Grand Challenge Results!
From the Stanford site, we have a DARPA Grand Challenge winner:
October 9, 2005 -- Today Stanley and the Stanford Racing Team were awarded 2 million dollars for being the first team to complete the 132 mile DARPA Grand Challenge course. Stanley finished in just under 6 hours 54 minutes and averaged over 19 miles per hours on the course.
The teams that finished, together with their times are:
| Team | Time |
| Stanford Racing Team | 6 hr 53 min |
| Red Team | 7 hr 4 min |
| Red Team too | 7 hr 14 min |
| Grey Team | 7 hr 30 min |
| Team Terramax | 12 hr 51 min |
The team my son was rooting for (and me too as his dad), Monstermoto, did not finish and was way down at the bottom. It was a really cool all-terrain vehicle with sinister-looking headlights and an appealing video scored with really cheerful music. Poor Monstermoto.
The Independent reacts with a mixture of wonder and dread at the race results.
Not so long ago, the notion of cars driving themselves seemed no more than science fiction. This weekend, it took a giant stride towards becoming reality as a handful of robotically controlled vehicles completed a 132-mile obstacle course around the Mojave desert. ... To some people, the development of robot fighters suggests a future not unlike the one imagined in the Terminator movies - where highly sophisticated machines are turned into merciless killers unburdened by considerations of conscience or their mortality.
posted by wretchardthecat at 10/09/2005 05:30:00 PM
16 Comments:
"fighters [as] merciless killers unburdened by considerations of conscience or their mortality."
Beslan?
Iskandaria?
Could the Independent's staff be any more clueless??
If I were a long-haul truck driver, I'd think about getting into a new business.
IT doubles in power and precision every year. The cars' processing power, sensors, and software will all follow Moore's Curves up that geometric progression.
The 18-wheeler drivers and Greyhound drivers will go first, but soon enough it will come for the taxi drivers. It sounds silly to think that these cars could navigate downtown traffic, but just give it a couple years. It'll get better and better, and it will never be distracted by talking on the cellphone, and it will never drink & drive.
One day, not as far into the future as you think, driving will be like riding horses or sailing - one of those hobbies some people do in their free time.
I just came back from a sci fi flick at my local theatre at the mall. The movie is called Serenity.
Great flick. Worth every bit of the A- on Yahoo. (Oh, the A- is the ticket buyers score--which is a more reliable score-- than the critics score.
There's some serious themes in the movie as well a lot of great action and adventure. And the characters are well drawn. They're not like the wooden egyptian sarcophogi drawings of the last three star wars but rather the whole quirky characters of the original three episodes.
Kurizweil has come up with a new book that seems to push forward the timeline for his singularity. The book is talked about in detail at slashdot.
DARPA's methodology for funding the robot cars imho is the most ingenious part of the race. DARPA offers a prize and a competition. I've seen GAO reports that put the amount of research that DARPA gets on their million dollar prize as something over 10 times what they would have gotten if they had simply paid to have it done. (And that includes administration. That is 1 million plus the cost of administration got DARPA 10 times as much research bang--as they would have got by just going out and paying some scientists to invent the robot cars.)
(In past years I've run this idea past water desalination people but so far without much luck.)
In a similiar vein GM said last thursday that their fuel cell car will be competitive with a gasoline car by 2010.
GM has been saying this once every couple months since last fall.
Buried in the above article are two interesting tidbits. The first is that.... "Currently, Burns said, the cost of producing the hydrogen for use in vehicles is 1.3 times the cost of producing gasoline for vehicle use."
Two things have happened recently to cut the cost differential. One is the recent tripling of the cost of gas and the other is a new procedure this year that cut the cost of hydrogen production in half.
The other interesting thing in the article is that GM has run the numbers for creating an initial infrastucture to support hydrogen.
"GM has done a study showing that building about 12,000 hydrogen filling stations throughout the country’s metropolitan areas would put a station within easy access of every driver of a fuel-cell vehicle. Also, a hydrogen station every 25 miles on the interstate could also be established.
Burns said GM estimated that would cost $12 billion."
nigelhuffy,
Don't underestimate the utility of slurpees. Americans are as productive as they are at least in part because they are happy; and they are happy at least in part because they have slurpees. It's not a coincidence that unlimited, free slurpees are one of the perks for working at the Googleplex.
