Wall Street has been using algorithms for years
to trade the market. Not likely an algorithm will prevent
a stock market crash however,
M.I.T. might reduce intersection crashes.
The New York Times
By: Paul Stenquist
11/30/2012
A green light at an intersection may offer a driver the legal right to proceed, but it offers no guarantee of safe crossing. Researchers at the Massachusetts Institute of Technology have developed an algorithm to predict when an oncoming vehicle would be likely to run a red light.
With the proliferation of spatial sensors, advanced telematics systems and other hallmarks of so-called smart cars, algorithms could help prevent drivers from getting into accidents, according to Jonathan How, a professor of aeronautics and astronautics at M.I.T.
Data from the National Highway Traffic Safety Administration showed that of 7,000 traffic-related fatalities in 2008, 700 of them were the result of drivers running red lights at intersections, according to a media release from M.I.T. Half of those killed were not the drivers of the offending vehicle, but were passengers, pedestrians or the driver of another vehicle.
Developed by Georges Aoude, formerly a student of Professor How, the M.I.T. algorithm used parameters like the vehicle’s rate of deceleration and distance from the traffic light to determine which vehicles were most likely to run the light. If the algorithm were to be put in place, that data would be generated by means of vehicle-to-vehicle communication, a smart-car technology being researched by the Transportation Department in conjunction with a number of vehicle manufacturers.
“If you had some type of heads-up display for the driver, it might be something where the algorithms are analyzing and saying, ‘We’re concerned,’” Professor How said in the release. “Even though your light might be green, it may recommend you not go, because there are people behaving badly that you may not be aware of.”
The algorithm was tested at an intersection in Virginia. Applying it to data from more than 15,000 approaching vehicles, the researchers observed that the algorithm accurately identified red-light violators 85 percent of the time. The identification was achieved a couple of seconds before those vehicles reached the intersection — sufficient time, the researchers said, to allow drivers to react.
The research team is investigating ways to use the algorithm in a closed-loop system that would provide a recommended response to the driver whose vehicle may be in danger. They also plan to modify the algorithm to test its efficacy in air traffic control applications.
Until M.I.T. perfects their algorithm,
BE ATTENTIVE when entering all intersections.
Assume Nothing and Stay Alive!