Honda "Drive by Wire" question... what if the power goes out?

Bucky wrote:
>
> Pete C. wrote:
> > I have no problem with electronic linkages replacing mechanical ones.
> > Where I have a problem is when they try to interject computers in
> > between and have them second guess my judgment.
>
> I know where you're coming from. I personally have only bought manual
> cars because I hate software deciding when to shift for me.

Also the reliability of being able to trash the transmission and still
limp along.

>
> > No computer in any even
> > remotely affordable vehicle can ever (in my lifetime) have as much
> > sensory input as a human and therefore does not have the information to
> > even begin to second guess the human.
>
> This is partially true. You won't get as good feedback from an
> electronic linkage as opposed to a physical linkage. Especially for
> cases like braking and steering. But there are certain scenarios where
> a computer's sensory input is far superior to a human's. For example,
> the airbag. Do you want a computer sensor and software deciding when to
> deploy the airbag, or would you rather do it manually based on your
> human response?

First off, I do not allow air bags in any vehicle I own.

Second off my point was that no vehicle priced in anywhere near an
affordable range even for the rich can have sufficient sensor inputs to
even come remotely close to the sensory inputs of a human.

The car computer can not detect ice ahead in the road, it can only
attempt to detect when the wheels are already slipping and even then it
has difficulty if you are not also attempting to brake.

The car computer can not detect potholes or other obstacles in the road
ahead where you need to swerve to avoid them.

The car computer can not detect when bumpy conditions are causing a
wheel to bounce off the road and lock during light braking as opposed to
actual wheel lockup during hard braking. This is a current problem with
ABS systems on heavier vehicles with stiff suspension.

None of the sensors on current autos are redundant and the computer
quite readily gets very confused when there is a sensor failure and can
make the wrong decision. This has been known to happen when a wheel
sensor fails on an ABS system causing the computer to think the wheel is
locking and the ABS system reduces the braking force and causes and
accident.

Simply put, sensor technology is either not sufficiently advanced to
detect a condition, or is far too expensive for deployment on a common
auto for the computer to have sufficient information to try to second
guess the human operator.

Pete C.

Sparky Spartacus wrote:
>
> Pete C. wrote:
>
> > Bucky wrote:
> >
> >>Don Bruder wrote:
> >>
> >>>Tell ya what, Ellie... You drive that "fly-by-wire" unit. Just do me a
> >>>favor: Warn me where you'll be so I can avoid that
> >>>deathtrap-looking-for-a-victim.
> >>
> >>As others have mentioned, the Civic only has "throttle by wire". So in
> >>case of electronic failure, it would just slow to a stop. The car you
> >>really need to be watching out for is Mercedes, who has had "brake by
> >>wire" in its higher end models since 2003.
> >
> >
> > Brake-by-wire is little different than the brake-by-oil which is the
> > norm on pretty much every car and light truck and even some heavy
> > trucks. Of course the brake-by-oil is a split system and you still have
> > the cable operated parking/E-brake as backup. Presumably the
> > brake-by-wire system has similar redundancy and the good old cable
> > operated backup.
> >
> >
> >>>I want, and will settle for *ABSOLUTELY NOTHING LESS* than *DIRECT*
> >>>mechanical or hydraulic control of all major vehicle functions,
> >>>particularly steering and braking. Something that works *NO MATTER WHAT*
> >>>in every situation short of total catastrrophic failure.
> >>
> >>I'm sure most people, including me, are uncomfortable with the concept
> >>of taking away direct control. But hypothetically, let's say mechanical
> >>catastrophic failure occurs 1 in a million chance. And after much
> >>maturation, electronic "drive by wire" systems only occur 1 in 10
> >>million chance. Which system would you prefer? (Again, that is just
> >>hypothetical, it may not be feasible to implement such a reliable
> >>electronic system cheaply). I would take the more reliable one.
> >
> >
> > I have no problem with electronic linkages replacing mechanical ones.
> > Where I have a problem is when they try to interject computers in
> > between and have them second guess my judgment. No computer in any even
> > remotely affordable vehicle can ever (in my lifetime) have as much
> > sensory input as a human and therefore does not have the information to
> > even begin to second guess the human.
>
> I would disagree here - the automated systems probably do a better job
> than *most* humans, but not the very best (which we all are here, of
> course).

