I'm actually on one of the teams (Georgia Tech), and lane following is a challenge. There are bigger ones in my experience, though. Following a road is not too terribly difficult, but it's the places where lane markings are faded or confusing (think intersections) that really provide the bulk of the problems.
The cars are going to have to be smart regardless. No matter what kinds of sensors are in the roads, there are going to be cars that aren't equipped with sensors, and have to be accounted for. Start including pedestrians, fallen trees, and any other unpredicted obstacles. Failures of the system would need to be recoverable too -- imagine what would happen if the sensors in the roads malfunctioned or went out. If all the cars were dependent on it, this would be a huge problem. But in a situation where the responsibility is on each car, at least other cars could avoid it. I can see a hybrid system taking place, but there will always need to be smarts built into the car itself (and probably a disproportionately higher portion, too).
We are. I haven't worked on that part of the car, but from what I've heard the SICKs combined saturate a 1GB ethernet link. One of the big challenges we're going to have to overcome for the car is data storage. For testing purposes, we'd like to record the output from all the sensors for individual runs. That way software can be tested offline. But with those consuming 1Gbps, the video cameras consuming about 750Mbps each, and other sensors, even storing the data becomes a huge task. We're probably looking at transfer rates somewhere around 3-4Gbps with everything said and done. Thankfully SAIC has offered to lend us a 1TB storage device we can throw in the car during testing runs.
Autonomous lane following by itself isn't actually even all that impressive. If I recall, there was a German who in the 80s set up a car to drive itself along the highway, with only minimal human intervention. It's interacting with other cars and navigating complex (i.e., non-highway) lanes.
The maximum speed any car will be allowed to do on the course is 35mph (might be 30mph, it's been awhile since I looked). There will be blocked streets so cars will have to replan their route, and we can probably assume DARPA is going to throw a traffic jam at us.
And of course, a lot of the purpose of this challenge is exactly what you stated in your post: we have to detect and avoid other cars, use safe and proper driving etiquette for passing others, follow the rules of the road (i.e., four-way stops, etc.), and dynamically adapt our course to the conditions. Chiefly, this last requirement means noticing obstacles (construction, accidents, etc.) and rerouting, but this could also incorporate predicting traffic jams. For instance, if a heavily-traveled section of the map has several intersections/stops, we might be able to predict this or at least notice it when we get close and take another path.
We don't have to deal with pedestrians at this point, but give it another generation or two. Right now it would probably be pretty hard to find any volunteers to do it.:)
I've posted elsewhere in this story, but again to prefix my comments, I'm a member of the Georgia Tech team, Sting Racing.
The course plotting part of the challenge is actually probably the easiest part. It's roughly analogous to you reading a map beforehand -- we're given a file detailing all the aspects of the course (course segments, how many lanes are in the segments, etc., and zones where free driving is allowed) plus a mission file giving the different waypoints we have to reach. This is, relatively speaking, easy.
The difficult part is determining where the edges of the road and its lanes are (GPS is terrible at this; most of the time it's accurate to 10 or so feet unless you're using extraordinarily expensive differential units) which is mostly done using visual scanning. Of course, we also have to detect other vehicles or obstacles in the path (using LIDAR and vision) and also determine the correct "pose" of the vehicle. Then we have to take that information and use it to modify the path we've already decided to take. These problems as it turns out are far, far harder than just plotting courses.
To prefix my comment, let me just say that I am currently on the Georgia Tech team, Sting Racing. According to the rules, we will be on the course simultaneously with the other cars. The other other vehicles allowed on the course will be professional drivers. AFAIK, this is not being done in an actual city, but a small-ish mockup is being constructed for the purposes of the event. I could be wrong, though.
I have to agree. It might not have been resolved this way, but he probably could have been better off trying to explain by explicitly mentioning the dollar sign. Basically explain to them that $0.002 is spoken as "0.002 dollars", and to reach cents from that figure requires dividing by 100.
Not saying anything about the number they get, then assuming there was a high number of direct attempts (i.e., 100,000) the attacks would likely have an even worse chance of working than if it were low. One of the primary reasons users are vulnerable to social engineering attacks is that they're rare (per individual). If this was something that happened routinely to every employee once every month or two, they'd probably be easy to spot. Of course, the additional volume might outweigh the drop in successes-per-attempt, but it's interesting to think about nonetheless.
You don't have anything from the Playstation product line I take it?
