Carnegie Mellon University
Skip navigation and jump directly to page content

Aug. 3: CMU Contest Will Award Electric Car For Most Efficient Power Management Strategy

Contact:  Byron Spice / 412-268-9068 / bspice@cs.cmu.edu

Carnegie Mellon Contest Will Award Electric Car
For Most Efficient Power Management Strategy

Researchers Seek Ways To Maximize Electric Car Performance

PITTSBURGH—Researchers in Carnegie Mellon University’s electric car conversion project, ChargeCar, have announced a contest to find the most efficient methods for managing power in their electric vehicles. The grand prize of the ChargeCarPrize contest is an electric car.

“The number of variables that could possibly affect an electric car’s performance and the strain on its batteries is virtually infinite,” said Illah Nourbakhsh, associate professor of robotics and director of the Robotics Institute’s CREATE Lab. “Crowdsourcing is our best hope for sifting through those variables to find the optimal method for managing the flow of current between the motor and the power storage system. A contest seems the best way to draw a crowd and tap its wisdom.”

ChargeCarPrize is free and open to virtually anyone anywhere in the world. Contestants can download a software package, data files on driving behavior and some examples of power management policies from the project website, http://chargecar.org. Some basic knowledge of Java programming is required to encode a contestant’s power management algorithm, said Alex Styler, a robotics graduate student supervising the contest.

“But even if you don’t know anything about programming, you could learn what you need in a day,” Styler added. “The most important quality for winning this contest is intuition, not programming skill.”

ChargeCar is a community-centered research project of the CREATE Lab that seeks to revolutionize commuting with electric vehicles. Working with Pittsburgh-area mechanics, researchers are developing methods for converting gas-powered cars into affordable electric vehicles practical for commuting. The electric cars will employ artificial intelligence to manage their power, thus improving efficiency and reducing wear-and-tear on batteries.

In addition to batteries, the ChargeCar project will use supercapacitors as part of each vehicle’s power system. While batteries have trouble handling the surges of power associated with starting or stopping a vehicle, supercapacitors can rapidly discharge current for acceleration and can rapidly store current produced by regenerative braking. Supercapacitors can thus improve vehicle performance, while reducing wear-and-tear that can dramatically shorten the life of expensive batteries.

The trick, Nourbakhsh says, is finding the best way to handle the flow of power between batteries, supercapacitors and motor. When should power be transferred from battery to supercapacitor, or vice versa? When should the motor draw power from the battery? When should it draw from the supercapacitor? To make these decisions, the system can rely on knowledge of the driver’s normal routes, driving habits, traffic conditions, road conditions, geography, time of day and even weather conditions.

“Any source of information is fair game for people designing power management policies, as long as the source is publicly available,” Styler said.

The contest includes eight “secret drivers,” whose detailed driving histories will be recorded each month. Contestants will upload their power management policies to the contest site by the end of each month; the policies will then be tested against the data from the secret drivers. The top scorers will be listed on an online leader board; the leader atop the board each month will win a $250 gift card and the top four scorers will be eligible for the grand prize.

All entries in ChargeCarPrize will be open source and will be publicly available online each month, so contestants can learn from each other as they try to improve their algorithm’s performance. Data from the secret drivers also will be published online each month.

The length of the contest could be as long as 18 months, Styler said, but it could be terminated earlier if it appears that an optimal algorithm has been found. Contestants will be given advance warning of the contest’s termination and all contestants who have qualified for the grand prize will be given the opportunity to submit a final algorithm.

The ChargeCarPrize winner will receive an electric vehicle, though Styler doesn’t know what specific vehicle that will be. “If the contest runs long enough, we may have time to build one of our ChargeCars as the prize,” he said. “Otherwise, we’ll obtain another electric vehicle as the prize.”

Rules and other information about the contest are available on the ChargeCar site, http://chargecar.org.

The ChargeCar project is sponsored by Carnegie Mellon alumna Donna Auguste and her husband David Hayes, Google Inc., The Heinz Endowments and Bombardier Inc.

Follow the School of Computer Science on Twitter @SCSatCMU.

###