Slashdot Log In
Inside the DARPA-esque Singapore Military Bot Contest
Posted by
Soulskill
on Sat Sep 20, 2008 01:05 AM
from the running-into-walls-on-the-government's-dime dept.
from the running-into-walls-on-the-government's-dime dept.
mattnyc99 writes "Earlier this summer we followed a war robot contest in England. But now, after the Russian onslaught in Georgia, this weekend's TechX Challenge in Singapore takes on a bigger meaning: can small countries keep up with military superpowers by upmodding existing robots for their own needs and then arming them? Researchers in the Far East seem to be struggling with their A.I. research right now, but this could just be the beginning of the 'little guys' fighting back. From the article: 'Chan says the agency wants to use more locally developed robots to help in homeland security and counterterrorist operations. The DSTA's goal is to improve robotic artificial intelligence so it can build machines to perform dangerous tasks — reconnaissance, surveillance and the handling of hazardous materials — that American robots already can. ... Back at Nanyang Technological University, Michael Lau acknowledges the urgency of the research but says the AI for urban warfare just isn't ready. "We don't really believe fully autonomous robots are possible yet," says the Evolution team supervisor. "How does a robot differentiate between friend and foe?"'"
We've discussed similar projects from DARPA in the past. Reader coondoggie notes that enthusiasts will be able to participate in the lighter side of robot warfare next month in Texas.
Related Stories
[+]
Science: DARPA Robot Contest Update 142 comments
rbrandis writes "DARPA has selected a wide variety of teams, after a series of last minute rule changes and a solid outpouring of anger, the final list of competitors for DARPA's Grand Challenge robot race has been set with 25 teams preparing to try and win a $1 million prize." The anger is exemplified by submissions like this one: Totally_Lost writes "Last spring we flocked to DARPA's Grand Challenge media event in Los Angeles to be told that they wanted everyone's participation in their Robot race this March. They told us that the race would be open to Mom and Pop garage sized participants - and Lied. This fall, nearly 100 teams completed technical paper submissions, with about half to be eliminated from the $1M prize race because they were too small to be 'real' competitors. Well, the rejected robot racing teams got together in Las Vegas last month, and formed the International Robot Racing Federation. This month IRRF is announcing its first competition with $1M in prizes pledged by sponsors, and lesser prizes too, to be offered in a REAL OPEN Challenge next September (providing the race that DARPA failed to deliver)."
[+]
DARPA Grand Challenge 3 127 comments
Meostro writes "DARPA announced the 3rd "Grand Challenge" today, The DARPA Urban Challenge. "To
succeed, vehicles must autonomously obey traffic laws while merging into moving traffic,
navigating traffic circles, negotiating busy intersections and avoiding obstacles." This year's new twist is two tracks for entry: the first is the same as the previous two challenges (develop on your own without Gov't. funding), but the second involves "submitting a detailed proposal for up to $1 million of technology development funds." Here is the PDF press release ."
[+]
Wall-E Lookalike Wins British War Robot Showdown 155 comments
longacre writes "Following in the footsteps of DARPA's Urban Challenge, in which robotic vehicles had to navigate a complex obstacle course without human intervention, the UK upped the ante with its own Ministry of Defence Grand Challenge: within a mock enemy village, robots were instructed to find potential targets and make distinctions between armed troops, roadside bombs and snipers. The winning entry, Team Stellar's SATURN system, actually consists of three vehicles: a low level drone and a tracked ground vehicle transmit reconnaissance data to a high-altitude robotic relay aircraft, which proceeds to phone that data home to a central processing center. Upon announcing the winner yesterday, MoD said they are 'carefully considering if technologies demonstrated in the final can be incorporated into future frontline kit for the Armed Forces. It is possible that the winning team will have invented a product that can be developed rapidly for the front line.'"
This discussion has been archived.
No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
Full
Abbreviated
Hidden
Loading... please wait.
Re: (Score:1)
wait, slashdot eats pants? i thought it w...
uhm..
did i say that out loud?
Russian onslaught in Georgia? (Score:4, Insightful)
Georgian onslaught in Ossetia?
Re: (Score:1)
Yeah, but "big scary bear" is catchier to the press.
Re: (Score:1)
In Other DARPA Words ... (Score:2)
Not Outside.
