2011年4月15日星期五

Smartphone forecast traffic jams

A Smartphone application can be user-defined alerts people warning get probably traffic trouble head. The software can predict road conditions 35-40 minutes in the future.

IBM tests Smartphone software developed to predict traffic jams and driver warning, before taking to the streets.

IBM said late Tuesday that its staff in the areas of San Francisco and Silicon Valley of Northern California have tested technology, that "will driver help ultimately all over the world," avoid polluted traffic.

Location detection automatically track involved in the pilot project of votes functions in their smartphones, where to go and when, according to IBM, smarter traveler have Program Manager John day.

The information is fed across the Internet in computers, the pattern such as commutes to the and to identify from the workplace.

Now is roadway sensors commonly used for online road maps analyzes collected data to determine conditions that usually cause difficulties.

For example, certain off-ramp or bridge consistently to secure entrance in a different area of traffic can result in traffic jams on one.

The results are combined to personalized forecasts of when a motorist apt run in highway headaches is to make.

"We wanted to provide analytics for predictive capabilities use;" That happen then get, "day said correlations with smaller slowdowns and large, AFP."

"So can you run a query at any time for a trip and predictions of 35 or 40 minutes in advance then link it looks like you, with a personal approach for the individual traveller."

IBM researchers worked with California State Highway authorities and a mobile Millennium team at the University of Berkeley, California in the project.

The Smartphone application can people custom alerts get traffic warning probably anger before it swings or other routine drives.

The service is powered by a "first-of-its-kind learn and predictive analytics tool" referred to the traffic prediction tool (TPT) from IBM research developed.

TPT analyzes congestion data continuously, commuter of locations and expected travel start times during a metropolitan area, which can affect commuters on the highways, railway lines and city streets.

"The idea is to learn, a traveller habits, they lead to the prediction model then see what they can expect traffic" day said.

"The goal was to make personal and equipped it much more it leave you just before they were."

IBM researchers present, integration of real time data from the bus and train systems in the equation, so that the service people could advise if it would be wiser, distract to public transport.

Privacy protection include start and to manage your dates online endpoint of the TRIPS agreement by the encryption as a people.

The pilot project has about five months go, on.

"The prediction opportunities are head and shoulders above what it is today," said day. "Everything out there shows you traffic as five reported or 10 minutes." "No one does predictive."

During tests in California, IBM is intent on building a system that can work around the world.

"In contrast to existing traffic alert solutions, we help the guesswork of which commute, take", said Stefan Nusser of IBM Almaden Research services.


View the original article here

没有评论:

发表评论