I'm guessing you're the one without a PlayStation. I've had my PS2 for nearly six years, and it's still working fine. Yes, I got disc read errors. Yes, I need to clean the laser and/or readjusting it every so often. It's a trivial thing to do after the second time you've done it.
Actually, just yesterday I managed to drop the outer shell when holding the rest of the case. The infamous ribbon cable pulled out of its holder, and I got quite upset. Most of the information I've read on the Internet previously had pretty much stated your PS2 was a brick if this happened. Not one to be discouraged, I found where the ribbon cable needed to be inserted, pried open the case even farther to get to it (requiring a total of 4 more screws to be removed), and got it back in. Reassembled the thing and it worked.
I don't know what your basis for comparison is, but PS2s are pretty damned durable. Yeah, they could be a little better about acknowledging the problems with the lens. But given some minor occasional maintenance, having a console last for 6 years is pretty impressive in my book. Additional kudos to them for making the whole thing easy to access and take apart. Considering the limited space requirements, it would have been quite easy for them to make the thing a tangled mess, and impenetrable to user efforts to repair it.
The problem becomes in what situations is force, even if non-lethal used. if we march on washington because we don't like the results of the next election and start getting zapped and tear gased, I don't think that is acceptable.
And that's the problem. If this sort of thing starts to happen, what are you going to do about it? What will you even be able to do about it?
Of course these points are not allowed to be brought up in any "science" program in the public schools.
Flamebait and a troll. Some teachers might not know about this, or bring it up, but nobody's prohibited from bringing it up. In any class I had where radiocarbon dating was discussed, they also clearly described its shortcomings, and exactly why we don't rely on a single method of dating and use several overlapping measures instead.
Clearly this metric is biased against Sony. I mean, after shelling out that kinda dough for the console itself, of course nobody has enough money to buy more than 1.5 games.:)
I didn't even bother pre-ordering a Wii. Calling up a local Wal-Mart, I asked how many PS3s they expect to be receiving the day it is released. 10-12, was the answer. How many Wiis? Around 150.
Absurd. You can't expect to wield supreme executive power just 'cause some watery tart threw a sword at you! I mean, if I went around saying I was an Emperor because some moistened bint had lobbed a scimitar at me, people would put me away!
Regardless of your stance, I encourage everyone in the US to exercise their vote today, but only if they've actually given serious consideration to the issues and candidates they vote for.
Remember folks, just because there are many races on the ballot doesn't mean you have to vote in all of them. An uninformed vote is worse than no vote at all.
Slashdot Mirror
I'm actually on one of the teams (Georgia Tech), and lane following is a challenge. There are bigger ones in my experience, though. Following a road is not too terribly difficult, but it's the places where lane markings are faded or confusing (think intersections) that really provide the bulk of the problems.
The cars are going to have to be smart regardless. No matter what kinds of sensors are in the roads, there are going to be cars that aren't equipped with sensors, and have to be accounted for. Start including pedestrians, fallen trees, and any other unpredicted obstacles. Failures of the system would need to be recoverable too -- imagine what would happen if the sensors in the roads malfunctioned or went out. If all the cars were dependent on it, this would be a huge problem. But in a situation where the responsibility is on each car, at least other cars could avoid it. I can see a hybrid system taking place, but there will always need to be smarts built into the car itself (and probably a disproportionately higher portion, too).
Wikipedia link: http://en.wikipedia.org/wiki/Ernst_Dickmanns
We are. I haven't worked on that part of the car, but from what I've heard the SICKs combined saturate a 1GB ethernet link. One of the big challenges we're going to have to overcome for the car is data storage. For testing purposes, we'd like to record the output from all the sensors for individual runs. That way software can be tested offline. But with those consuming 1Gbps, the video cameras consuming about 750Mbps each, and other sensors, even storing the data becomes a huge task. We're probably looking at transfer rates somewhere around 3-4Gbps with everything said and done. Thankfully SAIC has offered to lend us a 1TB storage device we can throw in the car during testing runs.
Autonomous lane following by itself isn't actually even all that impressive. If I recall, there was a German who in the 80s set up a car to drive itself along the highway, with only minimal human intervention. It's interacting with other cars and navigating complex (i.e., non-highway) lanes.
The maximum speed any car will be allowed to do on the course is 35mph (might be 30mph, it's been awhile since I looked). There will be blocked streets so cars will have to replan their route, and we can probably assume DARPA is going to throw a traffic jam at us.
