Auto accident fatalities are heading downward, and EDRs can help move them toward zero.
More than 50 years ago, the director of a Florida police academy told me the entire rationale for police traffic enforcement was supposed to be safety and only safety. Enforcement should be keyed to specific “accident” causes.
Nevertheless, as any motorist can attest, almost universally the object of traffic enforcement at the local level always has been revenue enhancement, with rare exceptions and sometimes harassment, which takes several forms, including DWB or driving while black.
This practice flies in the face of the responsibilities of the public safety officials charged with reducing crashes and their resulting injuries and deaths.
At the local level, these are the police accident investigators, whether beat cops or specially trained crash experts.
At the national level since 1970, this has been the mission of the National Highway Traffic Safety Administration (NHTSA), which is currently enmeshed in a controversy about its effectiveness arising from its handling of Toyota pedal entrapment and unintended acceleration matters, where as many as 60 fatalities are alleged. (Click Here for: Is NHTSA Underfunded?) FARS and EDRs
Today we have the two tools that could be extremely useful to study road crashes and get to their actual causes. These are the Fatality Analysis Reporting System (FARS) data collected from states nationwide by NHTSA, and the crash information collected in life-threatening crashes by Electronic Data Recorders (EDRs), or “Black Boxes,” in most vehicles – but not all – on the road today.
In essence, the FARS data tells what might have happened as surmised after the accident. It is the result of the police investigation with information that is gathered on the scene by a harried police officer who simultaneously must get the injured off to a hospital, direct traffic, clear debris and arrange for the wreckage to be towed away, and perhaps even notify the next-of-kin of those killed. Accordingly, the information may not be bulletproof as to details (example: a car identified as a four-door Mustang in one case.) but nonetheless, we salute the dedicated officers involved in this arduous task.
Then there is Black Box data, if the EDR (sometimes also called event data recorder) is properly programmed. An EDR provides information on a variety of vehicle parameters for the few seconds before and after an impact severe enough to trigger deployment of the vehicle’s air bags.
EDRs give solid scientific, objective, factual data about what happened during the crash within the capability of the programming and memory. (Here, recent Congressional testimony strongly suggests Toyota may have been cutting a few corners when compared with best practices in place at other automakers.)
GM led the EDR way
After General Motors had full-scale EDRs mounted in all its new cars and light trucks by Model Year 2000, NHTSA issued a Notice of Proposed Rulemaking (NPR) for EDRs in the summer of 2004, establishing a “voluntary” effective date of September 1, 2008, for the 18 light vehicle manufacturers marketing in the U.S. at the time.
NHTSA asked for comments on:
- A minimum set of specified data elements, such as speed, braking, crash forces.
- Common technical specifications for that data.
- Demonstrated EDR survivability in crash tests.
- Vehicle manufacturers recorded data is accessible for crash investigators to retrieve.
- Standardized owner manual language concerning the device.
In all, NHTSA identified 18 required data elements, plus standards for another 13 “if equipped” data elements. The recording tape loop, modeled after that in airliners, should extend from eight seconds before the impact to six seconds afterward, to cover up to three consecutive crash events to record possible rollovers or multiple impacts.
For example, the 2000 version of GM’s EDR recorded seat belt use, engine speed, impact speed and direction, change in velocity, braking and throttle operation, and air bag readiness.
Ten years later, Toyota’s EDR still may not record all these elements. In addition, there is only one tool in the U.S. (that is correct: one) which NHTSA can use to download the data. Even then, NHTSA needs the Japanese automaker to tell it what the data say.
GM worked with a California service diagnostics company, Vetronix, to develop software and hardware for convenient downloading of EDR data. By 2004, Ford was also aboard with the Vetronix equipment. Nearly six years ago, the California Highway Patrol already owned eight of these $2500 Vetronix units, while the Michigan State Police had four in service.
Well, that was then and this is now.
For whatever reasons, arguments or protests – presumably, by the affected companies, legislators worried about voter backlash, and product liability attorneys – NHTSA delayed the applicable date for the EDR regulation. This is just one area where safety advocates and critics charge NHTSA with being a “lapdog” of the auto industry. In hindsight, it is a tough charge to refute.
Apparently four years of lead time – 2004 to 2008 – was still not enough, and NHTSA’s website now indicates the applicable date was extended for an additional four years, until September 2012.
It is unknown to TheDetroitBureau.com whether or how much the specific data standards were watered down from the original proposal, which of course could have been flawed from a practical viewpoint.
From what Toyota has publicly stated either in Congressional testimony or news releases, as of February 2010 it had only ONE device in the U.S. capable of downloading data from EDRs installed in the millions of Toyota and Lexus vehicles on U.S. roads, although it was promising a handful more for NHTSA use. This is a scandal given the deaths and serious injuries incurred by Toyota owners, with 10 million or more vehicles now subject to safety defect recalls globally.
