Slashdot Mirror

What does the military need a robo-car for anyway?

Supply convoys. Do you remember that story from last year where troops refused to go on a supply mission because it was too dangerous? From DARPA's 2004 FAQ:

Q: How will autonomous ground vehicle technology benefit the military?

A: The military is looking toward a future in which manned and unmanned systems work together on the ground and in the air to provide enhanced capabilities for U.S. forces. For an example of the utility of unmanned ground systems, consider Operation Iraqi Freedom. The combat troops moved quickly toward Baghdad, and were followed by supplies and materiel. Protecting the supply lines was critical. In the future, unmanned systems may be able to conduct resupply missions without using humans as drivers, and without requiring additional troops for protection.

"Even as such, they're a waste - there are far more impressive things upon which a group of talented young engineers could focus their efforts."

Like researching autonomous vehicles for the military...

And ACIP is just one of several programs funded by DARPA IPTO (Information Processing Technology Office).

Just wait till the Playstation 4 comes out with massively parallel streaming (CELL) processors for graphics, AND cognitive processors for intelligent characters.

And ACIP is just one of several programs funded by DARPA IPTO (Information Processing Technology Office).

Just wait till the Playstation 4 comes out with massively parallel streaming (CELL) processors for graphics, AND cognitive processors for intelligent characters.

Exactly. It'll pay a prof and several grad students for a year or two. Not a lot of money.

Besides, Darpa is funding a large number of different AI projects right now, as they have been for the last, I don't know, few dozen years... Why is this newsworthy?

So the organisation responsible for proposing the Total Information Awareness program wants to develop a super AI with the ability to make decisions on the fly regarding the deployment of military resources? Fantastic.

Seems like a great system, but I just cant understand this statement: "Currently, Tor development is supported by the Electronic Frontier Foundation. Tor was initially designed and developed as part of the U.S. Naval Research Laboratory's Onion Routing program with support from ONR and DARPA."

*Puts on tinfoil-hat* isn't the guys at *.mil making their jobs harder by doing this? anonymous "terrorists" communicating freely without any traces, or do they already have this covered in the system? a honeypot?

Exoskeletons a la DARPA The military has been interested in combat exoskeletons for some time. DARPA has been poking its nose around this idea for years.

I think the concept has some potential. How about some thing other than hunting. How about patrolling the streets of Fallujah with a rigged unmanned HumVee from your home ? Real life Counter Strike anyone ?

The concept sure is interesting. Couple it with a DARPA challenge HumVee and soon it would be like playing real life Counter Strike. Patrol and shoot from your computer..NICE !

I run one of the DARPA Grand Challenge teams, and I'm up-front about the military implications. Some of the academic teams don't want to admit they're part of a weapons program. But they are.

Of course you realize that exactly that was tried with the DARPA Grand Challenge [DARPA], right?

The Joint Unmanned Combat Aerial System (J-UCAS) has nothing to do with air-to-air and cruise missile targeting. It also is not aimed at air-to-air combat of any form. It is designed to do suppression of enemy air defenses (SEAD) and in the Navy's case, certain low level strike missions as well. Both the Air Force-derived X-45 series (built by Boeing) and the Navy's X-47 series (Northrop Grumman) have flown as part of this program. Check out the DARPA site for more details.

The U.S. military (through DARPA and other agencies) has been funding research on micro-UAVs for years. Not too hard to think about what they would be good for if you change your mindset to try to solve the "problem" of being able to kill people more effectively.

Some interesting links:
Pretty scary justification for "why bother": http://www.darpa.mil/tto/mav/mav_auvsi.html
For those that don't know, an "ACD" is a program where they actually build a working weapons system out of the technology: http://www.darpa.mil/tto/programs/mavact.html
This is one of my favorate MAV's: http://www.aerovironment.com/news/news-archive/was p62.html
A general links page that shows the amount of research being done in this area: http://www.casde.iitb.ac.in/IMSL/amitay.html
FAS collects a ton of information about U.S. military programs and systems: http://www.fas.org/irp/program/collect/mav.htm

The U.S. military (through DARPA and other agencies) has been funding research on micro-UAVs for years. Not too hard to think about what they would be good for if you change your mindset to try to solve the "problem" of being able to kill people more effectively.

