You’ve undoubtedly seen someone—or been that someone—walking along carrying a gas can having run out of fuel. What happens when an EV is out of energy? This project is working to keep that from being a question that needs an answer. . . .

Apparently, the battery management systems in EVs are little better than crap shoots when it comes to providing accurate information as to whether there is sufficient charge in a battery to get from point A to point B.

One reason is there are lots of parameters that need to be taken into account in order to make the assessment, ranging from the condition of the battery to its discharge rate, from the traffic between A and B and the ambient temperature.

To address this overall uncertainty, engineers at the University of California, Riverside (UCR), have developed a metric, State of Mission (SOM). Mihri Ozkan, a UCR engineering professor described it: “It’s a mission-aware measure that combines data and physics to predict whether the battery can complete a planned task under real-world conditions.”

Some models use physics equations that don’t deal with changing environments. Some use machine learning models that provide output but no information regarding how that output was achieved.

The SOM takes into account both information from the batteries’ performance over time as well as electrochemistry and thermodynamics.

Cengiz Ozkan (also a UCR engineering prof): “By combining them, we get the best of both worlds: a model that learns flexibly from data but always stays grounded in physical reality. This makes the predictions not only more accurate but also more trustworthy.”

But there is still a hurdle that the UCR team needs to get over as the SOM development continues.

“Right now,” Ozkan said, “the main limitation is computational complexity. The framework demands more processing power than today’s lightweight, embedded battery management systems typically provide.”

It always comes down to sufficient power, one form or another.