Further, there evidently were two problems with the Toyota offer to provide more EDR downloaders to NHTSA:
- Unlike the Vetronix units used for GM and Ford vehicles, which dealership service technicians as well as police investigators could access, only a Toyota technician could handle downloading the EDRs in Toyota and Lexus vehicles – not NHTSA engineers.
- It seems that the Toyota EDR was not designed to collect nearly as much vital data as the GM and possibly Ford units. (Ford did not respond to our requests for comment.)
Given that, in combination with the FARS data, output from the EDRs would provide crash investigators the most complete, accurate and objective information of serious crashes, it is hard for me to imagine why Toyota dragged its feet on this vital safety adjunct. The fact that all safety and recall decisions are made in Japan and the U.S. executives are mere figureheads in this important area is a factor.
While we can all rejoice in the significantly reduced number of 2009 highway crash fatalities, there does not seem to be any systematic relation of crash causes to NHTSA regulations. Although the information is available, it just does not seem to be collectively analyzed. (Click Here.)
Are there Federal Motor Vehicle Safety Standards that do not contribute to real-world safety, and if so should these be dropped? At what speed should crash tests be run for maximum effectiveness? Up to now, it has just been guesswork.
Finally, everyone knowledgeable about highway crashes believes that some proportion of single-vehicle, single-occupant fatalities are really suicides. In addition, how many other crashes are the results of anger or other mental abnormalities, or just plain excessive fatigue?
Except for alcohol or substance abuse, we do not know what factors occurred before the crash, and neither Black Boxes nor FARS can tell us. For answers to those questions, NHTSA and police agencies need forensic investigation of serious crashes, yet to be facilitated.
It is time.
Pros: accident investigation, defect detection, defense of OEMs against frivolous lawsuits, statistics for .gov, insurance companies benefit also
Cons: privacy of consumer may be compromised–libertarian organizations will fight hard against EDR rules, tinfoil hat crowd will freak out.
TAP: What’s the tinfoil hat crowd?
“Tinfoil hat” is internet slang for the conspiracy theorists and paranoid, such as those who think that 9/11 was an inside job, or those who think that the NSA is watching everyone’s internet use.
I think the trial lawyers will be a real impediment here. They have the most to lose…
TAP – LOL. I use tin foil to cover my chicken in the oven so it won’t overcook. I didn’t know it could protect me from cosmic rays…
As to the trial lawyers, it all depends on whether they are acting for plaintiffs or defendents. EDRs are, in effect, lie detectors. No more “I was only going 20” or “The brakes didn’t work” or, per Toyota, “The accelerator stuck wide open.”
The Libertarians ought to love EDRs. “The truth shall make you free.”
Assume for moment that there really is something wrong with either the programming of a throttle by wire system or the various sensors that it uses as inputs for its logic. I’m not saying there is, in Toyota’s case or anyone elses, but just assume for the sake of argument that there is. Now if the EDR “black box” records the same signals from those same sensors as the car’s ECUs, you could debug the programming and find a problem in it if there is one. But if the problem was in the sensors or all the wiring that connects those sensors to the car’s various ECUs, the EDR would probably record the same bogus inputs that the ECUs were acting on. In the hypothetical case of a sensor incorrectly reporting that the driver had stomped on and held the gas pedal wide open, both the car’s ECUs and the EDR are going to believe the driver floored it. That’s not going to help either the driver’s court case or the manufacturer’s trying to track down an elusive problem.
And yes, I are an engineer. I have a dusty old degree in EE and I’ve been a software engineer for almost my entire career except for a brief foray in into the not so lucrative world of auto repair when the tech collapse of the early 2000s made it impossible to get a job writing software.
Bryan: Good points up to a point. Isn’t it helpful to eliminate the false data, i.e driver claims as to what was going on, and then move to your very real issues as to whether the EDR supports certain assertions?
And maybe we need some safeguards here – ones that go back to English common law, such as innocent until proven guilty. And, maybe, the driver owns the data and will only release it if fifth amendment rights aren’t violated, or some such provisionis in effect.
Another approach would be to release the data generically, stripped of specific driver data so it couldn’t be used against him or her.
Aviation uses this approach – if the pilot reports an incident voluntarily.
If it’s faulty sensor input to the ECUs/EDR (either because the sensor is sending the wrong thing or because of a possibly transitory wiring issue) you’re not going to eliminate false driver claims. I’ve seen this thing first hand. When Ford first came out with the 6 liter diesel in the super duty pickups, the things were having serious drivability issues left and right due to wiring issues that were impossible to pinpoint the cause of. They did a lot of buybacks from customers because they shops couldn’t fix the things.
