The Grand Challenge

Hello all you FIRSTers!

DARPA intends to conduct a challenge of autonomous ground vehicles (robots) between Los Angeles and Las Vegas in March of 2004. A cash award of $1 million will be granted to the team that fields the first vehicle to complete the designated route within a specified time limit. The purpose of the challenge is to leverage American ingenuity to accelerate the development of autonomous vehicle technologies that can be applied to military requirements. Many of the details of the event are being developed, and new information will be posted to their web site as soon as possible.

The website is: http://www.darpa.mil/grandchallenge

I am currently building a FIRST Coalition team from all those that may be interested to tackle this challenge, so if any of you are interested, please respond to this thread or email me at [email protected]. Remember…$1 million + Fame + Multi-million dollar contracts + Fun robot competition building and competition experience!

I’ve looked at the competition and it looks pretty cool.

They say that the course is travelable by a regular chevy pickup, so our robot doesn’t have to be insanely armored, and it shouldn’t be impossible. Also, it’s all autonomous, and even if your bot refuels itself it has to do it autonomously.

Sounds cool though, smart to get a team going of firsters.

Good luck!

A GPS controlled airplane might work. Hey, this thing gives me an idea for next years competition. The robots have to drive them selves to nationals and whatver ones get there can compete in the nationals which would be the challenge of driving back.

Unfortunately, this is a competition involving autonomous ground vehicles so the GPS airplane won’t work. O…If any of you are at the SoCal Regional, come to the station for team 1135 if you want more information.

*Originally posted by PsiMatt *
**O…If any of you are at the SoCal Regional, come to the station for team 1135 if you want more information. **
I’ll stop by.

I think you guys are underestimating the reality of this challenge.

Firstly, these courses are similar to something you’d see in professional dune buggy competition. Obviously, whatever you build is going to need to last. Even more-so than a FIRST robot. Simply, take the most robust FIRST robot you’ve seen and multiply it by 100.

Secondly, This is a challenge that is greater than 250 miles. They’re expecting 250 miles alone in a desert course. A typical car can only do 300 miles. Obviously you’re going to either need a very efficient engine, or you’re going to need a very efficient solar/electrical vehicle. You must also keep in mind that the time limit is 10 hours. That’s 30 miles per hour, even in the desert.

The programming and electrical engineering challenge of this task is astronomical. You need to design, build, and coordinate a guidance system of great proportions that can accurately depict hazards and correct its course. This fact alone would take someone with a PhD in either (or both) of these fields. GPS is accurate, but not that accurate. Let’s just say it’s the difference between being on the road and driving off a cliff, so you can’t rely strictly on that.

Needless to say, in the process of building the vehicle, major breakthroughs will most likely need to occur or will hapen not only in the programming aspects, but in the design itself. This is no easy challenge like FIRST. FIRST is a cakewalk compared to this. I don’t mean to be rude, but you simply won’t have the resources to complete a task like this. If you manage to go 10 miles in this competition, I myself will give you $1000.

For people thinking of critisizing me, don’t. Think, you only have a year to do this. Building a machine as complex of this is a large undertaking. There’s simply too much to learn before you can even begin to think about designing something like this. You have to learn Materials Science, Mechanical Engineering, Electrical Engineering, Physics, and Computer Science in less than a year. You have to explore all the possibilities. You’re better off designing stuff and waiting until the year after to do anything as massive as this. They said they’ll do the competition until 2007.

After briefly glancing at the rules, it seems to me that you’d want to retrofit an existing vehicle to complete this challenge. They say that it’s a course that could be accomodated by any commercial four wheel drive vehicle, so I’d start by acquiring one of those. Convert it to a gasoline/electric hybrid and throw in oodles of solar cells.

Install an onboard computer system that is capable of controlling the trucks acceleration, braking, and steering. Also interface it with the vehicles existing sensors so that mechanical problems can be detected.

For navigation, I’d expect that you’d need two-levels of systems to accomplish this task. The first, larger level would be global positioning system based and would be used to locate the vehicle in the most general sense. If the course is provided in advance, program the system to proceed through a series of waypoints along the route. If the course isn’t provided in advance, or if only their waypoints are provided with no information about road conditions, turns, etc., it’s much harder.
The second level of navigation would have to be used for local navigation. That is, you’d need a battery of sensors (maybe ground radar) that could detect and map out the surrounding terrain. The onboard computer would then need to extrapolate a route from that data based on where it is and where it needs to go, and then decide a way after accounting for ease to cross certain terrain versus time to cross other terrain.

Anyway, that’s where I’d start. It’s definitely not an easy undertaking and would require quite a bit of capital. But, if I were given the chance, I’d give it a shot.

In response to Jnadke above, I completely disagree. Sure it is great to be realistic, which you might be, but let us just have a chance. It gets us excited anyhow. There is nothing bad that can come from entering a competition which celebrates science and technology. Have we forgotten the meaning of FIRST? It is not about winning or losing, it is about what you learn in the experience and the good times you have while doing it. So what if the kids’ car goes 50 feet? What’s it to you? I bet from that project that made the 50 foot run they will learn something much more valuable than the winner would ever have learned. We all have dreams okay. It does no good for you or anyone else to inhibit them. So just let us race to Vegas alright?

