Professor Shekhar Collaborates with U of M Engine Laboratory on NextGen Automobiles
Distinguished McKnight Professor Shashi Shekhar is part of an interdisciplinary University of Minnesota team that has been awarded $1.4 million in funding from the NEXTCAR Program of the U.S. Department of Energy’s Advanced Research Projects Agency-Energy (ARPA-E) to research ways to boost efficiency of connected and automated vehicles. This project builds on Shekhar’s current grant from the National Science Foundation for investigating spatial big data for next generation routing services.
The University of Minnesota NEXTCAR researchers are partnering with UPS and hybrid-electric vehicle manufacturer Workhorse to improve the energy efficiency of medium-duty delivery vehicles through real-time powertrain optimization using two-way vehicle-to-cloud connectivity.
UPS has more than 100,000 vehicles that drive millions of miles per year. “Small changes can make a big difference in vehicle efficiencies,” said Will Northrop, the principle investigator of the ARPA-E grant and the director of the Murphy Engine Research Lab and mechanical engineering associate professor. “For example, in the past the company used big data analytics to study left turns. Developing routes for the delivery vehicles that avoided left turns saved 3 million gallons of fuel and 32,000 tons of carbon dioxide in one year,” he added.
Large delivery fleet operators already use extensive data analytics to assign routes for minimizing energy consumption. The project team will further improve the energy efficiency of their series hybrid-electric vehicle by optimizing battery state of charge and engine operating strategy in coordination with intelligent eco-routing. The overall goal of the research is to reduce energy usage in the Workhorse-produced UPS vehicles by 20 percent.
“These engines generate a very large amount of data. There are hundreds of variables measured every second,” said Professor Shekhar. “If we can analyze the data, we can find very interesting and useful patterns, which can help us reduce fuel consumption and emission.”
There are many ways to exploit the patterns in the vehicle on-board diagnostics data stream to reduce emissions and improve energy efficiency. One may compare alternative routes to select one minimizing energy consumption under an emissions budget creating next generation routing and navigation apps, which currently recommend routes minimizing travel time or distance. the vehicle computer may proactive control the power train based on the route ahead rather than react to current situation (e.g., hill climbing). Furthermore, one may co-optimize both powertrain control and route selection. Using cloud connectivity, the vehicle may also periodically download the most-efficient powertrain calibrations based on external data like traffic and weather collected while the vehicle is en-route.
“From all perspectives—if you look at business, consumer, or society, there is a lot of crossover potential in this project,” said Professor Shekhar.
For example, the European Union (EU) started the optiTruck project this year to reduce greenhouse gas emissions from heavy-duty vehicles by developing a prototype to demonstrate a fuel reduction of at least 20% while meeting Euro VI emission standards. This follows the EU REDUCTION project to reduce environmental footprint based on multi-modal fleet management systems for eco-routing and driver behavior adaptation. A recent McKinsey report on big data estimated that use of personal location data could save consumers worldwide more than $600 billion annually by 2020, with the biggest single benefit coming from fuel savings by avoiding congestion and choosing alternative routes.
To learn more, visit the Murphy Engine Research Lab website.
Adapted and reprinted with permission from the College of Science and Engineering.