Some interesting links:
Pretty scary justification for "why bother": http://www.darpa.mil/tto/mav/mav_auvsi.html
For those that don't know, an "ACD" is a program where they actually build a working weapons system out of the technology: http://www.darpa.mil/tto/programs/mavact.html
This is one of my favorate MAV's: http://www.aerovironment.com/news/news-archive/was p62.html
A general links page that shows the amount of research being done in this area: http://www.casde.iitb.ac.in/IMSL/amitay.html
FAS collects a ton of information about U.S. military programs and systems: http://www.fas.org/irp/program/collect/mav.htm

There was an article in a recent Car and Driver, about the DARPA sponsored Grand Challenge. AI piloted vechiles had to traverse a course in the California dessert. None of the vechiles made it very far at all. As much as we'd all wish it, this kind of thing is a long way off. Here's DARPA's official site.

Seems to me some of this technology might be able to be put to good use for the DARPA Grand Challenge 2005, in which autonomous vehicles race across the U.S. desert, driven by their waypoints and obstacle avoidance systems. I'm not at all surprised Cornell is doing some of this autonomous vehicle research.

Last year, The Ohio State University's TerraMax and Carnegie Mellon's Red Team did very well at the DARPA Grand Challenge. Here's some good coverage on Science Blog. There was some other really good blog coverage that gave a play-by-play breakdown of how each autonomous vehicle did the day of the event and what kind of troubles it ran into, but I can't find that via the Googling right now. :) There's also tons of previous Slashdot coverage on the Grand Challenge, and there's a pre-2005 event coming up very soon for interested people, I know.

New supercomputing advances on the way will radically redefine the industry.

I refer to a DARPA funded project created to fill in the performance gap between today's inadequate SC technology and tomorrow's (quantum, bio) still far in the future stages.

The project is called HPCS, which does not stand for High Performance Computing System, but rather High Productivity Computing System. The point is not to increase flops but increase value. The earth simulator, for example, is down for maintenance about 2/3 of the time and can only be reliably run in 8 hour chunks. The ASCII series may have high peak performance but averages only 5-10% of that. DARPA knows that if this is the state of the art, then there is work the do.

Three companies are currently doing research on proposals and are DARPA-funded through 2006. The three are Sun, IBM, and Cray. Two companies will continue from 2006. A working product will be delivered in 2010.

Many radical new technologies are in place, but as you can understand most of it is tightly under wraps. But do some reading on the DARPA page and you will find some interesting things.

Here is the link

- employee at one of the three aforementioned companies

TMR is a DARPA Advanced Technology Office program... other projects in the same office are here.

TMR is a DARPA Advanced Technology Office program... other projects in the same office are here.

if this was that easy i doubt that they would have the DARPA challege Your not going to get anythign easily to do this. I recomend you take the time to do it yourself, it would be good exercise too. Sometimes there something to be said about doing things manually.

The tech isn't there yet. There is something similar to this idea, it's a supposed "smart" floorvac called the Roomba which is just dumber than a pile of bricks. Not to say this is the leading technology in the field, but i'm fairly certain it's an indication as to where retail autonomics is at. Also remember the Darpa Project with autonomous ground vehicles, some of which, didn't even make it out the gate.

I hate to discourage your effort, but hey.. you get sun, you breathe fresh air not recycled by a case fan, and sweat off a few mountain dew pounds. It's a win-win situation!

...they already tried something like this, but with cars... DARPA. :)

And, well, it turned out a miserable failure. But good luck!

Why is it that when I click on the link (titled: www.darpa.mil/grandchallenge) at the bottom of the page at http://www.grandchallenge.org/media.html I get a "page not found" error?

Is it incompetence on the part of the creator of that page or is it because I'm surfing from outside the USA and DARPA figure it's not a good idea for potential terrorists to see what's going on with this challenge? :-)

Are non-American citizens allowed to participate? I tried looking at the Rules Page but it's not up yet. I don't recall if there was a stipulation which restricted participants to American citizens.

Given DARPA's great R&D track record in the past (Internet and what not), I would've liked to participate in the contest *purely* from a scientific curiousity point of view - and I bet a lot of nerds all over the world would like to overlook the fact that the contest is sponsored by a military agency (prize not withstanding - since it's US taxpayer money). Just as long as DARPA lives up to it's name and does not morph into OARPA - it's happened way too many times in the past.

Incidentally, the link to the official page is incorrect on that page. The site linked to in the article seems to be just a mirror of the darpa.mil site, however.

...one of the DARPA IXO programs, Cougaar, has developed a fair number of message transport techniques over the last few years. Good times.

DARPA invented the Internet. History of the light buib can be found here Karl Benz of Germany invented the first motor car. Antonio Meucci invented the telephone.