Bryan, you’ve touched on what we have wondered could be the hidden defect in Toyotas–a possible transitory wiring or electronic issue, which would misbehave sporadically and leave no trace. This is a very tough nut to crack with both Toyota engineers and frantic drivers telling different truths as each perceives it.
Then you wonder whether it is panic, publicity or whatever that suddenly results in myriad complaints about Toyota/Lexus run-away acceleration but not other makes in any meaningful way. We suspect it has not come up before, because Toyota owners have been true believers in their vehicles, and simply not reported problems which seemed to be against the conventional wisdom of invincibility.
Bryan: Scary thought – we have a whole generation of cars out there that can’t be properly diagnosed or serviced due electronics?
Automotive sensors, wiring, insulation, and (usually plastic) connectors live in a very harsh environment. They experience extreme heat, cold, vibration, caustic gasses and chemicals, moisture, dryness, dirt, and chafing against hard sharp surfaces on a daily basis. The more sophisticated automtive electronics become, the more components of the above types there are. And they are all provided by the lowest bidding, third party, barely solvent, often offshored vendor. And these vendors are changed or multi-sourced frequently – even during model years.
In the example that I gave above of the 6 liter diesels pickups, Ford redesigned the engine wiring harnesses and the components that they connected to at least three times during one production year. When parts had to ordered for them, you had to know which version of the 200x model year you were dealing with. In one of the drivability cases that I remember, a regional factory service guru told the dealer tech that the problem was probably a small hole in wiring insulation that resulted in some sort of intermitent arcing/shorting that we’d never be able to find. This was after extensive testing by both the dealer tech and the regional service guy. So they authorized the a buyback of the F350 from the long suffering customer.
Now imagine this same type of scenario with every major system on a modern car – engine management, braking, suspension, steering, drive train, etc. This stuff is complicated.
Bryan; I’ve spent most of my years in electronics and as much as I have worked with quality equipment,I have seen serious issues with quality control as the business chases third world countries for the lowest price. Think of how I feel about non-standardized equipment taking control over my brakes and steering!
I’ve recently looked at some of the quality coming out of China and it reminds me of the earlier crap out of Taiwan. Eventually with further investment their quality improved. But then it was on to the next cheapest source.
Although electronics have done a wonderful job controlling emissions and fuel consumption, cramming so much of it into vehicle control makes me nervous. I owned an ’85’ Audi 5000 and my daughter owns a 2010 Prius. My wife and I both almost went through the garage wall and my daughter came inches from the vehicle in front of her after braking over a sewer lid. Need I say more?
Every one of your points is pertinent. The industry needs to slow down in its pursuit of bells and whistles to appeal to our insatiable appetite for fun & flash. These complicated systems make the vehicles impossible to work on without serious training and test equipment.My opinion is that we are getting ahead of our ability to understand these systems and how they can fail.
We all just need to become better drivers. Darwin will weed out the bad ones and we can save
money on repairs. Making the car be a better driver is just another way of making us less responsible for our own actions. (oops,sorry that’s what lawyers are for!)
Hey Len,
All good points.
The problem for car makers is that they feel the need to cram ever more features that depend on electronics into their vehicles. In an ideal world, these features provide good things: better fuel economy, lower emissions, better performance, safer braking by way of ABS, safer handling by way of active suspensions. All make for better vehicles for all drivers – good, bad, and in between.
There are two problems with this though. Problem one is that design and implementation of these systems has not matured yet. Imperfect control software is an example of this. Problem number two is the one you highlighted so well. It’s also the one that we all deal with day in and day out with everything we buy and our ability to earn a living – the Wall Street driven race to the bottom in terms of costs over quality that all companies are engaged in these days. How much better and safer would all these complex electronics systems in cars be if the manufacturers spent an extra $500 per car just to beef up the reliability of sensors, wiring, and connectors? Just not buying them from the same manufacturers and/or countries that supply crappy disposable junk to Wal-Mart would be a good start.
Bryan; I don’t know what you do for a living but, have you purchased bearings for any purpose lately? Well I have. I bought a Dayton blower about nine months ago and it has already failed. I looked ,and guess where the motor was made.(hint it begins with chi and ends with na!) “Na” is what we should say to doing business over there.
I told my rep that they have a reputation for quality that could be lost in a hurry (I won’t say the company name but I’m sure you know it.)
I have also had bearing failures in my truck because I didn’t pay attention to its origin. Also in other products I work with, including pumps and controls. When I ask my vendors why the bearings don’t have numbers on them, I don’t get an answer… hummm
I wish, as Americans, we could replace the word price with quality.