• Jun 9, 2006

We had the pleasure of catching Dr. Sebastian Thrun's keynote presentation at the 2006 Sensors Expo on Tuesday, where he spoke to the audience about his experience leading Stanford's DARPA Challenge entry to victory last year and how the project's success may affect the way we drive in the not-so-distant future.

The technical content of "Stanley" is fascinating, of course. Due to the relative inaccuracy of GPS (2 meters of error doesn't cut it on a 2.5-meter-wide mountain pass), a variety of sensors were used for environmental recognition, allowing the vehicle to discern "good" terrain from non-drivable areas. A array of five lasers scan the area in front of the vehicle for obstacles, and their downward trajectory combines with the forward movement of the vehicle to allow for the return of 3D data. Unfortunately, the 20 meter range of the lasers and the 70 Hz scan rate limits the maximum practical speed of the vehicle to 35 MPH or so when relying solely on this navigational device.

To allow the faster travel speeds thought necessary to secure a victory, an optical camera was implemented to learn from the laser system how drivable terrain appears, eventually allowing the vehicle to see much further down the intended path and thus facilitating higher speeds. Doing so isn't as easy as it may seem - color can't be used due to the wide variety in coloration of drivable surfaces (roads may be brown, black, red, tan, or any number of other shades), and texture detection is nearly worthless as Thrun noted that the sky is smoothly textured but yet not a practical path. Radar is also used to provide ranging information.

[Click through for much more on this vehicle and the future of (not) driving...]

 

Eventually, the vehicle was able to autonomously travel fast enough to scare its human passengers, and so shock sensors were employed to slow the vehicle after hitting something too fast. It was admitted that this reactive system is imperfect compared to the anticipation provided by a human driver, but for now it's considered to be a workable solution.

Thrun stated that the team's success depended heavily on collecting physical data from actual driving to allow refinement of each system. Amazingly enough, the vehicle's autonomous navigation system provides all-weather capability, as proven by a video that shows the vehicle driving at high speeds across the desert in a torrential downpour (the vehicle's windshield wipers weren't even keeping up).

Thrun noted that each of the teams considering this to be much more of a contest versus nature and physics than against each other, and was careful to point out that the 11-minute gap between the finishing time of Stanford and Carnegie Mellon's second-place entry is not significant in the grand scheme of things. Instead, it's more important to focus on the fact that the competition evolved from a complete failure to one where five vehicles finished in only a year. This shows the rapid pace of advancement in autonomous navigation.

Looking past the competition and towards the future of transportation, Dr. Thrun threw out some sobering statistics. Traffic fatalities are the leading cause of death for those aged three to 33 years, and 1500 people die every year from collisions with trees - a slow-moving target if there ever was one. While gridlock remains an ever-increasing concern around the world, the average stop-and-go situation still involves roadways that are 92% free of cars; that is, due to a human's need to maintain some space to provide for sufficient reaction time, we are using less than 10 percent of a road's capacity. Essentially, people are poor drivers, and autonomous navigation will offer the ability to offload certain tasks from the wetware behind the steering wheel.

It was acknowledged that this would not happen overnight; rather, driving aids such as distance-following cruise control (including the upcoming stop-and-go versions that are useful in urban environments) and lane-departure warning systems would be the first step. Eventually, Thrun thinks that such technology would be able to improve urban environments by allowing the remote location of parking lots (your car might drop you off at the office and then park itself miles away), and could also improve the mobility of senior citizens as life expectancies continue to increase. Thrun injected a personal anecdote at this point, telling about how his family recently had to take the keys away from his elderly father. Such a collapse of one's social network often leads to a senior's demise, and as such autonomous navigation could lead to longer lives.

By implementing autonomous navigation in individual vehicles instead of trying to build "smart highways" (those with built-in navigational devices, such as magnetic strips), Thrun feels that the benefits of such technology can be brought to the average driver much more quickly and at a lower overall cost. It was suggested that perhaps special lanes could be dedicated to autonomous vehicles as a starting point, perhaps much as car-pool/HEV lanes are used today.

