MIT Algorithm Predicts Red Light Runners 348
adeelarshad82 writes "Researchers at MIT have developed an algorithm that determines which drivers will run a red light, within one to two seconds before a potential collision. The research, based on 15,000 cars at a busy intersection, monitored various factors to determine which cars were were likely to run a red light. They found that their predictions were correct about 85 percent of the time, which is about 15-20 percent better than existing traffic prediction algorithms."
Re:Just a matter of time... (Score:4, Interesting)
Nah. That'd cause a car to stop dead in the middle of the intersection and make a bloody mess of traffic.
Instead, you want a smart deployment of spikes which puncture only a single tire. More than enough to seriously piss off a guy running the light, while still letting him limp out of the way to the shoulder.
Re:Where's the Work? (Score:5, Interesting)
...As for the 15% error, did anyone consider cargo?...
THANK YOU!!! As an economic refugee of the "Great Recession", I ended up driving a tractor-trailer for a living - and wound up learning a few things along the way. One interesting fact I've learned is that a fully loaded (80,000 lbs) semi moving at 55 mph can take up to 300 ft to come to a complete stop (think about that next time you want to "brake-check" a truck...). I have, unfortunately, run across traffic lights in which the yellow phase was, for some strange reason, really short- even if the the semi is traveling the legal speed limit. This is not a situation you want to be in: your choices often boil down to:
(1) Stand on the brake in order to not run the impending red light (remember that 300-foot stopping distance? By the time you get stopped, your trailer in squarely in the middle of the intersection. And that's if you don't jackknife and end up wiping out 5 or 6 cars along the way).
(2) Run the light (Yes, it's going to be red by the time you hit it, meaning you will almost certainly incur the wrath of any red-light camera or nearby cop - but see option 1 for the alternative scenario)
This is probably the number two reason I try to avoid surface streets when possible (reason number one being the preponderance of infrastructure not exactly designed with a 75-ft long, nearly 14-ft high vehicle in mind). I figure any traffic engineer worth his salt is going to take these factors into consideration; a failure to do so is going to inevitably invite the occurance of an 18-wheeled clusterfuck and all that comes with it (major property damage, potential loss of life, etc).
Re:Why just predict, when you can prevent? (Score:5, Interesting)
That's a pretty big assumption. The advantage this thing has is that you can outfit the traffic lights to delay the green light while the runner crosses and prevent the problem without instrumenting every car on the road.
And ticket the jackass who might have killed someone.
Re:Just a matter of time... (Score:2, Interesting)
Re:Just a matter of time... (Score:1, Interesting)
What number of accidents are caused by red light runners? I'm talking the red light runners that squeeze through just after the light changes, not ones that run it 3-60 seconds after the change. Those two types are in different classes. The first just didn't want to stop and the oncoming traffic is probably not coming yet and the later was not paying attention and had some type of lapse when driving, similar to forgetting to stop and rear ending someone, drifting off the road, drifting over the line, blowing a stop sign etc... no amount of tickets, fines, signs, or warnings can predictably lower that from happening.
I've been driving for 25 years and just about the average amount per year as everyone else. I have NEVER seen an accident caused by a red light runner who "just missed" the light, I've seen and avoided several by idiots that we not paying attention and blew through even a few minutes after the light was red because they were not looking. Given my impression, this red light camera is nothing more then revenue generator and not about safety at all. Same as the police hanging out with speed detectors on wide open roads that have some unreasonable low speed limit, like a four lane divided road with a median and no houses or access roads around but the speed limit is 35. They are not there to protect the "kids" from getting hit because there are no kids there, they are there because they know they can get a bang for their buck catching people. Definitely not about public safety, it is about police funding.