OT: DARPA Robot Car Challenge

[Terry Pinnell]

Pretty disappointing results from that Million Dollar Grand Challenge.

Report here: http://www.theregister.co.uk/content/28/36234.html
Results: http://www.grandchallenge.org/statusboard/

Of the 142 mile course across the Mojave, one car apparently managed 7
miles, another 6, a few 1, and all the rest zero.

Pity - I'd been looking forward to a few hours of exciting TV
highlights ;-(


--
Terry Pinnell
Hobbyist, West Sussex, UK

 

[Paul Burridge]

On Thu, 18 Mar 2004 10:23:37 +0000, Terry Pinnell
<terrypinDELETE@dial.pipex.com> wrote:

Pretty disappointing results from that Million Dollar Grand Challenge.

Report here: http://www.theregister.co.uk/content/28/36234.html
Results: http://www.grandchallenge.org/statusboard/

Of the 142 mile course across the Mojave, one car apparently managed 7
miles, another 6, a few 1, and all the rest zero.

If they're only offering a lousy million bucks, then it's hardly
surprising. No serious team's going to emerge for that kind of dough.
Pay peanuts, get monkeys.
--

The BBC: Licensed at public expense to spread lies.

 

[Winfield Hill]

Paul Burridge wrote...

If they're only offering a lousy million bucks, then it's hardly
surprising. No serious team's going to emerge for that kind of
dough. Pay peanuts, get monkeys.

The top teams were very serious, and spent many millions each on
their entries. Furthermore, you can be sure many will be back
next time with more advanced vehicles. The challenge was in fact
very difficult. Some commented they weren't sure they could drive
the course themselves.

The prize would likely not simply have been the $1M, but landing
a good position for a piece of DOD's future spending in this area,
which will be considerable, given the congressional mandate for
one-third of ground combat vehicles to operate unmanned by 2015.

Thanks,
- Win

whill_at_picovolt-dot-com

 

[Joe McElvenney]

Hi,

There's a fair article in this months 'Scientific
American' (came today) on the subject if you are
interested. A browsing session in W.H. Smiths might be
worth the trip


Cheers - Joe

 

[Roger Gt]

"Winfield Hill" wrote
: Paul Burridge wrote...
: >
: > If they're only offering a lousy million bucks, then it's
hardly
: > surprising. No serious team's going to emerge for that kind of
: > dough. Pay peanuts, get monkeys.
:
: The top teams were very serious, and spent many millions each
on
: their entries. Furthermore, you can be sure many will be back
: next time with more advanced vehicles. The challenge was in
fact
: very difficult. Some commented they weren't sure they could
drive
: the course themselves.
:
: The prize would likely not simply have been the $1M, but
landing
: a good position for a piece of DOD's future spending in this
area,
: which will be considerable, given the congressional mandate for
: one-third of ground combat vehicles to operate unmanned by
2015.

The requirement is similar to the California for the number of
Electric Cars to be in use by 2004. First you have to design one
that really works, is affordable, and is practical!

However the procurement people are not acting serious about a
definition of the task. I have been trying to get a contract to
design to requirements for over seven years, there is no
unambiguous requirements and there is no money to pay for a
serious effort. I made a partial unsolicited proposal they said
they weren't interested, but also wanted to buy one key element of
my proposal and could not understand why it could not be tested
without the other elements involved. They seem to think a
simulation is good enough!
Then my proposal, slightly edited to remove my identification, was
circulated for bids.
When I contacted the department they claimed to have written the
entire document "before" receiving my submittal. Clowns and fools
in the procurement process make for a lousy and expensive circus!

Obvious I didn't have the resources to enter the contest, and
would not since there was no guarantee they wouldn't take what
they liked from each system and let the contract to their friends.

The last time I submitted something of value, they confiscated it,
classified the patent disclosure "TOP SECRET" and forbid me to
claim credit of discuss the invention with anyone.

It a moot point now, my team has scattered and two have died,
there is no possibility we could start up even if we could get a
PO or grant!

 

[Glenn Ashmore]

That's nothing new. Bantam Cars won the competition for the Jeep and
the contract to build them was given to Willis and Ford. No royalties
or anything. When Bantam raised a stink they got to build trailers for
the jeeps they designed.

Unless you have really good political connections, never trust the DOD
with a good idea.