During the Q&A session, it was asked why an American university didn't use a domestic vehicle. Thrun stated that Ford offered some level of support but generally was not interested in being associated with a military project or the aura of failure that surrounding the inaugural DARPA Challenge, and GM showed no interest in the competition. It was said that, generally speaking, there is far more interest in autonomous navigation coming from Europe and Japan than there is from the US.

Certainly, after watching this presentation and gaining a far better appreciation for what has been accomplished and potential lies aead for autonomous navigation, we await the upcoming Urban Challenge. For this event, Stanford will be fielding a VW Passat.

 



I'm reporting this comment as:

Reported comments and users are reviewed by Autoblog staff 24 hours a day, seven days a week to determine whether they violate Community Guideline. Accounts are penalized for Community Guidelines violations and serious or repeated violations can lead to account termination.


    • 1 Second Ago
  • 14 Comments
      • 8 Years Ago
      From what I read it sure seams like VW did like 90% of everything for the vehicle (including the robotics and much of the software). Maybe they should be interviewing VW all the time.
      • 8 Years Ago
      The quote from the Stanford guy about defects threw me off. I knew no one finished the 2004 challenge. I disagree that Stanford was the driving force after reading about the vehicle and the collaboration.

      My guess on why it was called the Stanford Racing Team is because it looks very bad for the American military to need the technology of a foreign company...

      "The purpose of the DARPA Grand Challenge2004 is to leverage AMERICAN ingenuity to accelerate the development of autonomous vehicle technologies that can be applied to military requirements." (source:
      http://www.darpa.mil/grandchallenge04/overview.htm)


      "DARPA’s mission is to leverage ingenuity and research to develop transformational technologies that give our armed forces a decisive edge." (source: http://www.darpa.mil/grandchallenge05/overview.html)

      With VW winning the competion they needed to change their story I see. No mention of "American" anymore. Anyway, I just am of the opinion that all your "advanced" technologies needs to be unique and not available to other countries or else they will send a better unmanned vehicle against (in time of war) yours and win or they will sell their better version to the people you are against. Just not smart. We are talking military purposes here.
      • 8 Years Ago
      The race grabbed a lot of attention in the US, but robot cars are an old hat in Germany. It was not VW who started this, but Mercedes, back in the 1980s, with software written at Bundeswehr University Munich under robot car pioneer Ernst Dickmanns. The site below has a pic of a fast Mercedes robot that drove "1678 km on public Autobahns from Munich to Denmark and back, at up to 180 km/h, automatically passing other cars":
      http://www.idsia.ch/~juergen/robotcars.html
      • 8 Years Ago
      "If you look at the people in the team on the following
      URL, you'll notice not all of them work for VW."

      Nice that the "Vehicle Lead" is from VW though, that says eons.



      "You can think all you want that VW did most of the work but by doing so, you're marginalizing the contributions of the other people in the group."

      Every f'ing day the things that GM and Ford engineers do are marginalized by import fanboys or uneducated domestic vehicle haters. The EV-1 was a lot of hard work and money (more than Stanley cost) yet there are still a ton of Americans that would say GM doesn't know anything about alternative fuels (just as one example). American import fanboys get on GM for using their own (owned Opel since early 1900's) German designs for cars here as if that is bad. The difference is that GM at least owns Opel. Stanford doesn't own VW. Example of what I am talking about: http://www.autoblog.com/2006/06/07/snagged-2007-honda-cr-v/#c1586322

      The first time I got the impression that VW did most of the work is from reading a forbes.com article when the race was not very old.
      • 8 Years Ago

      "Both vehicles were created in a collaboration between the Volkswagen research department, Volkswagen Group’s Electronics Research Laboratory (ERL) in Palo Alto, California, and Stanford University (hence the nickname of the prototype). Autonomous driving basically forms one of the main research subjects of the ERL. Its implementation represents an immense scientific and technical challenge. Many aspects of the autonomous automobiles will eventually be used in other, more conventional driver assistance systems."
      (source: http://media.vw.com/article_display.cfm?article_id=9616&uid=32503)

      Notice the ordering: VW's research department, ERL and Stanford. And notice that the ERL's main existance is autonomous driving. Put two and two together and basically Stanford wrote some software as VeeDub stated.