And if you insist on talking about the compeition side of things, just look at our robot. We made it to the final match and gave Kingman a run for their money - they even say so. We even won the KISS award from Archer. Our robot was completely student designed and built except for 4 welds and 2 simple CNC parts. And, in Phoenix, we got 3rd seed, semifinalists, and we killed our oponents.

Now, for this DARPA compeition, so what if it takes a pro dune buggy? They are actually quite simple. I’m sure any group of veteran FIRSTers would have no trouble building one. Or what about a truck? Trucks are so simple - especially old ones. All it would take is a little money and a little inginuity. And professional help is not as hard to find as you may think. It is definitely do-able.

I think this is going to turn into that thing where you have to construct a passenger vehicle to enter space. people will spend millions of dollars trying to develop the vehicle for the million dollar prize. I do agree that it would be very cool to try given the money to do so.

Cory

Many times in the past, those experienced in a field have declared that something was impossible, only to have someone without those preconceptions do the impossible. What I’m asking is for a team of FIRSTers to look at this problem from their viewpoints, which may be different and/or contain different preconceptions from the typical industry person. :yikes: I don’t think that should come as much of a surprise, really, but remember that the motto of FIRST is “For Inspiration and Recognition of Science and Technology.” Its mission is to inspire others to excellence and innovation, and participating in this project would be in accordance with the ideals of FIRST.

As for funding and such, our individual teams have been able to acquire sponsors for our comparatively small competition, and I see no reason why we cannot have sponsors and professional help on this project. No one ever said that the Grand Challenge would be easy (would it be called the “Grand Challenge” if it were?), but it is a chance to take what we have learned from FIRST and apply it to real life, to affect change in the greater world, and perhaps learn more than could have ever hoped to learn. And if it doesn’t work out, we will have learned, and we shall try again, as we in FIRST do time and again. When we face a challenge, the point is not to immediately look at the scale of the challenge and declare that it is impossible to succeed. The point is to try, to learn, to innovate, but not to give up without even an effort, for if we do not attempt a challenge, who will ever know what may have been?

if i remember correctly someone alreday posted about this. this contest is pretty big though, i remember reading anbout it in MIT’s technology magazine. That is way to complicated for me, but good luck to those who try.

Jnadke, you’ll give us a grand if we can make it 10 miles? If you can put your money where your mouth is, I’ll tell my team that. (It certainly would help with the cost of building something like this.)

10 miles isn’t a very long distance… I’ve run further than that.

The DARPA Grand Challenge looks like a difficult but fun project to work on. nod I mean, you’ve got to look at it in the same way rookies look at FIRST. You say “Oh God, that looks hard… I think I can… I think I can… I think I-- HEY! It works?!”

Someone was talking about using Skyway wheels for their DARPA project link:smiley:

Check this out. The size of the small truck’s tires in the pic is 6 in di by 4 in wide to get an idea for size.
And what about this You can get it with a 72 cc engine and it can do 40+ mph.

The hardest part of this project is the obstacle avoidance. The rest is just dirty work.

*Originally posted by sanddrag *
**Check this out. The size of the small truck’s tires in the pic is 6 in di by 4 in wide to get an idea for size.
And what about this You can get it with a 72 cc engine and it can do 40+ mph.

The hardest part of this project is the obstacle avoidance. The rest is just dirty work. **

…dirty work including devising a way to get enough fuel on board to make the entire trip.

Well, there is an alternative to carrying all the fuel on the 'bot. If we can manage it, we are allowed by the rules to have autonomous refueling, though I would imagine that it would be substantially more difficult to manage that as well…

Why is fuel an issue. Just mount more tanks on the rover. Make it real nice.

Seriously, the real challenge here is not mechanical. I think that we could easily get the controls mounted and designed. We could also easily mount the sensors. The difficult part is in the programming. We need to find a few “cracker-jack” programmers that can program a concrete block.

I think that if this is to be done, a group of the willing has to be assembled quickly to accomplish (or at least explore accomplishing) the goal. There are robotics teams that have the resources (if combined) to at least attempt this challenge.

Just remember. This rover has to go for longer than 15 sec. and it has to do it. You can’t just race for the controls.

The thing that bothers me about this competition is that it’s for the military.

I much prefer being a geek for geek’s sake.

*Originally posted by tenfour *
**Why is fuel an issue. Just mount more tanks on the rover. Make it real nice.

Seriously, the real challenge here is not mechanical. I think that we could easily get the controls mounted and designed. We could also easily mount the sensors. The difficult part is in the programming. We need to find a few “cracker-jack” programmers that can program a concrete block. **

Unfortunately, it may not necessarily be as simple as increasing the amount of fuel, though that it the most obvious solution.

To do things efficiently, it’s a matter of balancing fuel economy with fuel consumption. You could use one of those small RC Gas engines, but their consider how much fuel you’d have to carry on board to make the entire trip. I doubt that engine is capable of moving the weight of the necessary fuel over that distance, nevermind everything else.

Also, keeping something from breaking itself while operating in harsh conditions over ten hours is a mechanical issue. It’s not a cake walk for the design of the vehicle itself, either.