By Michael Kanellos, Greentech Media
In the future, your plug-in hybrid may be able to read your mind — and save you gas in the process.
Ford, which has been at the forefront among auto makers when it comes to integrating software applications into cars, is experimenting with an application that aims to leverage data amassed about your driving habits to increase the mileage of plug-in hybrids and standard hybrids.
Extending car mileage has become job number one at automakers now that the U.S. plans to raise the fleet vehicle average standard to 54.5 mpg by 2025. Ford and Toyota announced Monday that they will collaborate on hybrid technology for pickups and SUVs, and also work together to develop common standards for in-car Internet connectivity.
While the two companies focused Monday’s announcement largely on gas-electric trucks — Ford is No. 1 in trucks and Toyota is No. 1 in hybrids — software could become a more pervasive part of the relationship because it will become a crucial part of most new cars.
Ford’s latest project, codenamed Green Zone, taps Google’s remarkable Prediction API to create software that can determine where you’re going by examining where you’ve been. Say it’s 8 a.m. Tuesday. Your car knows it is the second in a five-day sequence in which you drive 23.5 miles to the same destination. The software crunches data about your driving habits, the topography of the drive, details about traffic and time-to-destination and information about how the car performs.
Then it attempts to maximize the power drawn from the battery pack and minimize use of the gasoline engine. When you get within a certain number of miles of your likely destination, you enter, from your car’s perspective, a “green zone.” Beyond that, it might go fully electric.
Further, if you typically charge up after the morning drive and don’t leave again until 5:00 p.m., the car can try other methods to squeeze out a few extra miles on electricity.
“We have this massive amount of data. The question is what to do with it,” said Ryan McGee, technical expert on vehicle controls architecture and algorithm design at the company.
The car continually refreshes its data by interaction with cloud-computing services.
The probabilistic principles underlying the experiment are similar to predictive algorithms exploited by search engines. In addition to using Google’s predictive APIs, Ford also works with Microsoft on in-car telematics and, like General Motors and others, has developed applications that let EV owners interact with their cars.
Quite a bit of work remains. Right now, Green Zone is in the proof-of-concept stage. The software designers also have to figure out ways to readjust predictions in the face of surprises.
Via Wired Autopia: http://www.wired.com/autopia/