And of course, a lot of the purpose of this challenge is exactly what you stated in your post: we have to detect and avoid other cars, use safe and proper driving etiquette for passing others, follow the rules of the road (i.e., four-way stops, etc.), and dynamically adapt our course to the conditions. Chiefly, this last requirement means noticing obstacles (construction, accidents, etc.) and rerouting, but this could also incorporate predicting traffic jams. For instance, if a heavily-traveled section of the map has several intersections/stops, we might be able to predict this or at least notice it when we get close and take another path.
We don't have to deal with pedestrians at this point, but give it another generation or two. Right now it would probably be pretty hard to find any volunteers to do it. :)
I've posted elsewhere in this story, but again to prefix my comments, I'm a member of the Georgia Tech team, Sting Racing.
The course plotting part of the challenge is actually probably the easiest part. It's roughly analogous to you reading a map beforehand -- we're given a file detailing all the aspects of the course (course segments, how many lanes are in the segments, etc., and zones where free driving is allowed) plus a mission file giving the different waypoints we have to reach. This is, relatively speaking, easy.
The difficult part is determining where the edges of the road and its lanes are (GPS is terrible at this; most of the time it's accurate to 10 or so feet unless you're using extraordinarily expensive differential units) which is mostly done using visual scanning. Of course, we also have to detect other vehicles or obstacles in the path (using LIDAR and vision) and also determine the correct "pose" of the vehicle. Then we have to take that information and use it to modify the path we've already decided to take. These problems as it turns out are far, far harder than just plotting courses.
Botched the URL. It's http://www.sting-racing.org/.
To prefix my comment, let me just say that I am currently on the Georgia Tech team, Sting Racing. According to the rules, we will be on the course simultaneously with the other cars. The other other vehicles allowed on the course will be professional drivers. AFAIK, this is not being done in an actual city, but a small-ish mockup is being constructed for the purposes of the event. I could be wrong, though.
There's an RTC battery in there. I would know, I've opened the thing several times. Maybe yours is just drained?
I have to agree. It might not have been resolved this way, but he probably could have been better off trying to explain by explicitly mentioning the dollar sign. Basically explain to them that $0.002 is spoken as "0.002 dollars", and to reach cents from that figure requires dividing by 100.
Not saying anything about the number they get, then assuming there was a high number of direct attempts (i.e., 100,000) the attacks would likely have an even worse chance of working than if it were low. One of the primary reasons users are vulnerable to social engineering attacks is that they're rare (per individual). If this was something that happened routinely to every employee once every month or two, they'd probably be easy to spot. Of course, the additional volume might outweigh the drop in successes-per-attempt, but it's interesting to think about nonetheless.
Actually, just yesterday I managed to drop the outer shell when holding the rest of the case. The infamous ribbon cable pulled out of its holder, and I got quite upset. Most of the information I've read on the Internet previously had pretty much stated your PS2 was a brick if this happened. Not one to be discouraged, I found where the ribbon cable needed to be inserted, pried open the case even farther to get to it (requiring a total of 4 more screws to be removed), and got it back in. Reassembled the thing and it worked.
I don't know what your basis for comparison is, but PS2s are pretty damned durable. Yeah, they could be a little better about acknowledging the problems with the lens. But given some minor occasional maintenance, having a console last for 6 years is pretty impressive in my book. Additional kudos to them for making the whole thing easy to access and take apart. Considering the limited space requirements, it would have been quite easy for them to make the thing a tangled mess, and impenetrable to user efforts to repair it.
Clearly this metric is biased against Sony. I mean, after shelling out that kinda dough for the console itself, of course nobody has enough money to buy more than 1.5 games. :)
Give them a break, that's the encore :)
I didn't even bother pre-ordering a Wii. Calling up a local Wal-Mart, I asked how many PS3s they expect to be receiving the day it is released. 10-12, was the answer. How many Wiis? Around 150.
Absurd. You can't expect to wield supreme executive power just 'cause some watery tart threw a sword at you! I mean, if I went around saying I was an Emperor because some moistened bint had lobbed a scimitar at me, people would put me away!
Regardless of your stance, I encourage everyone in the US to exercise their vote today, but only if they've actually given serious consideration to the issues and candidates they vote for.
Remember folks, just because there are many races on the ballot doesn't mean you have to vote in all of them. An uninformed vote is worse than no vote at all.
FWIW, voting systems that aren't first-past-the-post don't necessarily have to also enforce proportional representation.
The one off Holcomb Bridge? Shit, they still having the sale? :)
Google's data center?