News & Updates
The Washington Post | April 8, 2019
Coming soon: If you’re not sober, you won’t be able to start the car
New technology will prevent vehicles from moving for alcohol-impaired drivers
By Fredrick Kunkle
Someday soon, all of us may have to prove that we’re sober before starting a motor vehicle.
How could we prove that? you ask.
By doing almost nothing at all.
Since 2008, a government-funded research and development program has been designing a universal ignition interlock for all vehicles to prevent drunken driving. The built-in device would measure a driver’s blood alcohol level (BAC) to determine whether the person can legally operate a motor vehicle. If not, it would effectively take away the keys by preventing the vehicle from starting.
Such devices are already in widespread use for those charged or convicted of drunken driving. What’s different — perhaps even revolutionary — is that the built-in ignition interlock would make an instantaneous and precise reading of every driver’s BAC level when the driver attempts to start the vehicle. Eventually, the device could become standard equipment, just like air bags.
The device would take BAC samples in one of two ways. A breath-based system would gather a whiff of a driver’s ambient breath. A touch-based system would analyze the touch of a driver’s finger, perhaps from a vehicle’s starter button or the steering wheel. To win over consumers — and avoid the sort of outcry that occurred in the 1970s over an ill-fated ignition interlock intended to promote seat belt use — project officials say the built-in interlock has to be fast, precise and just about perfectly reliable in many different driving conditions. It also has to have safeguards against drivers who might cheat.
Officials behind the public-private effort to develop the technology — known as the Driver Alcohol Detection System for Safety (DADSS) — say the device will be ready for commercial fleets next year. Virginia’s Department of Motor Vehicles became the first state agency to use it in its fleet last year, and a private company, James River Transportation, is road-testing them in its fleet of Ford Flex crossovers.
To get a better idea of what’s involved with the technology, we sat down with Robert Strassburger, who is president and chief executive of Automotive Coalition for Traffic Safety. The coalition, which represents 17 automakers, is part of the public-private DADSS initiative with Virginia and the National Highway Traffic Safety Administration (NHTSA). Advocates say that if their work is successful, such a device — which requires understanding complexities involving the science of biology, spectroscopy, electrical engineering, consumer behavior and even politics — could save an estimated 10,000 lives a year.
Q: Can you update us on DADSS?
A: We intend to release by the end of 2020 a breath-based device for use in fleet applications and as a dealer-installed accessory. That is an important milestone in a couple different ways.
It’s along a pathway toward our ultimate goal, which is a device that can largely be deployed in all vehicles. We are beginning to prepare for the handoff of the technology from DADSS to other [automakers] and other manufacturers who may want to use the technology.
Q: You say the device will undergo real-world testing. Such as?
A: That would include baking these sensors to roughly 160 degrees Fahrenheit and freezing them to minus 40 degrees Fahrenheit.
Q: You’re also testing human subjects. Why?
A: All of this testing will help us to develop — for lack of a better word — a blood library.
That is one of the most significant challenges facing us in the development of this technology: How we, as individuals, absorb and eliminate alcohol is a function of our gender, our ethnicity, underlying health problems, [and] what we might be doing before or after we’ve consumed alcohol. All of that we have to understand.
Then we’re analyzing data to make to our measurement algorithms really robust.
Q: Aren’t there already lots of ignition interlocks on the market?
A: The interlocks that we have now are punitive devices. They are also zero-tolerance devices — meaning, if any amount of alcohol is present, they will lock you out.
They are very difficult to use — they require that you provide a very large volume of breath from the very depths of your lungs. Even people who use them regularly and are experienced in using them typically fail to provide a sufficient breath sample about 30 percent of the time.
And then, the technology has to be recalibrated roughly every year, dependent upon usage. If you use it more, you have to calibrate it more frequently.
Q: You’ve said those difficulties represent the kind of driver “hassle” that DADSS must avoid. Can you explain?
