MIT Researchers Fight Gridlock with Linux

nerdyH brings us a report about a Linux-based device being developed at MIT which aims to reduce traffic congestion as well as assist automotive research projects. "The current focus of the project is in developing algorithms that run on top of the portal application to help drivers plot the best route at a given time. For example, the team's MyRoute project includes applications that model delays observed on road segments as statistical distributions. Various algorithms then use these to compute optimal routes for different times of the day. 'Instead of asking the shortest time or shortest distance from point A to point B, you ask what route should be taken, say, for the highest probability of getting to the airport by a certain time depending on the time selected,' says Madden."

Comments

No live data?

[timeOday]

Within a few years I don't think we'll just be using statistics of past data, but rather real-time traffic data from cars that link into a real-time network. All it will take is a certain density of smartphones with GPS.

Re:No live data?

[farkus888]

that may take longer than you are guessing. mostly because uptake of smartphones is going to be quicker among the technically elite. judging by slashdot posts that particular crowd seems to dislike the idea of someone tracking where they are all the time. people who jump on this are probably not going to be getting gps phones till the free phone from their carrier has it. or you could go with the assumption that what is posted is likely far from what these people do in their real lives... for example how many of these privacy nuts use gmail?

Re:No live data?

[Jeff DeMaagd]

GPS isn't necessary. E911 location is probably "good enough" for this type of location and it's in every phone right now. I woudn't go for it because both systems suck down batteries. Another reason I wouldn't want to participate in a program that would help the Department of Fatherland Security anyway, they'll probably find some way to get unfettered access to the information.

Re:No live data?

[ptbarnett]

Within a few years I don't think we'll just be using statistics of past data, but rather real-time traffic data from cars that link into a real-time network. All it will take is a certain density of smartphones with GPS. The data is already available in many metropolitan areas:

http://www.xmradio.com/navtraffic/market_coverage.xmc

The methodology for gathering the info varies, but in Houston they use the electronic tolltags, even on roads that are not tolled. By measuring the time between stations, it calculates the average speed of cars on that segment.

The results are downloaded into the navigation system in my car, and depicted as green, yellow, or red bars adjacent to the route. However, I've never been able to determine if the GPS routing uses the speed information to calculate the fastest route, as I don't live in an area for which speed information is available. A couple of years ago, it was reported on Slashdot that Baltimore was going to test monitoring of traffic speeds using (presumably generic) cellphones. This article, although somewhat dated, reports initial results and also notes that the state of Virginia is doing the same thing.

Re:No live data?

[rvw]

In the Netherlands TomTom and Vodafone are experimenting with this. They track the movement of mobile phone users along the highway. That way they can see how the traffic is moving. You get the data on your TomTom if you take a subscription.

Re:No live data?

[funwithBSD]

I wonder.

I lost my cellphone this weekend, and I asked my provider: "Well, can't you track it with the built in GPS and tell me where I left it?"

Guess the answer.

So how do you get the data off if is not easily accessible by the provider?

(I entertain the possibility that they don't want to do that so I have to replace the phone)

Re:No live data?

[buravirgil]

Reactively re-routing will alleviate congestion, but as the number of users approaches 50%, this strategy diminishes in its return. Only a minority can benefit from reaction to a majority. Early adapters, People-In-A Hurry-with-the-Money, would experience benefit while the congestion, overall, is lowered...so I figure this plan has great appeal-- bureaucrats can insist on its necessity and their monied constituency can avoid the vulgate. Smarter routing Vs. Wider roads...

Re:No live data?

[CastrTroy]

Meh. Less people driving would probably be an equally good solution. More people carpooling or using public transportation would work a lot better. A good public transit system can get rid of a lot of cars on the road, and clear up a lot of traffic. The problem with most US cities, is that their public transportation services are atrocious.

Re:No live data?

[toddestan]

It could tell you if it was left at work, at home, or at a friend's house, and so forth. It would be atleast a start, if you have no idea where you left it.

here's an idea

[ImTheDarkcyde]

Eliminate green arrows from 12am-5am! As an overnight worker I can not tell you how many times I've had to sit at a red light twice because the sensor didn't want to trigger for just one car - and I know the one time I decide to go on a red arrow there is going to be a cop right around the corner.

Real time rerouting for evacuation

[Geoffrey.landis]

I wouldn't want to participate in a program that would help the Department of Fatherland Security anyway, they'll probably find some way to get unfettered access to the information. Hmmm.... Trying to think of a rationale that might be used for why Homeland Security would have to want to look at the traffic jam data, they could say that they need the data to practice so that, if it ever happens that a disaster means that they need to evacuate a city, they could develop a system that would allow them to do wo without clogging the streets.

But actually, that would be a good us of the system; if there ever were an evacuation, it would be useful to have a system to reroute around the inevitable traffic jams...

I wrote about this idea last year it's about time

[FromTheAir]

It is about time we applied techonolgy to solve the worlds problems. What if all the business leaders coordinated to shift their employee arrival and departure times slightly so as to create a well orchestrated movement of workers? With a little funding I could build a system to coordinate the daily rush hours. The system would take into account departure points and destinations, the course and various elements on the course, and perhaps the best course, preferences and people would select a travel spot. Even better all the cars can communicate with each other, knowing their drivers intended destination an artificial intelligence could coordinate suggesting the best time to move into the middle or left lane, sending some cars down an alternate route to relieve some congestion which in the end results in everyone getting home quickly which is the incentive to participate. How many billions of gallons of gasoline could be saved if we simply knew the most energy efficient route especially when driving in unfamiliar territory? Far more money would be saved than the cost of the system and the benefits would be extraordinary.

density, GPS units, and stupid traffic lights

[SuperBanana]

Problems currently exacerbating traffic in Boston:

Density. There are simply too many cars for the roads. Public transit is partly to blame; it's gotten better and more convenient (the Charlie Card was a major improvement- people can board busses *really* fast now, and I'm seeing fewer and fewer here-is-what-I-found-in-my-change-jar types), but the T still has miles to go in terms of reliability, routes, and just plain cleanliness.

GPS units (and Yahoo/Mapquest/Google maps) which go for the shortest/fastest route, not the *best* route or route with the most 'bandwidth'. For example, it's technically shorter to cut *through* longwood medical area, but it's much faster to go around it- even though there are more traffic lights, they're all in your favor.

Stupid traffic lights. There's a major intersection near me where, despite the complexity of the intersection (five streets), it's not wired with pads to tell how many cars are where. One car on a tiny side-street fucks up traffic on roads which are classified as arterial (y'know, the ones you can't park on during a snow emergency.) Boston and surrounding cities have hundreds of such intersections- but you'll only find the "smart" ones where rich people live.

Neat idea but...

[blacklabelsk8er]

I think the EasyWiFi technology that they had to develop in order to get this to work is actually the cooler aspect of this project, especially since its compatible with normal WiFi hotspots.

I could see it becoming a handy addition to any Wifi setup for a mobile device.