      "In terms of basic technology, the off-road vehicle is almost the same as the series production model and has simply been modified with complete underbody protection and more powerful shock absorbers. However, Volkswagen Group Research has also transformed the Touareg into a mobile high-tech laboratory. Numerous sensors and a combination of four laser detectors obtain all the data needed for the driverless car to find its way both quickly and safely. The systems were supplemented with stereo visual display units, highly-developed 24-GHz radar equipment and an extremely accurate, satellite-supported GPS navigation system, which shows the exact location of the vehicle digitally to the exact millimetre." (source: http://media.vw.com/article_display.cfm?article_id=9701)

      No mention of Stanford in that paragraph mentioning technology.

      "Software defects were a major cause of failure for teams in the 2004 DARPA Grand Challenge and we are working with Coverity to eliminate any defects in Stanley," said Dr. Alex Aiken of the Stanford Computer Science Research Group. "We were able to identify and subsequently address software issues using Coverity's technology that we were unable to find through manual testing. The Stanford Racing Team is very impressed with the results from Coverity." (source: http://www.coverity.com/news/nf_news_09_13_05_story_13.html)

      If Stanford wrote the software then they must be the reason they didn't make it through in 2004 according to this. At least Stanford got the right software to detect their bugs for 2005.


      This is what I have read to make me believe it was more VW than Stanford.

      • 8 Years Ago
      Great read! Although Ford may not have offered any official support, MIT's Team Manticore did field a modified F-150. The didn't make it past the qualifiers.
      • 8 Years Ago
      The excellent Nova episode on these cars certainly doesn't show Volkswagon doing any of the work.
      • 8 Years Ago
      Lithous, 2004 was the first year of the competition and noone won. In fact, all the entrants failed miserably. As Mike Sullivan stated, Stanford didn't even compete in 2004. Also, Stanford is the driving force behind the project and not VW. That's why their team was called Stanford Racing Team, and not VW Racing Team. While VW did play a role, Stanford's contributions were a greater factor in the victory. Anyone who knows robotics knows that it's all about the software and being able to handle different environments. Without Stanford, VW may have done just as good as perhaps Berkeley during that event.
      • 8 Years Ago
      Interesting bit on Thrun & Gerdes' "steer by wire" prototype by way of Intel CTO Justin Rattner, a big fan of the race: http://blogs.zdnet.com/OverTheHorizon/?p=8. I'd pay to take that thing for a spin... sigh.
      • 8 Years Ago
      To the above comment:

      Stanford did not compete in 2004. For those competing in 2004, software was a big component of the failure to complete the challenge, but not by Stanford.
      • 8 Years Ago
      I have heard there is collusion between the big auto makers and big oil and that electric cars are out of consideration for now. I personally believe both GM and Ford could get back on their feet financially if they would produce a car with the following characteristics:
      - Plug-in electric capable of going at least 300 miles on one charge. Hopefully, a fast charge, such as with a Super capacitor in lieu of a set of batteries.
      - Must be large enough to feel safe in.
      - Competitively priced with other smaller cars such as the hybrids.

      You had one that was close to this, the EV1. But, rumors are you got them all back and tore them up for some unknown reason.

      When can I get one. I drive two Toyotas because I feel I get much better reliability that the GM products I've driven in the past. If you were to put a car as I described on the market, I'd return to GM.
      • 8 Years Ago
      "Thrun stated that Ford offered some level of support but generally was not interested in being associated with a military project or the aura of failure that surrounding the inaugural DARPA Challenge, and GM showed no interest in the competition. "


      GM was to busy backing us... MUHA HAHAH

      http://grandchallenge.njit.edu

      We were unable to compete because our vehical was not ready, un-like stanford... we had to do everything our selfs.

    • Load More Comments