is considering a similar law.

Meanwhile, new federal regulations took effect this month that limit commercial truck drivers' hours. Truckers now can work 14-hour shifts, including loading, unloading, breaks and up to 11 hours of driving. Those shifts must be followed by 10 hours off the clock -- two hours more than the previous regulation. Federal regulators estimate 75 lives a year could be saved by the new rule.

There are about 5,000 fatal accidents a year involving big trucks, but there is no good way to know how many are due to truck-driver fatigue.

Fatigue scientist Martin Moore-Ede of Circadian Technologies estimates it could be as much as 50%. That's because many police reports cite "ran off the road" as a reason for the crash without knowing the cause.

The Insurance Institute for Highway Safety says truckers who have been on the road for eight to 10 hours have double the crash risk of those on the road fewer than eight hours.
Long-haul trucker Bill Grimes, 54, of Columbus, Ohio, figures he's on the road about 300 days a year and admits he gets tired. "But when that happens, I get off the road and catch a sleep," he said while getting ready to climb into his rig at a Milan, Mich., rest stop and head to Minnesota.

"You know what Dirty Harry said," Grimes said, referring to the Clint Eastwood film character. " 'A man's got to know his limitations.' Trouble is, most people don't."

Timeline of a sleepy driver

The staff at the Virtual Test Track Experiment lab kept this log of reporter David Kiley's time in the driving simulator. They began the log when he arrived at the lab. He entered the simulator 20 minutes after arriving:

30 minutes: Begin drive, training to demonstrate interventions and reaction time task.
45 minutes: Self-rated sleepiness a 7 out of 9 (sleepy, but no difficulty remaining awake). Performed reaction time task. 5% of the time did not react to lights displayed directly in front of him. An alert driver should be able to react to the light almost every time.
Throughout drive utilized several compensatory mechanisms to remain awake: eating, singing, postural adjustments, shaking head, drinking water, slapping face.
1 hour, 5 minutes: Eyelids noticeably droopy, slow blinks.
1 hour, 6 minutes: Experienced steering wheel torque and steering wheel vibration lane departure warning.
1 hour, 10 minutes: Stopped drive due to leg cramps.
1 hour, 40 minutes: Restarted drive. Fell asleep sitting in car waiting for simulation to begin.
1 hour, 45 minutes: Departed lane far enough to get a horn warning. (The horn warning is not one of the conditions we're testing. Instead, it's a last-ditch attempt to wake up the driver if they are so far off the road that a severe crash is imminent.)
1 hour, 46 minutes: Experienced steering wheel torque lane departure warning.
1 hour, 47 minutes: Kept departing lane and getting horn warnings throughout remainder of drive, sometimes as often as once a minute. Eyelids droopy, eyes closing.
1 hour, 56 minutes: Experienced steering wheel torque and head-up display lane departure warning.
2 hours, 15 minutes: Self-rated sleepiness a 9 out of 9 (extremely sleepy, fighting sleep). Performed reaction time task. 40% of the time (compared to 5% at start of drive) did not react to lights displayed directly in front of him. An alert driver should be able to react to the light almost every time.
2 hours, 20 minutes: Stopped drive due to extreme drowsiness and repeated lane departures.
2 hours, 25 minutes: Begins to discuss the experience with Jeff Greenberg. Appears more alert after walking around.
2 hours, 26 minutes: Sits down.
2 hours, 27 minutes: Falls asleep during the discussion.
2 hours, 30 minutes: Being driven home, Kiley begins to ask questions about the experience, falls asleep in mid-question. Sleeps for the remainder of the ride home.

Close this window.