The next step is a rerun of the DARPA Grand Challenge, this time for $2 million. Another article available at The Register. (I suspect we'll be seeing a story on this shortly.)

Submitted AC for your karma-free pleasure!

Just out of curiosity, any that you can think of off the top of your head? :)

Well, DARPA Solicitations is a good place to look.

All you have to do now is replace these cones with mines, add some pattern recognising AI, and you have the Self Healing Minefield.

For something intelligent on this topic, see this DARPA/Boeing presentation. DARPA has a number of "smart airfoil" projects. They've tried shape memory alloys. They've tried ferroelectric fluids. They've tried piezoelectric materials. It looks like the first flight test will be a piezoelectric system on the rotor blades of an MD-900 helicopter.

It's not yet clear that it's worth the trouble, but R&D continues.

For something intelligent on this topic, see this DARPA/Boeing presentation. DARPA has a number of "smart airfoil" projects. They've tried shape memory alloys. They've tried ferroelectric fluids. They've tried piezoelectric materials. It looks like the first flight test will be a piezoelectric system on the rotor blades of an MD-900 helicopter.

It's not yet clear that it's worth the trouble, but R&D continues.

"It navigates the hall way with a verity of instruments: Sonar for accurate long distance coverage. IR for close obstacle avoidance. Shaft encoding for accurate navigation once location is determined. Web cameras for reading room numbers off the wall."

They should have entered this bot in the DARPA Grand Challenge. Looks like it might have a had better chance then most of the other entries.

This really isn't all that new. The U.S. Naval Postgraduate School has been
sending their Infosec students to play Capture the Flag at Defcon for the last couple years as well as
this year's Interz0ne conference. In
fact, there was only one team (Anomaly - and they won ironically) that didn't
have government personnel or contractors on their team.

Also, Immunix, a DARPA funded hardened Linux version has also
been put under fire during CTF for the last couple year. (Their team placed a
solid second both times).

The Feds have learned over the last couple years that they
are behind the ball in terms of normal unclassified security training for their
personnel. These conferences have been really good at given them some real
world training that they normally don't get.

It's nice to see my tax dollars being put to a good use for
a change. Plus it makes the "Spot
the Fed" game MUCH easier.

For example, this DARPA initiative on Dynamic Mobility "--biologically inspired appendages to demonstrate multifunctional, dynamic, energy efficient and autonomous locomotion to enable revolutionary mobility capabilities such as running over multiple terrains, climbing (trees, cliffs, cave walls), jumping and leaping, and manipulating the world with an appendage in tasks such as grasping and digging. "

This is in contrast to Japan where I think more of the money is in consumer products. Not surprising to see the different focus in robotics.

And in other news, The DoD runs The site www.darpa.mil is running Microsoft-IIS/5.0 on Windows 2000 .

grandchallenge.org has timed out for Netcraft, but I assume they run the same thing.

I am not an expert in programming, but I work on fault tolerances in building processes. Tolerances come in two general categories, one is 'slop' the other is a factored tolerance where you predict where and how the processes will interact.

Slop, AKA 'quick and dirty', is the easiest and the cheapest way to design. It requires little forethought and marginal design ability. You begin by applying generic and standardized processes, you do not venture into any customization, you do not get fancy in any way. You specifically avoid all situations where two things have to line up and interact. This method is inneficient, not in terms of design resources, but in maintanance and operational resources (in computing, I imagine it would be memory, bandwidth et cetera). In architectural terms, this would be a shanty town. A million people can live in a shanty town, but it isn't pretty.

My preferred method is a precisely defined solution.
1. You start off by designing for the inherent demands of the overall system, which look at the long range extremes of usage, you decide how robust you want your system to be and game out 'all' likely failure points at intersections and interactions of the overall architecture. You benchmark past experience and develop overall standards for performance. The precise details of this overall architecture will determine day to day maintanance and operation requirements. You need to be aware of not only the the way elements within the system interact, but the physical properties or specifications of each element to be sure you know what you are doing and how they will perform. You need to be an expert.
2. As a co-continuous problem you design each particular system to a level of robustness based on its appropriate use. You need to assign safety factors, and acceptable failure levels so as not to waste resources while maintaining protection. In some cases failure can be catastrophic and people can die (the space shuttle systems, 911 call centers) or worse lose money (banking). The best example of a fault tolerant system I can think of besides the internet is the Interstate Highway System, which was originally designed as a transportation system that would remain operational in the event of a nuclear strike (its form isn't an accident, it was designed to specific criteria). In architectural terms, this kind of fault tolerance is a skyscraper city. Skyscrapers are durable and beautiful but are extremely expensive to design and construct. A skyscraper could stand for 100 years without maintanance of any kind, where a shanty will fall to pieces in 5 years.