Re:In Other DARPA Words ... (Score:5, Informative)
I would argue that yes, robots, in the form of UAVs and UCAVs can help level the playing field. It all comes down to money. A MiG-29 Fulcrum might cost $25 million and a Su-27 Flanker costs $35 million. A Predator UCAV (Unmanned Combat Air Vehicle) costs a mere $8 million and can be armed with laser-guided bombs, Hellfire air-to-ground missiles, and there are tests underway to outfit them with Stinger air-to-air missiles. Now, a Predator might not be as effective as a MiG-29 or a Su-27, but you can buy 3 or 4 for the same price, and take risks with them since you don't have to worry about losing the pilot. Keep in mind that training a pilot for a fighter or attack aircraft takes millions of dollars and years of time, so they are not easy to replace; and their capture creates all sorts of political headaches. A handful of Predator drones (or their equivalents) might not win a war, but it would force your enemy to reconsider their plans a bit. Of course, a U.S. F-22 costs $137 million, and the F-35 costs $83 million... so the leveling of the playing field affects the US as much, if not more, than the Russians.
As for ground-based robots, I doubt they would be very useful for a small nation like Georgia. Again, it comes down to money. Say a robot costs 1 million dollars. For that price you might be able to get 1,0000 M-16s or 5,000 AK-47s and hand them out to civilians to form a militia. Any one of them will be smarter, more mobile, more adaptible, and more stealthy than your robot. One gets shot, have the next pick up his rifle. If you're being invaded by your larger neighbor, there probably won't be a shortage of young, testosterone-fueled kids with more balls than brains who are willing to defend the homeland. One man suitably motivated by nationalism or religion, and willing to risk/sacrifice his life to take out his target, can take out millions of dollars worth of high-tech gadgetry, as we've seen in Iraq and Vietnam.
Parent
Re: (Score:1)
I would argue that yes, robots, in the form of UAVs and UCAVs can help level the playing field. It all comes down to money.
Actually, the United States already has such a weapon. It's called a printing press and it can shoot FRNs halfway around the world.
Re: (Score:1, Troll)
Actually, the Georgians were using robots- they were using UAVs (Unmanned Aerial Vehicles) for reconnaissance. They did lose at least one to the Russians, but UAVS are inexpensive and importantly, the loss of the UAV does not result in the loss of the trained personnel that operate it. The Russians, on the other hand, didn't have any UAVs, and instead were using Tu-22 Backfire bombers for reconnaissance, which resulted in one of the planes being taken down by Georgian air defense.
And they lost zero man hours, too? Are you Hereby declaring "intellectual property rights" on that bottle rocket that was just launched, and recovered by the enemy?
Who's the Fucking Iron Eagle, here? Sir, Doug Masters, Bitch Lieutenant, Sire.
It's obvious they both want to touch (how do you say from russian to german to american "touch my Wienerschnitzel.com") base.
Ooooh $35 "million". What was that, like the original FDIC $100,000 insurance limit, pre-inflation?
Keep in mind that training a pilot for a fighter or attack aircraft takes millions of dollars and years of time, so they are not easy to replace; and their capture creates all sorts of political headaches.
Tell that to "The British Versus the Zulus,",
Re: (Score:1)
One man suitably motivated by nationalism or religion, and willing to risk/sacrifice his life to take out his target, can take out millions of dollars worth of high-tech gadgetry, as we've seen in Iraq and Vietnam.
Which is why it makes me angry when people in free societies tell everyone that those societies are to blame for all that is wrong with the world and all war is wrong.
Re: (Score:2)
Just wait for the far cheaper Chinese knock-offs
The Empire strikes again?! (Score:1)
Have those developers already received overtures to run the robots on embedded Windows?
P.S. Free Xboxes for the development environment, wheeee
Astromech? (Score:2, Funny)
Re: (Score:2)
There's no such thing as "artificial" intelligence. Duh.
Police enforcement is about to get a lot easier (Score:1)
Killer robots (Score:2)
We need to make sure that someone can be held personally responsible for the actions of the killer robots. Political leaders, preferably, but realistically it's more likely to be line military personnel. Whatever. If the robot screws up and kills the wrong people, we need to hold its operators responsible for war crimes.
Re: (Score:1)
Re: (Score:3, Informative)
Really, you can establish any number of rules and regulations to try to hold someone accountable if shit happens, but how many lawyers will be able to convice the jury that it was a "computer bug", something not forse
Re: (Score:2)
Do you really expect a country that refused to apologise when one of it's warships entered another countries territorial waters then shot down a civilian airliner killing everyone on board, would provide evidence on it's own soldiers.
To be fair, there's probably not many countries that
Re: (Score:1)
The pattern we have always seen is new technology always leads to new ways to excuse the killing of civilians and other inhumane acts. If a man with a sword sticks it through an innocent kid, everyone would agree a war crime has been committed, if a man with a gun shoots the kid, some people will be willing to accept it's an accident, the kid was caught in the crossfire, stray bullet, if the man drops a bomb on the kid from a plane, it's just collateral damage, unavoidable, certainly not the pilots fault. If he does it from hundred of miles away with a robot, fuck knows, he'll probably get a medal or something.
If someone sticks a sword through a kid, the people that find out about it would think it's a war crime. Considering the way information spread, or rather, didn't, when swords were common, that wasn't very many. Vastly more people currently consider killing anyone in a war a war-crime.
