Ford is joining Google to develop cars smart enough to know where you’re going and how best to get you there so you aren’t wasting time in traffic.

The automaker is tapping the power of Google’s remarkable Prediction API to create cars that determine where we’re going by examining where we’ve been. It’s the latest example of the auto industry calling on Silicon Valley as software takes on more tasks in our cars. It also underscores the tech sector’s growing interest in the automotive sector as Microsoft, Cisco and IBM develop new technologies and markets.

Ford has been way out in front in this arena since introducing Ford Sync in 2007, and the Google partnership announced today at the tech titan’s I/O developer conference could strengthen its lead.

In a nutshell, the Prediction API provides pattern-matching capabilities to existing cloud-based data sets, using it to “predict” probable outcomes for current events. Using it in an automotive setting would allow our cars to effectively learn from our behavior and adapt to it.

“We take examples from the past and apply machine algorithm learning to find the best model that fits the situation,” Travis Green, Google’s Prediction API product manager, said.

So in theory, if you opt in to using the service, a computer in your car would create an encrypted record of your driving data — where you’ve gone, when you’ve gone there and what route you took — and tailor it to your driving profile. Eventually, your car “remembers” where and how you drive.

Using that information, it can tap the Google Prediction engine to analyze huge data sets and get a sense of where you might be headed when, say, you start the engine at 5:15 p.m., then suggest an ideal route to avoid congestion or other headaches.

But that’s not all. The car would then optimize itself for the trip — Ford didn’t say exactly how — to adjust the powertrain and improve fuel efficiency.

“Once the destination is confirmed, the vehicle would have instant access to a variety of real-time information so it can optimize its performance, even against factors that the driver may not be aware of,” said Ryan McGee, who works within Ford’s vehicle controls architecture and algorithm design division. “This information can ultimately be used to optimize vehicle performance attributes such as fuel efficiency and driveability.”

It’s all still highly theoretical, and still in the preliminary stages of R&D. But Ford is applying the Google Prediction API to two years of in-house driving data. It almost certainly will be years before we see cars with this sort of technology, but Ford says it is committed to the project.

“Ultimately, we’re looking to make cars smarter,” McGee said. “We’re looking to make them more efficient.”

Image: Ford