Fault tolerance needs to be put in perspective. It is only useful based on how much it costs to fix and maintain a system and its 'aesthetic' value (it is always hard to sell an ugly or stupid design). A $1 million system for a $1 hundred dollar problem is not effective unless it is a critical application (a missile or a laser for eye surgery). It may well be cheaper to employ a person or ten to keep some system up, rather than make it 100% bullet proof (most web sites can fail for a few minutes an no one will care). It depends on the situation. As far as aesthetics of your work, I don't know anyone who wants to live in a shanytown, they have a tendency to get cholera, malaria et cetera and die, it isn't necessary to live like that. We invented trailer parks to solve that problem and it is now Americas #1 form of affordable housing.

Our friends at Darpa are researching this problem and have a good outline of these issues.

Now you can finally make a data port that connects directly to a person. You can theoretically send and receive neural signals which can interact and control a machine...or perhaps the other way around. If you think I full of crap, check out this link. Join the Army and you too can be a cyborg!

If a team leaves every 5 minutes, (and assuming the first few hundred yards is relatively easy going - you find that on most courses of any nature), then we are going to have an awful lot of bunching at the first point the vehicles start dropping below 25mph. Interestingly, the rules state that the team in front (i.e. being passed) has right of way, unless E-stopped.

Well, see DARPA funded the initial development of this toy we've all come to love and refer to as "the Internet". So whenever DARPA starts pumping money into newer and even more exciting toys, all the geeks' little ears perk up... "Wow, the things I could do with an autonomous robotic off-road vehicle! Why, I might even get laid!"

Now, personally, I'd rather have my own Predator drone, but I can see the appeal of stomping around the boonies in a vehicle you don't have to steer while you're tipping your beer back. Or while going into the cooler for another beer. Or while taking potshots at what appear to be deer in orange reflective vests. Or ever steer, for that matter. Frequently lost? Embarrassed to ask for directions? This is the vehicle for you! However, it requires a special breed of owner: somebody with several million dollars burning a hole in his back pocket, who is also not afraid to admit that his off-road vehicle is smarter than he is. Much, much, MUCH smarter.

By the way, google sez DARPA is the Defense Advanced Research Projects Agency Note the ".mil" suffix; this denotes it is run by the greatest toymakers in the world: the US Military Industrial complex.

I think this was the whole idea, DARPA wants to make sure they have some kind of "race" on their hands, not just one team putting along till the end.

Keep in mind that they're doing sequential starts. So it won't really be a race, just a long course with a few vehicles strung out one behind another. Far behind. And lots of disabled vehicles which have been shut down and pushed off the course (those few which even made it out of the starting gate).

DARPA is setting up a live update page where you'll be able to get a map of the course and watch in real time where the vehicles are during the race.

According to this schedule the first vehicle (which will be CMU's Sandstorm) is scheduled to depart at 6:15 AM PST on Saturday.

The Wired article was written on Tuesday night; the Caltech and Ohio State teams went ahead and qualified on Wednesday. At the time Wired was writing, it was correct that only one team had qualified. As of last night, three teams have done so. See the DARPA media page for updates on who has made it so far.

The Wired article also speculates that even teams which don't complete the qualifier will be allowed to try the race, but I haven't seen any confirmation of that on the DARPA site. Of course, if a robot can't make it through a one mile practice track, it's unlikely to get across 150 miles of desert. But letting them try would make for a more exciting race day.

From the press release:
The results of the attempts of today's group break down as follows:
SciAutonics II, Team Cal Tech and Virginia Tech completed the course.
Team CIMAR , Team ENSCO, TerraMax nearly completed the course.
Axion Racing, Digital Auto Drive, The Golem Group, Palos Verdes High School, Team CajunBot, TerraHawk partially completed the course.
The Blue Team, Rover Systems, SciAutonics I and Team Phantasm terminated their attempts.

In any case there isn't much harm in allowing anyone to compete. It looks like many of them will fail in the first few minutes leaving it to Red Team.