The point of my story? You cannot know what really is and is not important to people, unless you let the market decide. McGriddles are important.
What's most remarkable here is that a few very inexpensive 'skunk works' efforts have been able to stitch together existing technologies and components to produce such enormous performance strides in only two years.
Only in America. This gives an indication of the stunning acceleration in the application of technology to warfare we would be likely to see if we ever got into another major war, just as we did during WWII.
In fact, if you take UAV's into account, robots are already flooding onto the battlefield in the GWOT. If America's productive resources are ever fully engaged in a conflict then even without the use of nuclear weapons the results would, after the first three years, most likely once again be utterly devastating for any likely foe - even on a global scale.
Battlefield robots also affect America's combat potential in strategically significant ways. Firstly they could only ever be effective when tightly and dynamically networked with other battlefield resources, in ways in which America is technically and organizationally a long way ahead. Secondly, as a force multiplier they exert a relatively small impact in a conflict in which we limit ourselves to taking minimal casualties, but an enormous potential impact in any future conflict to which we might become fully committed (i.e. willing to take casualties on a WWII scale).
The 'demonstration effect' of this sort of contest may also be deeply significant. If both Hitler and the Japanese had not so disastrously underestimated, out of pure ignorance, the overwhelming warmaking potential of the US economy then WWII might never have happened.
Cedarford,
I don't imagine combat robots will have to be used more than once, any more than Desert Storm had to happen more than once. After that, any potential asymmetric enemy looks for other ways to engage in combat. One possibility is that the enemy may develop IADs (Improvised Autonomous Devices) -- made with the latest AQ Khan technology -- crude robots effective against shopping malls and similar targets.
Santayana once said "only the dead have seen the end of war". We have yet to find a way to put a stop to men's hatreds. The means are new but the ends as ancient as Lucifer's.
What the hell do hamburgers and milkshakes have to do with creativity, morality, innovation, stamina or courage? What a snooty pseudo-intellectual way to put down a big bunch of people which has nothing to do with anything! Jerk probably thinks that French wines are the pinnacle proof of ... what? Something important, doubtless.
Cardozo, The long haul drivers would just shift over to regional drivers. At the moment long haul is classified as irregular route long haul, for an automated vehicle to be efficient it would have to compete with the railroad on the same types of routes that the rail follows today. The infrastructure doesn't exist to copy what the average trucker does every day.
As far as giving up time and money, there is a project that can be assisted at the present. The project is Einstein at Home and they're looking for gravity waves from extraterrestrial sources. Through the use of a program called BOINC they can use your system to become part of a supercomputer. There are other projects eg. SETI@home, a cryptography project and others.
I understand that the DARPA effort was spurred by direction from Congress that at least 30% of military ground vehicles be unmanned within 10 years.
And I guess that by "unmanned" they don't mean parked like the busses in New Orleans.
UAVs were so long in coming not due to technical reasons but rather the inability of a pilot-dominated US Air Force to accept the concept.
From the technical standpoint, air-launched precision-guided UAVs should have been ready in time for Korea; that they were not even ready for Vietnam is simply astonishing.
Arming of the Predator UAV occurred only because (1) The CIA was doing it anyway and (2) the USAF Chief of Staff told his people to get off their duffs and do it. This in a Service that is the most technology-oriented of them all - or at least should be.
The greatest leaps in military "technology" are not made in hardware but in concepts.
It takes an enormous effort for the Services to toss aside old ideas -or even to just accept how they must be changed.
Despite the Rise of the Machine fears of the Independent, my guess is that autonomous vehicles have largely a civilian future. It's hard to remember, helping your child research his homework that the Internet was developed for military purposes. And it's just as funny to realize, when out boating, that the GPS system which had its origins in the darkest places of the human heart takes fishermen home to the winking light on the distant shore.
I guess all real fire is like that. You intend it for burning out your enemies and wind up cooking your supper with it.
nigelhuffy:
Go into any American fast food restaurant and you will find it a model of efficiency, cleanliness, decency, respect, and comfort compared to the other 99.99999% of what human beings have to endure or for that matter, have created.
Utopia could be defined as the United Nations and most of the countries on Earth being run at least half as well as a typical McDonald's. At which I usually do not eat, by the way.
You would pose that the people who conceived and created the robotic vehicles do not dine at such places?
Absurd!
Hell, most of them probably used to work there.