Roger Gt wrote:

However the procurement people are not acting serious about a
definition of the task. I have been trying to get a contract to
design to requirements for over seven years, there is no
unambiguous requirements and there is no money to pay for a
serious effort. I made a partial unsolicited proposal they said
they weren't interested, but also wanted to buy one key element of
my proposal and could not understand why it could not be tested
without the other elements involved. They seem to think a
simulation is good enough!
Then my proposal, slightly edited to remove my identification, was
circulated for bids.
When I contacted the department they claimed to have written the
entire document "before" receiving my submittal. Clowns and fools
in the procurement process make for a lousy and expensive circus!

Obvious I didn't have the resources to enter the contest, and
would not since there was no guarantee they wouldn't take what
they liked from each system and let the contract to their friends.

The last time I submitted something of value, they confiscated it,
classified the patent disclosure "TOP SECRET" and forbid me to
claim credit of discuss the invention with anyone.

It a moot point now, my team has scattered and two have died,
there is no possibility we could start up even if we could get a
PO or grant!

--
Glenn Ashmore

I'm building a 45' cutter in strip/composite. Watch my progress (or lack
there of) at: http://www.rutuonline.com
Shameless Commercial Division: http://www.spade-anchor-us.com

 

[John Jardine]

Winfield Hill <Winfield_member@newsguy.com> wrote in message
news:c3c5uc01mln@drn.newsguy.com...

Paul Burridge wrote...

If they're only offering a lousy million bucks, then it's hardly
surprising. No serious team's going to emerge for that kind of
dough. Pay peanuts, get monkeys.

The top teams were very serious, and spent many millions each on
their entries. Furthermore, you can be sure many will be back
next time with more advanced vehicles. The challenge was in fact
very difficult. Some commented they weren't sure they could drive
the course themselves.

The prize would likely not simply have been the $1M, but landing
a good position for a piece of DOD's future spending in this area,
which will be considerable, given the congressional mandate for
one-third of ground combat vehicles to operate unmanned by 2015.

Thanks,
- Win

whill_at_picovolt-dot-com

Gosh!. Wish I had their uncritical faith in modern technology. I can't even
trust my security lights to ignore the local wildlife. Never mind imagining
some gun-toting, microsoft controlled, terminator unit.
regards
john

 

[Glenn Ashmore]

John Jardine wrote:

Gosh!. Wish I had their uncritical faith in modern technology. I
can't even
trust my security lights to ignore the local wildlife. Never mind imagining
some gun-toting, microsoft controlled, terminator unit.
regards
john

That was my first thought. The opening scene from Terminator I.

--
Glenn Ashmore

I'm building a 45' cutter in strip/composite. Watch my progress (or lack
there of) at: http://www.rutuonline.com
Shameless Commercial Division: http://www.spade-anchor-us.com

 

[Ken Smith]

In article <c3dnts01qpp@enews3.newsguy.com>,
Chris Carlen <crobc@BOGUS_FIELD.earthlink.net> wrote:

I think it is great to have this competition. But I don't think it is a
viable military plan to use autonomous robotic vehicles.

Perhaps the vehicle isn't really the main thrust of the project. for a
vehicle to navigate through fences and trees etc, it needs a vision system
that can figure out what is out there and logic that can find a path
through the clutter. If you can do that fast enough, you can make a
self guided weapon that would go in the right window, down the right hall
and blow up the right general without havinf to have detailed plans of the
building to program it with. Decapitation of the other side is a quick
way to win a war.


--
--
kensmith@rahul.net forging knowledge

 

[Mike]

On Thu, 18 Mar 2004 19:03:24 -0800, Chris Carlen wrote:

Winfield Hill wrote:
Paul Burridge wrote...

If they're only offering a lousy million bucks, then it's hardly
surprising. No serious team's going to emerge for that kind of
dough. Pay peanuts, get monkeys.


The top teams were very serious, and spent many millions each on
their entries. Furthermore, you can be sure many will be back
next time with more advanced vehicles. The challenge was in fact
very difficult. Some commented they weren't sure they could drive
the course themselves.

The prize would likely not simply have been the $1M, but landing
a good position for a piece of DOD's future spending in this area,
which will be considerable, given the congressional mandate for
one-third of ground combat vehicles to operate unmanned by 2015.

