Developing more advanced intelligent transportation systems

You are sitting at a traffic light waiting for it to turn green and see the light just ahead turning red. You groan to yourself, "Why can't they talk to one another? Why can't they see how infuriating it is to blindly switch signals, ignoring the flow of the traffic they are supposed to regulate?"

Well, maybe you are about to groan no more.

If everything works according to plan, the University of Toronto's Intelligent Transportation Systems (ITS) will soon create what civil engineer Baher Abdulhai, head of ITS, describes as "a traffic light that learns what to do based on its own experience."

"It's like having a Robocop standing there, someone who's been doing this for 30 years and can do it while asleep," remarks Abdulhai, whose own first traffic experience was with the hypercongestion of his native city, Cairo.

"Because the lights are experienced, when the environment changes - say, all of a sudden a new mall or a new condo gets built - they can say to themselves, 'Okay, this is not what I knew before.' So the Robocop traffic lights start updating their knowledge and improving their functions in the field. And they do this without an engineer or a traffic specialist giving them advice about what they should do."

Abdulhai, who has a PhD in intelligent transportation systems from the holy land of the discipline, the University of California at Irvine, is optimistic that we will see such artificial intelligence lights in the next five years for three reasons. For one, the first element of a self-educating device - a diversion control monitor that tells drivers when conditions such as traffic or an accident suggest they should get off a freeway and travel instead by alternative routes - has been developed and lab-tested and is ready for field testing.

Secondly, and just as important, the data to determine what is and isn't working are already being generated at the University of Toronto ITS centre, in collaboration with the Ontario Ministry of Transportation and the City of Toronto. This will give a sense of what to do next.

Thirdly, and right now, fibre optic cables feed real-time data from traffic detector loops buried in roads all over the Greater Toronto Area. These loops can be linked with traffic videos from cameras spread throughout the GTA.

The results are displayed on 20 screens that monitor traffic in different parts of the city. The virtue of the two-pronged monitoring technique is that if one of the detectors indicates that something is amiss, researchers can then switch to the video feed and view the ebb - and often not flow - of the congestion.

And, in the future, U of T's ITS will travel down bigger and faster roads. The institute has just received funds to create an information network in which the data being gathered in Toronto and other major cities around the globe are made available over the Internet to 14 universities through a virtual network known as ONE-ITS. The idea is to create a platform where people everywhere can use the diverse data collected at U of T and elsewhere to design a host of new technologies.

"We want to make all the resources available to everyone so other people can create newer applications we haven't thought of," says the man who has been jokingly labelled a "roads scholar."

He imagines a software package where somebody has configured traffic patterns leading to different Toronto golf courses and has linked them together with data showing how busy courses are and, in so doing, indicating the total time it will take to play a round.

Impressive, yes. But equally impressive are the centre's recent efforts to expand the notion of intelligent transportation systems to vehicles without engines.

One of Abdulhai's graduate students is working on a Toronto area mapping project aimed at cyclists. The application calculates street grades linked with GPS devices. This information would direct cyclists to routes where they have to climb the fewest hills to get where they want. Consider it a path of least resistance, if you will.