Many in the industry resisted the just-adopted rule from the National Highway Traffic Safety Administration, which asks auto-makers to ratchet up their protection of unbuckled vehicle occupants.
The fear is the step could again lead to overpowered airbags, instead of pressing consumers to wear seat belts. This is part of a larger concern for safety experts: that improved technology might actually encourage drivers to be irresponsible.
That issue weighs heavily on the minds of Terrence Connolly, executive director of General Motors' North American safety center, and Tom Terry, GM's director of safety affairs and regulations. The two spoke with Automotive News Staff Reporter Aaron Robinson recently at the GM Technical Center in Warren, Mich.
If you could re-write any government regulation, which would it be and how would you change it?
Terrence Connolly: The one that's on our mind right now is FMVSS 208, which is the new frontal-crash standard. There seems to be a great advocacy for going back to a 30-mph unbelted barrier, which is fundamentally what drove the overpowering of airbags to begin with. The industry is of one mindset, along with the Insurance Institute for Highway Safety and American Medical Association, that a 25-mph unbelted barrier is much more appropriate.
Our primary concern might be described this way: With the safety system of the vehicle partially disabled - that is, the occupant doesn't put on a belt - should the vehicle and occupant performance have to be the same as with the total safety system in place? We think no, and our worry is that it could easily drive the re-powering of airbags to a level that's not good for small occupants.
Tom Terry: I would go back even further than that to when 208 was first written. Remember, that started out as a passive restraint standard because most people weren't wearing their belts. If you go back and look at the history of airbag regulation, it's been driven all along by the fact that people refuse to wear their belts. That's why we had automatic belts, and then Congress mandated airbags. We're the only country in the world that went this way. If I were to re-write it now with 70 percent seat belt usage, I would write it as a manual belt standard requiring people to wear their belts.
If you design the car to meet the 30-mph unbelted standard, how will that make vehicles less safe?
Connolly: It gets down to the basic physics of an occupant traveling down the highway. In a vehicle moving at 30 mph, that occupant has a kinetic energy associated with their own mass and the vehicle's speed.
At the point at which they hit something reasonably rigid such as a barrier, the occupant restraint system has to dissipate that energy. That's a matter of decelerating the occupant with some force, mostly on the upper chest and body, through some distance.
If you've got a large occupant of about 180 pounds, who is unbelted, that restraint system has to decelerate that mass through a fairly deep distance of the bag. So the bag has to be fairly deep.
For a small female who is sitting fairly close to the steering wheel, she has a lot less kinetic energy. The bag is too deep for that female and will throw her backward and create these neck extension moments that are above the human tolerance.
The industry has complained that the five-star rating system for crashworthiness is too simplistic. How would you rank the safety of vehicles?
Terry: It needs to be in a form that people can understand and digest. There's not a clear agreement on what that is.
We spent a lot of time on the original federal New Car Assessment Program, even deciding how to take two fundamental metrics - a head number and a chest number - and turn them into something that anybody can understand.
There were all sorts of theories as to what that should be. That's when they finally arrived at stars, which basically said, 'Well, they're all good, it's just that some are better.'
It sounds good to provide consumers a lot of information, but how do you integrate all that into a decision that makes sense?
We've put a lot of effort into generating ways to develop stars, but I think not enough effort in making it meaningful to the customer.
Is development of injury protection systems taking a back seat to crash-avoidance systems in safety engineering?
Connolly: No. I don't think we could really say that.
If you look at the people who would classify their profession as 'safety engineering,' most of them are involved in crashworthiness.
GM spends immense amounts of money on engineering strong crashworthiness into our vehicles every year. We also do spend a fair amount of money and engineering time on crash avoidance.
It's done in a whole variety of communities. It's done through the chassis engineering community, through the lighting engineering community and through the instrumentation engineering community. It involves every part of the vehicle.
How much are you depending on suppliers to develop crash-avoidance technology?
Terry: At GM, we feel it's our responsibility ultimately to integrate the safety performance into the vehicle.
Hopefully, we get a well-engineered and well-validated component or even subsystem from the supplier, but we're still going to feel it's our responsibility to engineer the vehicle safety performance.
What's the strategy to put crash-avoidance technology into vehicles at GM?
Connolly: We're interested in anything that helps steering and braking, the fundamental issues of control.
Visibility is another one. It includes not only visibility out of the passenger compartment, but lighting.
I'm talking about ways to use lights more intelligently to give not only the driver better visibility, but more signals to people around the vehicle about what it's doing.
Some people are looking pretty creatively at hydroscopic glass that doesn't collect rainwater so visibility in climate conditions is better.
Terry: There's another challenge, which is to make sure that the driver takes advantage of the system and doesn't overdrive because he thinks he has more capability.
There's a theory in the safety arena called the 'risk compensation theory.' If you give someone ability to stop in a shorter distance, then they'll follow the car ahead of them in a shorter distance. If you give them stability enhancement, they'll go around corners faster because they'll believe they can avoid the crash.
So we'd like to give them all these additions, but have them drive basically the same so that the incremental benefit that we put in can be realized, and that they can't negate it by overdriving the systems we put in.
What can GM do to make sure people are using these new safety systems properly?
Connolly: We definitely try to educate the customer on how to use any new technology.
For example, the 2000 Cadillac DeVille night vision system also comes with a video in the glove box.
It's not just an owner's manual that people don't want to sit down and read, but rather it's a video that's fairly entertaining and informative to watch.
Are you seeing more unconventional ideas from suppliers?
Terry: Manufacturers are looking for the breakthrough that meets all the demands of low cost, high tech and low mass.
We have a group here called New Devices. Their job is to take people's ideas and turn them in to the corporation. So if you as an individual have this brilliant idea, you come in and it will get a fair review.
We look for those other opportunities besides what comes in from big Tier 1 or even Tier 2 suppliers.
Connolly: The most promising development project that we've run, which is going to result in a production program, is with a small company called Flex Point. When we walked into it, they were about a five- or six-person firm out in Utah and they had the best piece of technology. It happened to come out of the Nintendo power glove from virtual reality.
We teamed them up with a major Tier 1 supplier to be able to implement the technology in a reliable high-volume manner.
What is the status of your efforts to improve computer simulation for crash and injury prediction?
Terry: We have an exercise going on with the Canadian government.
We are trying to develop the first standard that will involve compliance by simulation.
It isn't a dynamic test like a crash test. It's a belt-fit simulation where the government wants to prescribe certain characteristics of belt-routing and pressure on an anthropomorphic test device.
We've been working with them to allow us to do that through simulation.
We're optimistic that it will emerge as the first standard where you actually can show compliance through simulation, rather than from a physical test.
Whether we'll ever get to that on a full dynamic is possible. But we're going take that small step first.