Stanford's latest stunt for extreme autonomous driving is a tire squealing high speed 180 degree spin into a parking spot. Looks like Junior's all grown up.
A good robot car knows how to make the tough decisions. Stanford’s Junior, the robot car that took second place at DARPA’s Grand Challenge in 2007, has learned some new tricks. Namely it can decide when to follow traditional physics models of how to drive, and when it should simply try to repeat a past performance. What’s the first task Stanford students asked it to test its capabilities on? The same maneuver that every slick operator from James Bond to the Blues Brothers relishes: a tire squealing 180 degree spin into a skin-tight parking space. We’ve got some great videos of this move that you have to see. Check them out below.
The DARPA Urban Challenge (aka Grand Challenge) has provided the prize money necessary to generate major interest in getting cars automated. During the contest in 2007, Junior traversed a simulated town without colliding into moving and stationary objects. Since then, Stanford’s taken on automated parking and we’ve seen one of the university’s cars perform well for that task. Junior’s latest sliding maneuver isn’t really about parking, however. It’s about handling the extremes of driving. Just as Stanford’s new vehicle, Shelley, is looking to race up Pike’s Peak at breakneck speeds, Junior is testing the limits of what robot cars can do. These cars are pushing the boundaries of automated systems and they’re looking awesome while doing it:
As recently discussed in the Proceedings of the International Conference on Robotics and Automation and presented at ICRA 2010, Junior’s slide into private parking spot is all the more impressive for the way that it was performed. Standford has given the car the ability to choose between two different ways of driving. Junior can follow traditional physical models of driving or it can try to replicate a move its seen before. The latter choice is good for hard to understand but repeatable tasks, such as making James Bond look like a pansy. Check out the explanation in the following video:
If you wanted further proof of the power of Junior’s new decision making, watch the video below. You’ll see how a closed loop (physically well understood) approach fails, as does the open loop (imitation, non-physical). When combined, the multi-modal approach succeeds with flying colors:
Junior’s baby bear methodology can place it within two feet of its target every time. That’s remarkably accurate, though still not so good as to warrant you handing your keys over to Stanford anytime soon. As Junior continues to be refined, we may see its capability to hit its target improve, but that’s really besides the point. The real story here is Junior’s ability to cleanly switch between various methods of driving. There’s real power there. An operator can show Junior how to perform a difficult maneuver (like the 180 slide) and Junior can decide when best to follow that demonstration. That means Junior could have an entire language of extreme driving maneuvers it could unleash when called upon, but that it could also drive reliably under normal conditions without them. If we want robots to eventually take over much of our driving duties, this sort of decision making will be critical to keep passengers safe. Watching Junior slide into a parking spot is fun, but it’s also a sign that the cars of the future will be able to respond to any adverse condition with remarkable driving talent. I can’t wait until one of these guys gets into a car chase with police.
[image credit: Kolter et al, ICRA 2010]
[video credits: Stanford, IEEE Spectrum]
[source: Stanford, Kolter et al, ICRA 2010]
Comments
I’m interested in how this can be applied to transit or to combinations of the car and transit, such as PRT and GRT. It would be nice to have a decent PRT vehicle that runs on highways and railroads and feeds into public transit networks (and even groups up with other such vehicles to enter densely populated areas, such as Newark or Hoboken or their train stations).
I’m interested in how this can be applied to transit or to combinations of the car and transit, such as PRT and GRT. It would be nice to have a decent PRT vehicle that runs on highways and railroads and feeds into public transit networks (and even groups up with other such vehicles to enter densely populated areas, such as Newark or Hoboken or their train stations).
add memristors, optical gates, quantum computers and the cop car will be able to guess where you are going before you leave. The manual driver is limited, the future is not. Autonomous Cops is a scary thought let alone autonomous robberes lol.
add memristors, optical gates, quantum computers and the cop car will be able to guess where you are going before you leave. The manual driver is limited, the future is not. Autonomous Cops is a scary thought let alone autonomous robberes lol.
Very cool. I’m waiting for NASCAR as much as for a police chase. Add a bunch of horse power and appropriate braking capability, along with the dashcam. Get one of those drivers to teach a little bit about fender-bending and how to be nice to your pit crew would be awesome.
Very cool. I’m waiting for NASCAR as much as for a police chase. Add a bunch of horse power and appropriate braking capability, along with the dashcam. Get one of those drivers to teach a little bit about fender-bending and how to be nice to your pit crew would be awesome.
There are two problems. First this is a learned maneuver. So after the computer has been shown a 180 degree spin-parking, it can do that maneuver, but if the 180 degree approach isn’t available, a say 140 degree spin should be possible. But because it hasn’t been shown this maneuver, it won’t be able to do it.