Thanks,
- Win


I find this idea of unmanned vehicles looney. The likelyhood of human
piloted vehicles (barring being attacked by an enemy) completing their
missions is very high, thanks to human intelligence. The likelyhood of
a robotic vehicle completing a mission will always be substantially
less, perhaps even really quite mediocre by 2015. It's also downright
dangerous to anyone in the path of the darned thing.

What I think makes much more sense, because it capitalizes on the
virtues of both the robotic as well as the human capabilities, is remote
controlled vehicles. This seems so obvious that I find the DOD plan
really quite disturbing.

I think it is great to have this competition. But I don't think it is a
viable military plan to use autonomous robotic vehicles.

Yeah, right. You're still pissed because they didn't hire you to drive the
cruise missile, aren't you?

-- Mike --

 

[Jeff Liebermann]

On Thu, 18 Mar 2004 10:23:37 +0000, Terry Pinnell
<terrypinDELETE@dial.pipex.com> wrote:

Pretty disappointing results from that Million Dollar Grand Challenge.

The first few solar powered racers in about 1975 were a total
disaster. The first autonomous robotic soccer matches were equally
defective. I vaguely recall that the first Robot Wars episode took
weeks to film due to multiple breakdowns. Each took a few years and
numerous mistakes to fix the problems, and give the participants a
better view of reality. Patience please.

If you've ever tried to design something totally new, without any
example to plagerize, er... borrow ideas from, or without any real
experience, you'll understand the problem. I once had to design
something with virgin and state-o-de-art technology. Nothing worked,
everything broke, and everyone was questioning our competence. The
2nd incantation was better, but far from perfect (or even useable).
It took 5 tries, 5 broken deadlines, and 5 cost over-runs, to get it
right, all of which involved discarding some bad assumptions and
learning the limitations of the technology. In other words, NOBODY
does it right the first time.

May I humbly suggest you withhold judgement, learn what you can from
what didn't work, and stay tuned for another attempt in September.
http://www.irrf.org

--
# Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
# 831.336.2558 voice http://www.LearnByDestroying.com
# jeffl@comix.santa-cruz.ca.us
# 831.421.6491 digital_pager jeffl@cruzio.com AE6KS

 

[Quack]

What I think makes much more sense, because it capitalizes on the
virtues of both the robotic as well as the human capabilities, is remote
controlled vehicles. This seems so obvious that I find the DOD plan
really quite disturbing.

Okay so if you were to use remote control cars, and the enemy 'jammed'
your transmissions, do you want your cars to sit there stupidly doing
nothing ?
Maybe this is a backup system for remote controlled vehicles.

Not as good as a human/remote control driver sure, but perhaps better
than nothing at all.

Alex.

 

[Daniel Haude]

On Thu, 18 Mar 2004 19:03:24 -0800,
Chris Carlen <crobc@BOGUS_FIELD.earthlink.net> wrote
in Msg. <c3dnts01qpp@enews3.newsguy.com>

I find this idea of unmanned vehicles looney. The likelyhood of human
piloted vehicles (barring being attacked by an enemy) completing their
missions is very high, thanks to human intelligence.

But then, even current military operations don't appear to be guided by
too much of human intelligence, either.

--Daniel

--
"With me is nothing wrong! And with you?" (from r.a.m.p)

 

[Terry Pinnell]

Chris Carlen <crobc@BOGUS_FIELD.earthlink.net> wrote:

I find this idea of unmanned vehicles looney. The likelyhood of human
piloted vehicles (barring being attacked by an enemy) completing their
missions is very high, thanks to human intelligence. The likelyhood of
a robotic vehicle completing a mission will always be substantially
less, perhaps even really quite mediocre by 2015. It's also downright
dangerous to anyone in the path of the darned thing.

What I think makes much more sense, because it capitalizes on the
virtues of both the robotic as well as the human capabilities, is remote
controlled vehicles. This seems so obvious that I find the DOD plan
really quite disturbing.

I think it is great to have this competition. But I don't think it is a
viable military plan to use autonomous robotic vehicles.

I assume that both limited autonomy and R/C are planned. For 'routine'
stuff like staying on a compass bearing, avoiding a lake, skirting a
wall, etc, the thing could be left largely to itself. For cleverer
stuff, like distinguishing bear from man, shooting the right target,
etc, accepting verbal help from locals, that would be via R/C
interfaces.