The results of the attempts of today's group break down as follows:

• SciAutonics II, Team Cal Tech and Virginia Tech completed the course.
• Team CIMAR , Team ENSCO, TerraMax nearly completed the course.
• Axion Racing, Digital Auto Drive, The Golem Group, Palos Verdes High School, Team CajunBot, TerraHawk partially completed the course.
• The Blue Team, Rover Systems, SciAutonics I and Team Phantasm terminated their attempts.

Here is the link to DARPA's coverage on the events: results

DARPA is continuing to lower the requirements to qualify for the race in hopes to allow more than about 5 contestants. They are also overlooking some of the safety rules in the pit area to allow the teams to do as much troublshooting as possible without a DARPA rep. being present. Originally teams were not allowed to move their vehicle without a rep. present or even put it into autonomous mode. The teams are now being allowed to drive their vehicles in manual mode and park them in autonomus mode in an attempt by DARPA to allow as many teams to qualify as possible.

The big news today is that three more teams have successfully finished the Q&D course.

Team Caltech. The Caltech team is the first to completely navigate the course using a NavCom GPS receiver.

Sci-AutonicsII is the first to unequivocally clinch a position in the Grand Challenge competition by successfully navigating the Q&D course two times. Sci-AutonicsII has decided to pass on their originally scheduled second run of the Q&D course.

Virginia Tech is the fourth team to completely navigate the course.

Team CIMAR (ASI & Florida State) made two more runs on the Q&D course. Their second one was made with reflexive obstacle avoidance functioning but without IMU integration. They were on their way to complete the course until they passed under a walkway that briefly blocked GPS. The dropout caused the vehicle to veer hard right and swap paint with the retaining wall. The damage was only superficial despite brutally applied custom pinstipes to one of the Preco RADAR sensors.

Attendance was about the same today except it didn't appear that there was as many media representatives present. Again temperatures were in the 90's. I acquired a media pass today and was allowed access to almost every area of the speedway including the pits and the start line. This will allow me to film each entry up close and interview members of the teams. DARPA is also publishing the daily events here and here.

Vehicle inspections on the rest of the field were performed today. DARPA is reacting as fast as they can to modify the rules and give every opportunity to each of the teams in hopes they will be able to qualify. DARPA is now allowing the teams as many appearances on the Q&D course as requested by the teams. The Q&D that was scheduled today became an opportunity for teams to iron out their problems on the track.

The Blue team with the CyberRider (the motorcycle entry Web Site) was the first on the Q&D course. It traveled about 20 feet when it fell over and exposed it's greasy side.

Team ENSCO Web Site traveled to the first major turn and failed to navigate the first sharp turn. This is also the same place Team TerraMax web site failed today and the Sci Autonics web site team.

The first turn appears to be difficult for the vehicles that make it there.

The ASI/Florida State Web Site had two more runs today. They disabled all of the perception systems and successfully ran about 1/3 of the course. The first run, the vehicle was driving like a "drunk sailor" according to a member of the team. Florida State took the recoded path data to tune their vehicle controller. ASI reported the vehicle tracked much better on the second run. I get the impression that a few more Q&D course test runs will be performed before the vehicle is ready to qualify.

Team Caltech Web Site made two more runs on the Q&D course. Well, both runs were consistent, but not as good as yesterday. Both times, cleared the start line and banked hard left as if it were going to the last way point. I hope to find out today what happened.

There were several other teams that attempted the Q&D course today with similar or worse results.

The highlight of the day was CMU's Red Team web site. As anticipated they made it to the finish line of the course and electrified the spectators and increased pressure to the rest of the field.

That's the highlights of the second day of Q&D testing.

Attendance was about the same today except it didn't appear that there was as many media representatives present. Again temperatures were in the 90's. I acquired a media pass today and was allowed access to almost every area of the speedway including the pits and the start line. This will allow me to film each entry up close and interview members of the teams. DARPA is also publishing the daily events here and here.

Vehicle inspections on the rest of the field were performed today. DARPA is reacting as fast as they can to modify the rules and give every opportunity to each of the teams in hopes they will be able to qualify. DARPA is now allowing the teams as many appearances on the Q&D course as requested by the teams. The Q&D that was scheduled today became an opportunity for teams to iron out their problems on the track.

The Blue team with the CyberRider (the motorcycle entry Web Site) was the first on the Q&D course. It traveled about 20 feet when it fell over and exposed it's greasy side.

Team ENSCO Web Site traveled to the first major turn and failed to navigate the first sharp turn. This is also the same place Team TerraMax web site failed today and the Sci Autonics web site team.