A: DADSS recognizes that 99.9 of drivers in the U.S. are not the problem and therefore won’t tolerate any inconvenience. We want to make sure our system is very easy to use, doesn’t hassle sober drivers, doesn’t require a lot of maintenance, etc.
So in the case of the breath-based system, you simply sit in driver’s seat and breathe normally. That’s all that’s required.
Q: There’s no mouthpiece?
A: There is no mouthpiece.
The other problem with those mouthpieces [besides some drivers seeing them as uncomfortable or intrusive] is they’re plastic and you can only use them about five times. We have a lot of concern about the use of [drinking] straws. If everybody were required to use a conventional interlock, that would lead to a lot of plastic waste.
The technology will be available for commercial fleets next year.
Q: How does the breath-based system work then?
A: Your breath is drawn into this optical cavity, where there is a mirror on each end, and an infrared light bounces back and forth in that cavity and measures the amount of alcohol present.
Now, the very first thing we did to try to keep the size of the device down was to bounce the light back and forth, and hence the mirrors on either end. But now we’ve shrunk that down to roughly half of what it used to be, again, by bouncing the light back and forth. Our ultimate goal is to shrink that again by half.
Q: What about the touch-based system?
A: The concept with the touch-based [interlock] is you would touch a vehicle control, and, with the mere act of pushing the starter button, we would know what the driver’s blood alcohol concentration would be.
If you’ve ever been to the doctor or the hospital and they clip that thing on the end of your finger that measures your pulse and the oxygen content of your blood, that’s a similar kind of concept. We’re looking below the surface of the skin at your capillary bed and measuring how much alcohol is in your blood that way.
[But] we’re not asking to clip anything on the finger. You just push the starter button or, say, grab hold of the gear shift to change gears. Some may choose to integrate this into the steering wheel.
Q: What other requirements have you set?
A: Our performance specifications are that this technology will live in the vehicle for the life of the vehicle, require little or no maintenance, and not require any recalibration at all. It would be very, very easy to use. Our ultimate goal is [that it be] fully passive — just sit in the driver’s seat and breathe.
In the case of the fleet device, we’re making an assumption that a small amount of inconvenience would be tolerated because there would be an employer-employee relationship. If the system failed to obtain a sufficient sample, you would be required to turn toward the sensor and puff at it like you were blowing out a candle.
Q: When you were testifying before a congressional panel recently, former NHTSA administrator Joan Claybrook — who helped lead the campaign to install air bags in cars — took you and the auto industry to task for dillydallying on technology that could save thousands of lives. She suggested that you worry more about hassling consumers than saving lives. What was your reaction?
A: Frankly, my reaction was — unfortunately, Joan often has this habit of being dismissive of concerns the industry has raised, and that comes at our detriment. She did that in the context of air bags, and that led to the first generation of air bags not being beneficial for some occupants.
[In the late 1990s, critics accused Claybrook of dismissing early warnings from automakers that air bags might kill or injure small passengers, including children — a criticism borne out when early air bag models caused the deaths of 39 children. Advocates for the passive restraints urged modifications but also pointed out that many more lives had been saved and urged modifications to increase air bag safety.]
[From the archives: From life-saver to fatal threat: federal government considers allowing manufacturers — and consumers — to pull the plug on airbags]
Q: You also mentioned that the project must avoid a rerun of the interlock rebellion that occurred after the federal government ordered automakers to install seat-belt ignition interlocks, which prevented the vehicles from starting until drivers buckled up. Can you explain the controversy and why you think it’s instructive here?
A: In the mid-1970s seat belt usage was between 10 and 15 percent. Not many people used their seat belt.
The [seat-belt interlock] systems that were put on vehicles for just one year, the 1974 model year — my dad had one, and I knew how to disable it because it performed so poorly — were not well-designed, and they did hassle drivers.