If a kid is caught in the crossfire, it is an accident. Nobody's going to target anyone but the enemy while under fire. Bombing civilians is a high-up decision, and is unrelated to a single soldier that goes on a killing-spree
Re: (Score:2)
We need to make sure that someone can be held personally responsible for the actions of the killer robots. Political leaders, preferably, but realistically it's more likely to be line military personnel.
The people who have the technology to build any type of advanced A.I. robots won't be subject to any war crimes. The winners in these types of military/political/economic battles never are.
No! (Score:1)
I don't welcome any robot overlords. The Singularity is growing more real by the day.
Re: (Score:2)
Overly constrained (Score:2)
My, my, aren't we picky. Well, if you're really arming most kinds of robots, because they can't see they can't differentiate, thus it's not relevant to them.
Re: (Score:2)
It doesn't really matter if they can see or not because all soldiers look pretty much alike (green in the countryside, brown in the desert, and white in the arctic). You basically just have to know which way to shoot.
Who's surprised? (Score:2)
It's hard to build a robot to discern friend from foe.
Where's the news?
Re: (Score:1)
It's hard to build a robot to discern friend from foe.
Look in the telephone directory. Simple!!!
Re: (Score:2)
And if Kansas decided it was going to secede from the United States, do you honestly think they would go a week without federal troops (from different states, don't make the "Tiananmen Square" mistake) walking the streets?
As I understand it, North Ossetia (part of Russia!) and South Ossetia were once one country. North Ossetia is technically part of Russia (as South Ossetia is technically part of Georgia). I wonder how things would have went if North Ossetia declared independence from Russia?
Oh, I'm sure Ru
Re: (Score:2)
Oh, I'm sure Russia would have just let the North Ossetians have their land back.
Parent poster has a point. You must of missed this old news: "Russian parliament votes to recognise independent Ossetia"
"The Russian parliament has voted unanimously to recognise Georgiaâ(TM)s breakaway regions as independent, in a move that will increase tensions with the US and other Western nations."
http://www.telegraph.co.uk/news/worldnews/europe/russia/2618728/Georgia-conflict-Russian-parliament-votes-to-recognise-independent-Ossetia.html [telegraph.co.uk]
I did not see it reported in the US as well, but no surprise
Re: (Score:2)
You'd have to be pretty nuts to think that Russia actually intends to let Ossetia keep the land.
It's definitely NOT about reuniting Ossetia. It's either about:
a) Screwing Georgia out of land
b) Be friendly with Ossetia for economical reasons (don't important pipelines go through Ossetia?)
c) Reunite them and then reintegrate it into Russia
Nukes, anyone? (Score:2)
No leveling of the playing field (Score:2)
As always, I don't really see how they might. Have firearms leveled the playing field between superpowers and the others ? Superpowers will probably have the most effective, most scary, most immoral war robots, while the smaller countries will either buy sub-par export models from them or try to mod their roombas.
Re: (Score:2)
also, the headline is bullshit.
russia and georgia went into war with mostly the same old soviet armament (georgia even had some western made weaponry and some western upgrades for their jet bombers). also georgia had combat drones, while russia had to use old strategic bombers as reconnaissance aircraft.
still, after only two days georgians fled and left most of their weaponry (both soviet and western) behind.
Friend or Foe? (Score:2)
> "We don't really believe fully autonomous robots are possible yet," says the Evolution team supervisor. "How does a robot differentiate between friend and foe?"
Same way the USA does. That wont take much AI. How many lines of code are there in "shoot first, ask questions later?"
Re: (Score:2)
You are a moron if you have never seen the UAV/Apache videos.
The Apache pilots go through this thorough process of identifying the target, whether they have weapons or have been firing mortars.
Sure, they check they have weapons. Not much good when they're shooting the hell out of friendly forces though, is it? Take a look at this link [youtube.com]. Of course it doesn't matter much when the US refuses to cooperate with the inquiry anyway.
To answer the question... (Score:1)
can small countries keep up with military superpowers by upmodding existing robots for their own needs and then arming them?
no, they can't. Which is why Georgia got hammered, and every small country that isn't as dumb as Georgia, is going for good ol' unconventional warfare.
My friend is on it. (Score:1)
AI controlled kill-bots (Score:2)
Yes what a brilliant idea, just imagine if lots of small countries had AI controlled kill bots, this would clearly help against the tanks and kill bots of their larger foes.
Seriously have we learned nothing from our education^H^H^H^H^H^H^H^H movies? AI controlled kill bots have only one true enemy....
US.
Semi-seriously though, given the fragile state of AI and the issues we already have with soldiers making bad decisions is it really smart to start delegating the kill/not decision to robots?