Engineers at UC Riverside have developed State of Mission (SOM), a new metric to improve battery life predictions in electric vehicles by considering environmental factors. This innovation promises more accurate and reliable battery management.
Current battery management systems often leave electric vehicle drivers guessing if their vehicle’s remaining charge will cover the distance, especially under demanding conditions. Engineers at the University of California, Riverside aim to eliminate this uncertainty with their innovative diagnostic metric, State of Mission (SOM).
SOM is designed to provide real-time, task-specific predictions by incorporating both battery data and environmental factors like traffic patterns, elevation changes and ambient temperature. This means instead of a simple percentage display, drivers receive actionable insights about their battery’s capability.
“SOM fills that gap,” Mihri Ozkan, an engineering professor at UC Riverside who helped develop the system, said in a news release. “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.”
The breakthrough, published in the journal iScience, employs a hybrid approach.
Traditional battery diagnostics rely heavily either on rigid physics equations, which often fail to adapt to changing scenarios, or on opaque machine learning models.
SOM combines the flexibility of machine learning with the foundational principles of electrochemistry and thermodynamics.
“By combining them, we get the best of both worlds: a model that learns flexibly from data but always stays grounded in physical reality,” added Cengiz Ozkan, a UC Riverside engineering professor who co-led the study. “This makes the predictions not only more accurate but also more trustworthy.”
To validate their framework, the team utilized publicly available battery datasets from NASA and Oxford University, which included a vast range of real-world usage patterns, temperature fluctuations, current and voltage data, and long-term performance trends.
The results showed significant reductions in prediction errors compared to traditional methods.
Instead of a basic “percent charged” estimate, SOM offers more advanced, forward-looking insights. For instance, it can inform a driver if they need to recharge midway through their journey or indicate that a drone flight is unfeasible under certain conditions.
“It transforms abstract battery data into actionable decisions, improving safety, reliability and planning for vehicles, drones and any application where energy must be matched to a real-world task,” Mihri Ozkan added.
Despite its promise, the model’s current computational complexity exceeds the capabilities of most existing embedded battery management systems.
However, the researchers remain optimistic, expecting that with further optimization, SOM could soon be integrated into various applications such as electric vehicles, unmanned aerial systems and even grid storage solutions.
“Right now, the main limitation is computational complexity,” Mihri Ozkan said. “The framework demands more processing power than today’s lightweight, embedded battery management systems typically provide.”
Looking forward, the team is planning field tests for SOM and hopes to expand its applicability to different battery chemistries, such as sodium-ion, solid-state or flow batteries.
“Our approach is designed to be generalizable,” added Cengiz Ozkan. “The same hybrid methodology can deliver mission-aware predictions that improve reliability, safety and efficiency across a wide range of energy technologies from cars and drones to home battery systems and even space missions.”