Wrong. Computers can react faster than humans, but without a boatload of
very expensive sensors they do not have enough information to properly
make the decisions in all cases. When one of the non-redundant sensors
fails and in a common auto they are indeed non-redundant and produced by
the low bidder, the computer makes the wrong decision and in the case of
current ABS systems has been known to cause accidents.

Pete C.

wrote:
>

<snipped>

>Road injuries are a big problem,

Actually they are not, they are extremely low relative to the number of
vehicle miles.

> clearly our current
> cars are not safe enough!

Clearly our current driver training and licensing is insufficient. Try
upgrading training and licensing to the level of a class A commercial
license and watch the number of accidents plummet along with the number
of clueless drivers.

<snipped>

Pete C.

Sparky Spartacus wrote:
>
> Pete C. wrote:
>
> > Don Bruder wrote:
> >
> >>In article >,
> >> "Pete C." > wrote:
> >>
> >>
> >>>Mike Romain wrote:
> >>>
> >>>>I have always gotten a laugh out of such a foolish system.
> >>>>
> >>>>The way I understand it, basically if the power fails when you are at
> >>>>speed, you crash, just like on an airplane. I guess you could also put
> >>>>your head between your legs and kiss your ass good-bye like they do on
> >>>>planes too... ;-)
> >>>>
> >>>>Mike
> >>>>86/00 CJ7 Laredo, 33x9.5 BFG Muds, 'glass nose to tail in '00
> >>>>88 Cherokee 235 BFG AT's
> >>>>Canadian Off Road Trips Photos: Non members can still view!
> >>>>Aug./05 http://www.imagestation.com/album/in...?id=2120343242
> >>>>(More Off Road album links at bottom of the view page)
> >>>>
> >>>
> >>>Most of the fighter jets are fly-by-wire and they generally don't have
> >>>problems with failures of this system. The fly-by-wire system is
> >>>multiply redundant and rather expensive though. The main reason for it
> >>>is the fact that it allows faster response and allows the computers to
> >>>assist in stabilizing some inherently unstable aircraft designs.
> >>
> >>The F-16 - Proof that even a brick will fly if you can cram a big enough
> >>engine into it...
> >>
> >>
> >>>In the auto world drive-by-wire would be constrained by the price points
> >>>and the multiple redundancy would probably be sacrificed. At auto speeds
> >>>the faster response of by-wire technology is not needed, so the only
> >>>possible reason to use the more expensive technology would be to allow
> >>>the computer to try to compensate for a drivers lack of skill.
> >>
> >>Or more accurately phrased, to allow the computer to *ATTEMPT* to
> >>compensate for what it *PERCEIVES* as operator inability.
> >>
> >>When I turn the steering wheel, the wheels better move correspondingly
> >>*EVERY* time. Not "just when the engine is on", not "When there's a
> >>charged battery installed", not "When the computer thinks that what I'm
> >>doing is OK", but *EVERY* *SINGLE* *TIME* *NO* *MATTER* *WHAT*. While
> >>I'm behind the wheel, I will accept *ABSOLUTELY NOTHING* less than
> >>*TOTAL*, godlike control of that vehicle, subject to *NO* influences
> >>outside of my own decisions and actions.
> >
> >
> > Indeed, one of the reasons I demand a manual transmission.
> >
> >
> >>(By way of illustration, a few years ago in europe, a "fly by wire"
> >>plane decided it knew more than the pilots - Pilots said "We gotta
> >>hammer on the power and crank the bejeezus out of the controls so we can
> >>lift, or we're gonna crash!". Fly-by-wire system said "Sorry, you can't
> >>do that", and proceeded to "fix their mistakes" by throttling down and
> >>not permitting them to crank the control surfaces to the needed degree,
> >>which caused the plane to crash and burn. After something like that, I
> >>can't see *ANYBODY* with a functioning brain-cell wanting anything to do
> >>with getting into a machine that might decide at any time that what
> >>they're trying to do is "off limits".)
> >
> >
> > You're partly confusing two different things, the fly-by-wire i.e. no
> > mechanical link, and an automated control system. Not really the same
> > thing although the fly-by-wire makes implementing the automated control
> > easier. If it's a simple electronic replacement for a mechanical link
> > (with suitable redundancy) it's ok with me. Automated control trying to
> > second guess my decisions based on far less sensory input than I have,
> > is not ok with me.
> >
> > The hydraulic steering on a lot of tractors and construction equipment
> > that was noted by another poster is a good example of basic fly-by-wire
> > or in this case fly-by-oil technology. It makes no attempt to second
> > guess the operators decisions and simply replaces what could be a very
> > complex mechanical linkage with a couple of nice flexible hoses.
> >
> > The hydraulic brakes in cars is another even earlier example.
> > Brake-by-oil basically, and we still require the mechanical cable
> > operated backup system in addition to the split hydraulic redundancy. Of
> > course in recent years they've added the automated control a.k.a. ABS to
> > try to second guess the operator.
>
> This is a good example of who benefits & who is penalized, i.e., ABS
> undoubtedly saves more asses/lives than it costs.

