CARS - Affiliates News
Statement of Professor J. Christian Gerdes on Monday, March 9
The following provides a transcript of the statement that Professor J. Christian Gerdes made on Monday, March 9, 2010:
"Good morning.
I am an Associate Professor of Mechanical Engineering at Stanford University and the faculty director of the Center for Automotive Research at Stanford (CARS). I want to disclose right away that Toyota is an affiliate member of that program, and supports other research programs in the engineering school, as well as the Global Climate and Energy Project at Stanford.
That said, the analysis I'm here to discuss and opinions I'm here to express have been formulated with complete independence. These conclusions are my own and should not be interpreted or described as an official position of Stanford University.
I also want to note that I am not here today to defend Toyota or to attack or discredit anyone. I am here to clarify the contents of a report which I believe has been misinterpreted. I am concerned that these misinterpretations are driving public fear and public policy. I want to help clarify the science behind the report and the broader picture of modern vehicle safety.
That broader picture is indeed bright. Driving has been getting dramatically safer in recent years. Traffic fatalities in the US have been declining since 2005 and fell from approximately 41,000 in 2007 to 37,000 in 2008. That trend continued nationally through the first half of 2009 (the most recent data I have) and many states reported record low fatality levels last year.
What is causing this improvement? One likely cause is the deployment of electronic safety systems such as Electronic Stability Control. Careful research by the National Highway Traffic Safety Administration estimates this technology would save between 5000 and 9000 lives a year if it were available in all cars. To ensure this technology is in all cars, NHTSA has boldly mandated ESC for the 2012 model year. Actions such as these by regulators and manufacturers are making driving safer.
But accidents still occur. Recently, several tragic accidents discussed in media reports and congressional testimony have raised concerns about the overall safety of vehicle electronics and Toyota vehicles in particular. As with any vehicle safety issue, it is imperative that these concerns be fully investigated and properly addressed. This investigative process, however, should be careful not to raise unnecessary and overbroad concerns about vehicle electronics and the lifesaving systems built upon them. There is a responsibility to public safety to build this investigation upon established facts, sound scientific reasoning and solid engineering analysis.
On Wednesday, February 24, Toyota representatives contacted me and asked whether I would provide an independent assessment of a report prepared by Prof. David Gilbert of Southern Illinois University at Carbondale. Such assessment is common in academia where peer review provides necessary checks on completeness and correctness before research results are published. They offered me access to any information I required from either Toyota or Exponent and asked for my honest opinion, whatever that was. For my review, I read the report and related Congressional testimony, participated in experiments at Exponent duplicating Professor Gilbert's demonstration and spoke on the phone with Professor Gilbert to confirm my understanding of his process.
In that process, Professor Gilbert creates a situation where a Toyota vehicle can have a wide open throttle without setting an error code and triggering the fail-safe system. To do this, he rewires the throttle pedal to include an extra component - a resistor. Based on the fact that the rewired pedal does not detect a short circuit fault that he then applies, the report concludes that there are circuit malfunctions that the Toyota system cannot detect.
I do not agree that this conclusion can be drawn from the information in the report.
Fundamentally, you cannot rewire a circuit and expect it to behave as originally designed.
- By adding an extra component, Professor Gilbert created a different pedal circuit than the one Toyota engineered.
- Professor Gilbert's circuit does indeed produce the failure he described.
- However, the report makes no attempt to describe a real-world condition that would cause the Toyota circuit to resemble Professor Gilbert's circuit.
- The report therefore contains no evidence of any real-world circuit malfunction that the Toyota system cannot detect.
To put it simply, the report draws conclusions about the Toyota circuit based on tests performed on a different circuit. It then establishes no link between the two circuits. Let me be clear - I am not stating that I disagree with analysis in the report that makes a link between the two circuits. I am stating the report contains absolutely no discussion of a link between the circuit tested and situations that might actually occur in a car. Material necessary to draw the conclusions reached is simply missing.
This report has been described in Congress and the media as demonstrating that there are holes in the Toyota system, a flaw in the design and that the system is not foolproof. The material in this report does not demonstrate any of those things. To be fair, none of these phrases are used in the report itself but this is the impression created. I am concerned that a misinterpretation of the facts by the public and officials could result in misguided policy and unwarranted fear.
Professor Gilbert was in no way wrong to run these experiments. In testing fail-safe systems, engineers do artificially simulate circuit faults. The engineer must then demonstrate that the simulated circuit is representative of the real-world condition being simulated. The report does not demonstrate that nor does it acknowledge that this is a simulation and not a real-world scenario.
The closest real-world scenario that would resemble the circuit in the report is what is known as a resistive short. I would conclude the following from Prof. Gilbert's study: It is worthwhile to search for evidence of resistive shorts and the other rare events necessary to produce his result. I do not, however, feel any stronger conclusion can be drawn from this material."
More information and coverage on the statement Professor J. Christian Gerdes made on March 9, 2010 can be found here:
Toyota March 2010 Webcast (VideoNewsWire.Com)
Toyota's Intended Rebuttal (MotorTrend.Com).
March 9, 2010 - 1:30PM