The first turn appears to be difficult for the vehicles that make it there.

The ASI/Florida State Web Site had two more runs today. They disabled all of the perception systems and successfully ran about 1/3 of the course. The first run, the vehicle was driving like a "drunk sailor" according to a member of the team. Florida State took the recoded path data to tune their vehicle controller. ASI reported the vehicle tracked much better on the second run. I get the impression that a few more Q&D course test runs will be performed before the vehicle is ready to qualify.

Team Caltech Web Site made two more runs on the Q&D course. Well, both runs were consistent, but not as good as yesterday. Both times, cleared the start line and banked hard left as if it were going to the last way point. I hope to find out today what happened.

There were several other teams that attempted the Q&D course today with similar or worse results.

The highlight of the day was CMU's Red Team web site. As anticipated they made it to the finish line of the course and electrified the spectators and increased pressure to the rest of the field.

That's the highlights of the second day of Q&D testing.

From http://www.darpa.mil/grandchallenge/media_feeds.ht m

On Saturday, March 13, DARPA will provide same-day coverage via satellite of the Grand Challenge start and highlights at the following times:

Live coverage of the start: 6:30 - 8:30 Pacific/9:30 - 11:30 a.m. Eastern
Video news release: 11:00 - 11:30 a.m. Pacific/2:00 - 2:30 p.m. Eastern

Coordinates for both feeds:
Satellite: AMC 9, Ku, Transponder 03
Space is: 36 MHz
Downlink Frequency: 11760.000
Downlink Polarity: Vertical

Hopefully someone will record these feeds and make them available online for all of us without satellite

• DARPA is seeking to promote innovative technical approaches that will enable the autonomous operation of unmanned ground combat vehicles. In the future, such combat vehicles will operate over varied terrain without the benefit of road signs, pre-programmed routes, etc. Autonomous vehicles must navigate from point to point in an intelligent manner so as to avoid or accommodate obstacles and other impediments to the completion of their missions.

To insure that teams didn't pre-plan, there were these provisions in the original rules:

• The Route Definition Data File (RDDF) will be given to all Participants approximately two hours prior to the first Departure Signal at a pre-Challenge brief.
• Only commercially available data (maps, images, other cartographic products) may be downloaded to the autonomous or safety vehicles prior to the challenge. Use of GPS is acceptable.

Then, when the general route leaked from the Bureau of Land Management, preplanning got completely out of hand. Now teams could predrive much of the route or overfly it. And they did. Two teams had the route laser-scanned from aircraft. This produced a very detailed topo map, with an elevation point every 25cm or so, along with equally detailed aerial photos.

On top of this, DARPA increased the number of waypoints from 1000 or so to 5000 or so.

At this point, it started to look like a breadcrumb-following exercise.

CMU will manually plan the exact route in the two hours before the race with a team of people at workstations in a big trailer. So even the planning is mostly manual. Their vehicle really does have some autonomous navigation capability, but it only uses it if the route doesn't match the mapped path.

So that's the history. From true autonomy to connect-the-dots.

• DARPA is seeking to promote innovative technical approaches that will enable the autonomous operation of unmanned ground combat vehicles. In the future, such combat vehicles will operate over varied terrain without the benefit of road signs, pre-programmed routes, etc. Autonomous vehicles must navigate from point to point in an intelligent manner so as to avoid or accommodate obstacles and other impediments to the completion of their missions.

To insure that teams didn't pre-plan, there were these provisions in the original rules:

• The Route Definition Data File (RDDF) will be given to all Participants approximately two hours prior to the first Departure Signal at a pre-Challenge brief.
• Only commercially available data (maps, images, other cartographic products) may be downloaded to the autonomous or safety vehicles prior to the challenge. Use of GPS is acceptable.

Then, when the general route leaked from the Bureau of Land Management, preplanning got completely out of hand. Now teams could predrive much of the route or overfly it. And they did. Two teams had the route laser-scanned from aircraft. This produced a very detailed topo map, with an elevation point every 25cm or so, along with equally detailed aerial photos.

On top of this, DARPA increased the number of waypoints from 1000 or so to 5000 or so.

At this point, it started to look like a breadcrumb-following exercise.

CMU will manually plan the exact route in the two hours before the race with a team of people at workstations in a big trailer. So even the planning is mostly manual. Their vehicle really does have some autonomous navigation capability, but it only uses it if the route doesn't match the mapped path.

So that's the history. From true autonomy to connect-the-dots.