Better driver training would save more lives and cost no additional
lives, unlike failing ABS and airbags.

>
> I can understand why manufacturers would put automated vehicle stability
> on unstable vehicles like SUVs, same logic as above.

SUVs are *not* unstable by the wildest stretch of the imagination. Under
any normal driving conditions they are as stable as anything else on the
road. Under limited emergency conditions they can become unstable, just
as a regular car can become unstable, when in the hands of an unskilled
driver.

Note that a standard 80,000# tractor trailer has a significantly higher
center of gravity than any SUV and you do not see them rolling over at
anywhere near the rate of SUVs. This is because of better driver
training. You of course do see semis rolled over, but the factor in the
majority of those cases was not the higher COG, but rather the braking
limitations of an 80,000# vehicle that has a pivot point 1/3 of the way
down it's length.

Pete C.

Sparky Spartacus wrote:
>
> wrote:
>
> > David E. Powell wrote:
> >
> >>Hi there.
> >>
> >>I have been looking at the new Honda Civic, it's pretty sweet, and the
> >>welds and everything are as nice as anything I have ever seen. I just
> >>have a question about the "Drive by wire" system that they are supposed
> >>to have.
> >>
> >>What happens if the Engine dies on you? In my current car, my timing
> >>chip went once and the engine went out. I had enough steering control
> >>left, even without power steering, to pull my car over before it came
> >>to a stop. If the drive by wire system has no "real" or active
> >>connection, how can it work if the engine or electronics quit on you?
> >>Are there any backups built into the system in case any of that stuff
> >>happens? And what if your battery dies and you need to push the car?
> >>Can you turn the steering wheel to adjust your wheels when you push the
> >>car?
> >>
> >>Much thanks, sorry to bother.
> >>
> >>David
> >
> >
> > ++++++++++++++++++++++++++
> > Tuesday night we had some interesting weather.
> >
> > Real heavy lightening.
> >
> > A friend was comming back from Hazard on the Mt. Parkway
> > and had a nearby lightening strike. His engine car died and
> > he was able to pull offto the side safely.
> >
> > It wouldn't restart. The engine wouldn't even crank.The EMP from
> > the nearby lightning strike killed the ECM, igniton system, alternator
> > diodes and regulator, fuel pump and the fancy after market radio/sound
> >
> > system. It was a 2003 Civic.
> >
> > I don't want to think what would happen if it was a "die by wire"
> > system.
>
> Do you wear seat belts or do you worry about being trapped in your
> vehicle in the 0.0001% of the time that's an issue?

Bad comparison, seat belts do not have the ability to cause accidents
like a failing drive-by-wire or ABS system or airbag can.