Second, because it is just repeating previous inputs, it cannot adapt the procedure to different conditions. Suppose the surface has a different friction coefficient? Maneuver is still possible, but where will you end up? Very difficult to calculate if you’re just following the previous inputs.
Third, because within the maneuver the car is just following the previous inputs, it will not be able to do any adjustments. Suppose you’re to drive through a portal with only 10cm (4 inches) to spare. A human can drive the car through that portal no problem. Easy does it. Similarly, this autonomous vehicle can do the same, as long as it can see the portal, and find its exact position. This works by finetuning the position during the procedure. This is entirely impossible when repeating a previously learned maneuver.
There are two problems. First this is a learned maneuver. So after the computer has been shown a 180 degree spin-parking, it can do that maneuver, but if the 180 degree approach isn’t available, a say 140 degree spin should be possible. But because it hasn’t been shown this maneuver, it won’t be able to do it.
Second, because it is just repeating previous inputs, it cannot adapt the procedure to different conditions. Suppose the surface has a different friction coefficient? Maneuver is still possible, but where will you end up? Very difficult to calculate if you’re just following the previous inputs.
Third, because within the maneuver the car is just following the previous inputs, it will not be able to do any adjustments. Suppose you’re to drive through a portal with only 10cm (4 inches) to spare. A human can drive the car through that portal no problem. Easy does it. Similarly, this autonomous vehicle can do the same, as long as it can see the portal, and find its exact position. This works by finetuning the position during the procedure. This is entirely impossible when repeating a previously learned maneuver.
Okay, now project *way* into the future: The police autonomous vehicle is chasing the autonomous get-away car. The hacker is inside the latter, and he says “Oh, I think I can hack the controls to enable manual driving.” The side-kick says “Manual driving? Are you crazy? Only computers are able to drive cars like this.”
Okay, now project *way* into the future: The police autonomous vehicle is chasing the autonomous get-away car. The hacker is inside the latter, and he says “Oh, I think I can hack the controls to enable manual driving.” The side-kick says “Manual driving? Are you crazy? Only computers are able to drive cars like this.”
The police will get these cars before ordinary individuals do. So, the first robot assisted car chase you see will likely end quickly.
The police will get these cars before ordinary individuals do. So, the first robot assisted car chase you see will likely end quickly.
This is how artificial intelligence should work. Some decisions are made with a higher level of rational concious thought, some are just ‘instinct’ or ‘muscle memory’ from obverving others and imitating them.
This is how artificial intelligence should work. Some decisions are made with a higher level of rational concious thought, some are just ‘instinct’ or ‘muscle memory’ from obverving others and imitating them.
“Beside the point,” not “besides the point,” please.
“Beside the point,” not “besides the point,” please.
A car chase with police will never happen. The law gets the control code, and shuts it down on the side of the road. End of story. It’s nice to dream, though.
It’s called hacking. It’s prevalent, even in the primitive, hardly automated world we live in today. The car chase will happen.
“A car chase with police will never happen.”
BAHAHAHAhahahaha! *wiping away tears*
oh man that’s rich… I bet you’d say a massive oil leak in the gulf spewing hundreds of thousands of gallons of crude a day could never happen. Your avatar pic looks like an old geezer, but you speak with the naivety of a teenager.
A car chase with police will never happen. The law gets the control code, and shuts it down on the side of the road. End of story. It’s nice to dream, though.
It’s called hacking. It’s prevalent, even in the primitive, hardly automated world we live in today. The car chase will happen.
“A car chase with police will never happen.”
BAHAHAHAhahahaha! *wiping away tears*
oh man that’s rich… I bet you’d say a massive oil leak in the gulf spewing hundreds of thousands of gallons of crude a day could never happen. Your avatar pic looks like an old geezer, but you speak with the naivety of a teenager.
(There’s someone in the car though… )
This is link of a video of Junior without a driver, or anyone in the car.
http://www.youtube.com/watch?v=BSS0MZvoltw
This video shows shelley driving at high speed with a passenger in the front passenger seat
http://www.youtube.com/watch?v=-CuVVZq9GfY
he’s there to hit the emergency stop button in case the program controlling the car runs out of control
(There’s someone in the car though… )
This is link of a video of Junior without a driver, or anyone in the car.
http://www.youtube.com/watch?v=BSS0MZvoltw
This video shows shelley driving at high speed with a passenger in the front passenger seat
http://www.youtube.com/watch?v=-CuVVZq9GfY
he’s there to hit the emergency stop button in case the program controlling the car runs out of control
Sorry, headline should read “Like a glove” per Ace Ventura.
Sorry, headline should read “Like a glove” per Ace Ventura.