And there was such an outrage just from that single model year that Congress in 1974 enacted legislation prohibiting manufacturers from using seat-belt interlocks. And that prohibition stood until a couple years ago, when it was lifted because we do have technology now to make a more intelligent system, and the industry is researching more intelligent [seat-belt interlock] systems.
Q: Claybrook has called you the “industry excuse man” when you make that argument. What’s your response?
A: Well, again, as Joan mentioned, we’ve often testified together, so she often has pointed words for me that I’ve become accustomed to.
But in many ways engineering is the science of trade-offs; we can never have the ideal everything.
I think oftentimes what really separates the industry from, for example, Ms. Claybrook, is not the goal — I think we can all agree on the goal — it’s really the path we take to get there and the problems we need to solve in order to get to the finish line.
Q: What’s the biggest hurdle that remains?
A: Quite literally, the performance [specifications] that we set for DADSS fill a ring binder about six inches thick. Our focus has really been on speed, accuracy and precision of measurement and the overall size of the technology.
Q: Why speed?
A: We want to make a very precise very accurate measurement within a third of a second. People are accustomed to getting in their cars, sticking the key in the ignition or pushing the starter button and having the car start right away. They’ll be intolerant of any delay.
Q: And precision?
A: What is driving the very precise, very accurate measurement is that there are roughly 350 billion trips taken annually by passenger car. We are not able to tolerate even a very, very small percentage of failures because that would still be a large number when you’re talking about a billion starts a day.
Q: What about size — why is that a concern?
A: Taking a university chemistry lab and making it the size of your smartphone and finding a place to stick it in your car where it can live for 20 years — when the car sits out in the hot sun or sits in Alaska and freezes, and works day in and day out reliably without fail — has been our challenge.
For example, the first version of the touch-based technology was a device about three feet wide, by four feet long, by about 18 inches high. And it was that big because we use 40 lasers to look at 40 different spots in the infrared light spectrum where we would expect to find alcohol.
Q: That’s a lot of lasers.
A: Our ultimate goal is two to four lasers looking at those 40 different wavelengths. If we get to that point, then the system becomes the size of a postage stamp — a really thick postage stamp — that can be integrated into the push to start button.
Q: What other problems have you encountered?
A: We were using fiber optics, which we can only bend so much before they fail to work.
Q: You mentioned that the device also generates a lot of heat. Isn’t that a problem?
A: How we manage the heat that’s given off by those lasers working in close proximity to each other was both a challenge and a help, because we actually used that to get double and quadruple duty out of the lasers.
We get them to hop to different frequencies using different temperatures. But there’s got to be a way to manage that, so they also don’t overheat and drift all over the place, which means we sometimes have to cool them.
Q: How much has been spent on DADSS so far?
A: Thus far it’s been roughly $65 million.
[The funding breakdown is 60 percent from the federal government, 27 percent from industry and 14 percent from the commonwealth of Virginia, a spokeswoman for the coalition says.]
Q: Could the DADSS technology be used to prevent other forms of impaired driving?
A: While our focus at the moment will remain alcohol until we finish the job, it is possible with additional research to come up with a variant of DADSS that could measure THC. [Tetrahydrocannabinol, or THC, is the psychoactive ingredient in marijuana.]
And there’s a lot we’ve learned about laser development and laser packaging and how we power the lasers and how we turn them on and off that would be directly applicable to THC.
Q: What about people who might try to game the ignition interlock or tamper with it?
A: Part of our research is also looking at how we would know for certain that the alcohol measurement that we’re making is coming from the driver and not anyone else in the vehicle.
Q: Wait — why would an intoxicated person ask a sober passenger to start the car so that the inebriated person could drive? Wouldn’t it be more logical for the drunk driver to just let the sober person drive?
A: I think most people would say it’s irrational, nonlogical. But there are instances where a child is asked to use an interlock.
Q: That’s hard to believe.
A: It’s really nuts. Unfortunately, it’s happened, and not infrequently.
To read the original article, click here.