Obviously, they're still a long way from getting the routine
self-contained technology right.

--
Terry Pinnell
Hobbyist, West Sussex, UK

 

[Guy Macon]

Chris Carlen <crobc@BOGUS_FIELD.earthlink.net> says...

What I think makes much more sense, because it capitalizes on the
virtues of both the robotic as well as the human capabilities, is remote
controlled vehicles. This seems so obvious that I find the DOD plan
really quite disturbing.

How do you communicate with the remotes? Radio? Jamming and
anti-radiation missiles. Lasers? Smoke. Fiber optics?
Fletchettes.


--
Guy Macon, Electronics Engineer & Project Manager for hire.
Remember Doc Brown from the _Back to the Future_ movies? Do you
have an "impossible" engineering project that only someone like
Doc Brown can solve? My resume is at http://www.guymacon.com/

 

[Fred Bloggs]

Glenn Ashmore wrote:

That's nothing new. Bantam Cars won the competition for the Jeep and
the contract to build them was given to Willis and Ford. No royalties
or anything. When Bantam raised a stink they got to build trailers for
the jeeps they designed.

Unless you have really good political connections, never trust the DOD
with a good idea.

This is not true anymore- just create a certified paper trail prior to
submittal- and you can bring the whole department to its knees. Rumsfeld
wants to fire the *GOOD FOR F__K__G NOTHING* "acquisition workforce"
altogether. Their performance record is the gold standard of being the
absolute worst- cannot go anywhere but up by eliminating the worthless,
ignorant, incompetent, corrupt, unproductive, overpaid, bureaucratic
SCUM! These are the vermin who brought you the Commanche fiasco- twenty
years and as many billions and you have absolutely nothing, Rumsfeld
was so eager to shed the boondoggle, he is cutting his losses and buying
$2B in contract termination costs. The "acquisition workforce" is a
*prime* Rumsfeld target to be eliminated in 2005 base realignment and
closure, and this is why it's important that Bush be re-elected-
Rumsfeld is going to send them packing as well as eliminate their legacy
of pathetically overcomplicated, counter-productive, bureaucracy
expanding acquisition regulations- all will be gone, and good riddance.

 

[Daniel Haude]

On Fri, 19 Mar 2004 10:25:55 +0000,
Terry Pinnell <terrypin@dial.pipex.com> wrote
in Msg. <eail50d9l0k4t95prbcfn5dgbd187309lj@4ax.com>

wall, etc, the thing could be left largely to itself. For cleverer
stuff, like distinguishing bear from man, shooting the right target,
etc, accepting verbal help from locals, that would be via R/C
interfaces.

Yeah, I bet the locals area really keen on getting barked at
by robots.

--Daniel

--
"With me is nothing wrong! And with you?" (from r.a.m.p)

 

[Sir Charles W. Shults III]

Well said.

Cheers!

Chip Shults

 

[Daniel Haude]

On Fri, 19 Mar 2004 16:07:03 +0000,
Terry Pinnell <terrypin@dial.pipex.com> wrote
in Msg. <pd6m50h5io7oen10gtr3p4pu0fu1883rip@4ax.com>

I had in mind a simple two-way intercom type of thing (sound and
video), with humans at both ends. No more a robot than the car park
attendant you speak with before he lifts the swing-barrier.

Except that this thing is in all likelihood armed and being used in an
invasion of the country whose locals are addressed through that intercom.
Just how long do you expect this contraption to last?

--Daniel

--
"With me is nothing wrong! And with you?" (from r.a.m.p)

 

[John Woodgate]

I read in sci.electronics.design that Jeff Liebermann
<jeffl@comix.santa-cruz.ca.us> wrote (in <hd8l50db9nhinpkq7jrd8cq36sr46f
05an@4ax.com&gt about 'OT: DARPA Robot Car Challenge', on Fri, 19 Mar
2004:

NOBODY does it right the
first time.

Yes, what works well (for us, if not dodos) is the Universe; NOBODY did
that right the first time.
--
Regards, John Woodgate, OOO - Own Opinions Only.
The good news is that nothing is compulsory.
The bad news is that everything is prohibited.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk