[vox] No electronic flaws in Toyotas?

[Brian Lavender]

On Tue, Feb 08, 2011 at 07:19:39PM -0800, Bill Broadley wrote:
> On 02/08/2011 11:16 AM, Brian Lavender wrote:
> > The government says that there are no flaws in Toyotas.
> > http://finance.yahoo.com/news/Government-No-electronic-apf-3420619474.html?x=0&sec=topStories&pos=6&asset=&ccode=
> >
> > According to the article, engineers at NASA "reviewed" 280,000 lines of software
> > code to look for flaws. I wonder if that means that Toyota's code is available
> > for download?
> 
> Heh, of course not.
> 
> > Or, do you have to sign an NDA that you won't reveal if you find
> > any vulnerabilities?
> 
> Well publishing a report saying "no flaws" seems like a pretty good vote of 
> confidence.  Or do you think there's a government conspiracy to have 
> unreliable automobiles in the USA?

No conspiracy. As seen in San Diego, people try to exploit these things.
There could be a real problem. Just because NASA engineers didn't find any
problems, one could exist. 

I believe that there may be a case when the vehicle suddenly accelerates,
probably in a case of hard braking. I am thinking that in the case
of where they represent acceleration using a vector. If the vector is
negative, one is decelerating. If it is positive, one is accelerating.
In the case of overflow for this vector, the vector would go from extreme
braking to extreme acceleration. Antilocks preventing brakes from braking
and full fuel going to the cyclinders.

For example, in 8-bit signed representation, the range is -128 to 127.
-128 minus 1 is 127 because of overflow.

A throttle cable with a butterfly flap would kill sudden acceleration
because the engine would be starved of adequate oxygen even if the
fuel injectors were blasting the engine with full fuel. The cable
provides a safety feature.

brian
-- 
Brian Lavender
http://www.brie.com/brian/

"Program testing can be used to show the presence of bugs, but never to
show their absence!"
Professor Edsger Dijkstra
1972 Turing award recipient