Papa Bear,
Computers are notoriously much better and faster than humans at tedious and repetitive tasks. So I think that autonomous vehicles will mostly be used to reduce the tedium and boredom.
The University of Parma (where I graduated in Industrial Chemistry) already developed in the late '90s a car with a guidance system that can autonomously follow a road, thus relieving the stress of keeping hands on the wheel and eyes on the road during higway driving. And many high-end cars have cruise control, keeping a set speed and proper distance from other vehicles. The future is already coming to us.
Wretchard: Relative to your exmaple of GPS, I can tell you that for many years the GPS Joint Program Office had a sign near its entrance that said "The purpose of this SPO is to drop a bomb within 10 ft of the target."
They have done rather better than that since then, of course.
I wonder though, what it says about the recesses of the human heart that the most innovative and efective technology often comes as a result of the darker aspects.
Cedarford,
The "worth" of an American combat soldier's life is about 375-400,000 including past training & survivor benefits.
It's probably worth more than that, if you calculate the present value loss of a lost young man's life. You lose his entire future, although we can only put a dollar figure on his lost earnings.
But on the other hand the benefit of a soldier's sacrifice is equally incalculable. To the extent that he keeps others safe, for example you or me, how much loss is he preventing? Someone once wrote that a soldier gives us his future so that other people can have theirs.
Most probable application for the immediate future: open pit mining. This technique could be the breakthrough required in the extreme far north of Canada. There are some amazing mineral ores north of the artic circle. The local climate is just so severe. Technology which would allow extraction in the depth of winter might make such deposits viable.
FDR and his top brass suckered the Japanese and the Germans.
Hap Arnold released the ultra-secret Rainbow Five white paper detailing an extremely bleak strategic mobilization timetable via a classic Washington 'leak'. He did so at the specific direction of the President and the assembled top brass.
Each Tuesday morning FDR routinely met with his service chiefs and other specifically invited officers. It was the beginning of the National Security Council. (Truman's codified this ad-hoc committee in 1947.)
The grim calculations based on a top to bottom analysis of the US war making power pointed to America being completely unable to react in any meaningful way for at least 18 months -- with 30 months a more realistic figure.
It was this leak plus many other disinformation operations that convinced Hitler that he had effectively won in Russia already and that America was hopelessly behind.
With war he figured he could shut down Britian's economy with his U-boat fleet.
Actually, the 18 month targets were met in six months. War production went right through the roof as all of the previous thinking was conservative. War production always astounded the brass -- none of whom was an industrial production engineer.
Japan was provoked in every imaginable way because Hitler refused prior bait.
In retrospect it is clear that Adm King schemed up the naval engagement at Pearl Harbor. An air raid on Pearl was American doctrine for a decade. It was the fleet exercise in 1930.
The President's idea was brilliant and simple. Get the weakling to attack an empty bag.
All of the new capital ships were missing from Pearl Harbor. Most of the battlewagons were on skeleton crews. The Navy had welders fitting additional AA mounts along battleship row all Saturday night into Sunday morning. (!)
All that and the Coast Guard ordered all merchant shipping away from the path of the IJN fleet, a fleet that couldn't keep its mouth shut. Even if America couldn't decode the signals triangulation would have -- did -- reveal their location.
Japan was to have attacked even earlier, but had to reschedule because of the need to get fresh code books distributed through the fleet. Just that alone gave the game away, since that is, was and always will be standard operating procedure prior to a strategic attack.
What was not forseen was that Japan would deploy novel shallow running torpedoes, nor that she could actually get a mini-sub into the harbor. ( This last point was only realized and proven in the 1990's with close analysis of battle photos. The last missing sub was finally 'found')
What was not forseen was that the Arizona would suffer a critical hit. Over 50% of Pearl's losses are due to that one blow.
What was not forseen is that the high tech radar would be ignored.
Any posture much beyond what was actually done would have compromized the fact that JN25 was broken. That secret was worth thousands of lives.
Stalin figured out the gambit. He had spies everywhere.
Britain fully doped it out no later than Spring 1943. And that is when America figured out that Britain was trying to tempt Japan in just the same way.
Britain did all she could to convince the Japanese that she was not to be a naval player against Japan anymore. This so as to lower the barrier for Japanese aggression against the 'all alone' US Navy.
Post a Comment
<< Home