Twenty auto makers recently agreed to equip cars with automatic brakes that activate when the driver has failed to do so. How does it work, and what are the advantages and disadvantages of the technology? We reached out to Brian Cooley, editor at large at CNET’s automotive site, Roadshow, for the answers.
KPCC: In less than six years, 99% of all cars sold in the U.S. will be able to sense when they’re moving too fast for the traffic ahead of them and automatically apply the brakes to emergency stop. Why did auto makers want to do this?
Cooley: The idea here is that since we had airbags, anti-lock brakes and stability control, we haven’t had a major breakthrough in safety. And those big three which all of us know on our cars at this point unless you drive a really old car really changed the number of fatalities in the U.S. and brought us down to what’s been the floor for the last few years of about 32,000 or 33,000 fatalities. That’s way down from where it used to be a decade or two ago before these technologies kicked in. But then we stayed at this stubborn level of a rather high number of fatalities. So the next big breakthrough that goes after a wide swath of human fallibility is to hit the brakes when we simply forget to, are too distracted to, are too drunk to or are too sleepy to.
KPCC: How does it work?
Cooley: It uses some sort of sensor that looks forward. It’s typically radar, but it can also be camera based, and this sensor is looking from the front of your car, the grille of your bumper, at whatever’s in front of it. It can be tuned to recognize only something as big as a car or to recognize something smaller like a pedestrian that will vary somewhat by automaker and then it very simply hits the brakes. And the way it does this is because the car has anti-lock brakes, it already has sort of an automatic braking pump. If you’ve ever hit the brakes really hard and felt the ABS kick in, you feel that kind of a chattering in your brake pedal. That’s this automatic braking pump doing its thing. Well that same pump can be used by a computer chip for automatic emergency braking the same way. It just gets a signal from the sensor in the front saying, you know what, we’re about to hit something, and the computer tells this automatic braking pump to hit the brakes. So part of the hardware was already in the car for a couple decades now.
KPCC: As a reviewer, it’s almost impossible to test these systems because it involves speeding toward another car.
Cooley: This is the hardest thing to review, isn’t it, because you have to be wiling to put yourself in harm’s way to basically be driving toward something and hope that the system kicks in and that your career as a reviewer isn’t about to get a black eye because it doesn’t work. It has to be well calibrated or it can cause alert fatigue.
A lot of the cars I’ve driven have three or sometimes four levels of warning. You can say warn me close, mid or really far out depending on how sensitive you want it to be, so that can be a driver option. And of course, it’s important for this technology to work well. The first time it doesn’t work, even though no car maker is going to guarantee that it works, they never do, they never can, you’re going to say this is junk I don’t trust it now.
So there’s always going to be a perception issue around any technology like this, but this is one that a lot of people will be very aware of. Unlike stability control, they don’t always know that it’s working. They just know that they didn’t flip their car. Or they didn’t spin their car, but most drivers don’t know what that’s like anyway. Whereas a lot of us do know what it’s like to make a panic stop, or to not, and to rear end someone. So we really have a relationship with what this technology does and can I think assess it either well or ill.
KPCC: Why is this happening now? Is it part of the march toward autonomous vehicles or is it because so many people are still using cell phones when they’re driving, even though a lot of states outlaw that?
Cooley: It certainly is a confluence of the fact that we have more distraction in the car. Distraction’s not new. We’ve had radios and conversations and lunch and makeup and newspapers certainly since the post-war automotive explosion, so distraction is only getting more complex obviously and more common. It’s about time we had this technology. We need it more now than ever.
Now one of the gripes you’ll hear about it going the other way is from a lot of the safety groups out there that will tell you, you know, this was an agreement by automakers to put this in all new cars by no later than September 1, 2022. Toyota, by the way, will get it out years earlier, they say, but for the most part it’s fall of 2022 when these car makers have agreed to do it. That's not the same as being regulated and ordered to do it.
And it also means that they can largely do it their way — per car maker — and not necessarily according to a specific standard that is going to be set in the federal motor vehicle code. So that’s got a lot of the safety watchdogs concerned. They say, you know what, the auto makers kind of got a pass here. Normally, this is regulated and they’re afraid it's going to be patchwork and uneven in terms of how well it’s implemented. We’ll have to wait and see
KPCC: The safety groups will never be satisfied though.
Cooley: What’s interesting here is this concept of you don’t get a second bite of the apple. If you get auto makers to agree to this, it’s much harder, the theory goes, to go to a politician and say let’s get a bill now that’s going to require automated braking and specify how it works. That politician may say I don’t want to spend a lot of my campaign time or my legislative time doing something that was already done. So that’s kind of a clever angle here if indeed the automakers have taken it that way.
They likely will not be regulated on this any time soon. This is one of the fist technologies to arrive from the future, wwhich is from the automated, autonomous car of the future. This is the first technology that really is going to step in and do something you should do when you apparently aren’t going to do it. And that is really robotic type stuff. It’s using sensors, it’s using logic, it’s using software that’s analyzing a situation in real time. And while anti-lock brakes and airbags do that also, this is the first piece that actually is coming from the future toolbox of self-driving cars.
KPCC: There's been a lot of speculation about what people will be doing in their cars once they no longer have to sit behind the wheel and control it. The level of amorous activity is likely to increase, won't it?
Cooley: You used to have to pull over to do that. Park somewhere. Now you can stay on your way , get there on time. This is clearly an area that auto makers do not want to see be squandered -- any degree of autonomy.
You know Google recently, one of their guys was talking at an autonomous car conference about how they use employees to do some testing of their cars on real roads, and they put cameras in their cars to watch their behavior. In the first hour or so, as they describe it, the employees are very nervous. They’re grabbing at the wheel. They’re looking around to make sure the car’s going to work right.
And they say within an hour, in many cases, a formerly nervous occupant was suddenly leaning their seat back to take a nap, was reaching around and turning completely behind them to get a laptop out of a bag in the back seat for like 30 or 40 seconds. Doing things that were immediately too comfortable with the technology. It seems as though human nature, after a while, in a short while, we pretty quickly start to trust these systems at least in some of the early human machine tests.