3 February 2009—Drivers use all manner of data these days to travel efficiently, and vehicles should follow their lead, according to University of Wisconsin–Milwaukee mechatronics expert Yaoyu Li. He predicts that vehicles privy to data in the latest GPS-enabled electronic navigators—which download real-time traffic data to update route suggestions on the fly—will provide substantial fuel savings in the decades to come.
Li has developed control algorithms that use route and traffic data to allow hybrid vehicles to plan how and when to use stored battery power so as to burn as little gasoline as possible. He hopes to enhance the plug-in hybrid vehicles already in development at major automakers, such as a grid-chargeable Prius that Toyota plans to lease starting later this year, or the Chevy Volt that General Motors promises for late 2010.
The idea has merit, according to Tom Robinson, senior manager, control and electronics, at automotive-systems supplier Ricardo, in Shoreham-by-Sea, England. ”If you know what’s forthcoming, you can inform vehicle systems to operate more effectively,” says Robinson, whose company has worked on similar algorithms for conventional hybrids.
An uninformed plug-in is almost certain to discharge its battery power either too quickly or too slowly. If it simply uses the battery until it is discharged, it will lack an electric option for later stop-and-go situations where running the internal combustion engine is inefficient. Alternatively, if the plug-in acts like a conventional hybrid and lives in the moment, blending its electric and gasoline energy based on the driving conditions that second, it is likely to arrive at its destination with leftover battery charge. Either way, the plug-in will have consumed more gasoline than necessary.
Li’s algorithms use data from electronic navigators to optimize the mix of combustion and electric propulsion to suit the trip. First, one algorithm cuts the driver’s chosen route into segments and, based on traffic data and the plug-in’s current state of charge, predicts how the vehicle should balance its use of electricity and gasoline in each segment. Li likens this macroscale algorithm to his family’s monthly spending plans for the year ahead. A microscale algorithm then takes over to make en route adjustments, much as Li might adjust his spending if friends pay a surprise visit to Milwaukee and blow his budget. ”As the vehicle approaches the next route segment, I use my current state of charge as a start point to solve a new optimization problem. I’m trying to force my actual expenditure toward my preplanned budget,” explains Li.