It’s said that the average car today has more computing power than the rocket that put a man on the moon.
A new research project by IBM and McMaster University in Hamilton, Ont., is working on the theory of a “cognitive car” that will connect the vehicle’s multiple microprocessors, which currently work in isolation, to provide important information for the driver.
Such a car might predict vehicle failures before they happen, save fuel by sending the driver on less-congested routes, or even “talk” to other cars behind it to warn about hazardous road conditions up ahead.
“The average car has a minimum of 20 microprocessors, all connected with multiple networks,” says Don Aldridge, general manager of research and life sciences at IBM Canada.
“The objective is to eliminate as many of the separate devices as possible and consolidate them into one, if we can.”
The research includes a grant award from IBM and is intended to study how using a single IBM multi-core processor could improve vehicle efficiency and safety. The processor, originally developed for video games, is now used for a wide range of applications.
As mainstream electric cars get closer to their market launch, such microprocessors could be used to help drivers charge their cars at optimum times.
“One of the challenges of electric vehicles will be fear of running out of juice,” Aldridge says.
“By tying into a simple GPS requirement, it would give options for charging wherever I happen to be. You also start tying into the system — am I potentially charging at a peak time, or not? The system should be smart enough to give you guidance and minimize the cost.”
A car connected in such a way could know its location and how much power is left in its battery, tell its driver where the nearest charging station is located and, depending on whether prices are peak or off-peak, how much it would cost to recharge.
The microprocessors could also simplify car-to-car communication.
“At the moment, you have an automobile that’s talking to the GPS system,” Aldridge says. “It could also communicate with automobiles behind it, so that you’d know that black ice is up ahead because the cars share that with each other.
“The research is just beginning and we are just at the early stages, but now we’ve started the process, and it will continue to evolve. When you start going from dozens of little computers all over the place to smaller numbers, and possibly one, it could reduce the price.”
