The European Grand Challenge

An anonymous reader writes "A European version of the DARPA Grand Challenge is being held in Germany next month. Instead of a race through the desert, the EU challenge is split into three events. Urban, non-urban, and landmine detection will be the 'courses', with multiple winners in each event. Interestingly Sebastian Thrun, winner of last year's Challenge, has been forbidden from taking part despite being a European citizen." From the article: "The trials will take place in and around Hammelburg, a mockup of a town used by the German military for training exercises. In the non-urban course the robots will have to contend with a one-kilometer route containing ditches, barbed wire fences, cattle guards, fires, narrow underpasses, and inclines of up to 40 degrees. The urban and landmine 500-meter trials will require the robots to negotiate doorways, stairs, partially collapsed buildings, and poor visibility from smoke or partial lighting. Along the way, they will also have to search for designated objects and report their findings back to base."

Re:COCKfuckers!

[LeftOfCentre]

Actually there is in fact such a thing as a European citizenship, or more specifically EU citizenship: http://europa.eu.int/comm/justice_home/fsj/citizen ship/fsj_citizenship_intro_en.htm [eu.int]

Re:Landmine Detection a Good Thing

[Buran]

That wasn't the point of the Grand Challenge, though. The GC was designed to help develop autonomous vehicles for things like long-distance supply missions. Your friend's complaint is like complaining that a digital SLR camera can't record movies even though it's good at what it was designed for -- still image capture. It can't capture movies because its design makes that physically impossible since it wasn't meant for movie capture. Similarly, the GC robots weren't designed for clearing minefields and no one ever said they would be useful for that.

Apples and oranges.

It's not as progressive as i'd like

[Xiph]

While like both the darpa and the euro challenges, I also find that they are more a showcase for solving tasks of yesterdays war, with newer and modern toys. So far in warfare, robots are used much like human operated vehicles were used in WWI, for reconnaissance purposes and the fact that they're spending so many resources on maneuvering excersises, is more of a showcase for their limited vision, than the capabilities of robot technology. When they make these big expensive robots for warfare, they forget the primary reason robots are used in the first place; robots are expendable, and partially autonomous.

The last part is where the european challenge at least gets something right. There's no need for fully autonomous vehicles on the battlefield, because the decisions you make on the battlefield require human accountability, when the situation is grave enough to throw away accountability, that's the time you launch the nukes.

So how do we make a robotic system that takes those two benefits into account?
My suggestion would be to use swarming, and standby robots. For instance, if I were to launch a robot air assault, with say 500 human controllers involved, i'd use standard hobbyshop vehicles, with advanced communication, some computing power and a weapon on each, keep it cheap, And i'd use somewhere along the line of 10,000 robots.
The robots can be dropped from a plane, or send off from the ground, the later option will be cheapest the former will have greater range.
The controllers will take control or partial control when they arrive, in early versions full control of a single plane, if there's no available controller, they'll go on standby somewhere close to the battlefield, When a robot goes down, they're allocated a new one from the pool of robots on standby.
In a more advanced scenario the robots would create a 3d representation of the battlefield and the controllers will just point out targets, and possibly hit the fire button (for accountability reasons).

But that's just one version, I think that a cool competition goal would be to design a system that can: Take out targets as fast as possible, as cheap as possible and as reliable as possible (reduce collateral damage), the targets can be anything from target dummies, over vehicles to other robots, in scenarios including regular, urban and guerilla warfare, police assignments and terrorist attacks.
The reason i stop here is that i don't have the vision to go further, not that others should not try to think beyond it.