Pete C.

notbob wrote:
>
> On 2005-11-18, Pete C. > wrote:
>
> > the term drive-by-wire to electronic throttle control which has been
> > around in the heavy diesel world but not hyped as "drive-by-wire"
> > confuses people.
>
> It's old news in cars, too. I recall being completely amazed upon
> reading the Helm manual for my girlfriend's 2 yr old '87 Cad De Ville
> and learning I could access and read and manually manipulate all the
> car's control voltages from the environmental control display. And
> sure enough, the throttle valve opening was a 0-10 volt range from all
> the way closed to all the way open. My first look at the wide world
> of car computers! Quite the revelation for me at the time.
>
> nb

I think you're referring to the Throttle Position Sensor, a.k.a. TPS,
not electronic throttle. Engine computers have had sensors to monitor
throttle position for years, at least since the advent of fuel
injection. This has nothing to do with electronic throttle control where
the computer actually has control of the throttle position.

Pete C.

Steve H wrote:
>
> I was taught this once, but like other stuff we don't use, you forget it go
> in a general since, it works like this:
>
> In the GM throttle body, it has 3 electronic signals to compare to, one
> being a mirror of itself. If it looses two of them then there may be an
> issue (I believe) If you were to have an catastrophic failure, the car goes
> to limp home mode and gives you a crawl speed.
>

I've not seen any redundant sensors in any GM vehicle I've worked on. My
current '97 truck with the 7.4l Vortec V8 certainly has no redundancy in
it's sensors. A single sensor each for throttle position, intake air
temperature, mass air flow, etc. Certainly if it looses one of the
sensors to the extent it can detect it, it will enter limp mode, but
absent redundant sensors, there are failure modes that the computer has
no way to detect.

Pete C.

"Sparky Spartacus" > wrote in message
...

> > That quote from Billy-Bob has nothing to do with advances
> > in design utilizing differing technologies. That only has
> > to do with capacity - he was basing his opinion because
> > of the current state of capacity. The technology hasnt
> > changed in computers,
>
> Wow, have you told Intel, AMD, HDD manufacturers, etc., etc.?

America, and indeed the world, pursue what is seen to be cutting edge
technology
just like codfish rush to bite an unbaited hook. Have things REALLY
improved
by quantum steps?

Software capabilities are not so greatly changed, and the chip technology -
though greatly evolved - has developed solely to service the software
which, indeed, has become bloated and glitchy.

You could run word processors, databases, spreadsheets, games, etc even on
the old
black and white Z80 machines. One company where I used to work ran the
whole operation
with two 10 megabyte harddrives and a Z80 network system.

Personal computers today do little that the old ones wouldn't do in some
form or
the other. Nor do they always do the job so terribly much quicker or
better,
although the microprocessors grunt along at multigigahertz speeds. We
garbage
mongers that feed the data into them are, oft as not, the limiting factor.

Mainframes had somewhat different requirements. They didnt have to cater to
the
executive gamer showoff computer-illiterate.

On Sat, 19 Nov 2005 01:16:33 -0600 notbob > wrote:

> On 2005-11-18, Don Bruder > wrote:
>
> > The F-16 - Proof that even a brick will fly if you can cram a big enough
> > engine into it...
>
> Tell it to a bumblebee.

Why - your mention of a urban legend is just that: legend...
http://quest.arc.nasa.gov/people/jou...bumblebee.html

--
remove MYSHOES to email

On Sat, 19 Nov 2005 07:24:47 -0500 Sparky Spartacus > wrote:

> > That quote from Billy-Bob has nothing to do with advances
> > in design utilizing differing technologies. That only has
> > to do with capacity - he was basing his opinion because
> > of the current state of capacity. The technology hasnt
> > changed in computers,
>
> Wow, have you told Intel, AMD, HDD manufacturers, etc., etc.?

So tell us what innovations have happened with processors
and hard drives? They've made processors speedier, and maybe
added more to the instructions set, or have increased capacity
from 32-bit to 64-bit, and have made hard drives with more
CAPACITY at lower cost to the consumer.

Yes, we have new drive interfaces, such as SATA, but that is
merely a higher rate interface that moves data at a HIGHER
CAPACITY.

My argument still stands - there has not been any true
innovation to computer hardware/peripheral components, they
have only gotten speedier moving those 0's/1's around and
they move MORE ( CAPACITY) 0's/1's than predecessors.

--
remove MYSHOES to email