The system, dubbed State of Mission (SOM), goes beyond traditional battery status checks. Instead of just estimating how much energy remains, it calculates whether a battery can successfully complete a task under current and forecasted conditions. That means factoring in terrain elevation, ambient temperature, traffic data, and even wind resistance—all of which can significantly affect energy consumption.
In real-world terms, this could help logistics companies plan mid-route recharges in advance, notify drone operators of flight risks due to wind or temperature, or allow autonomous vehicles to plan routes with higher reliability—reducing the risk of operational failures due to energy miscalculations.
The foundation of SOM lies in a hybrid diagnostic model that blends machine learning with physical battery models. While traditional systems typically rely on either static physics-based calculations or opaque machine learning models, UC Riverside’s approach uses both.
By combining the flexibility of data-driven models with the grounded accuracy of electrochemical and thermal laws, the system remains reliable even when conditions change unexpectedly—something conventional models struggle with. In testing, SOM outperformed existing diagnostic tools, improving accuracy in voltage (by 0.018V), temperature (1.37°C), and state of charge (2.42%).
This dual-layered framework allows for adaptive recalculations in real time. If a truck enters a colder climate or a drone experiences higher-than-expected headwinds, the system reassesses whether the battery can still complete the mission. That shift from static monitoring to predictive planning is where the real innovation lies.
Despite its capabilities, the SOM system is not yet plug-and-play for commercial settings. The current version requires more computing power than typical embedded battery management systems. That makes widespread deployment in low-power environments—like drones or small EVs—a challenge for now.
However, the team is actively working on optimizing the system for efficiency and expanding its use beyond lithium-ion batteries to include emerging technologies like sodium-ion, solid-state, and flow batteries. Long-term plans include field testing and collaboration with manufacturers to bring SOM into commercial battery management platforms.
If successful, this mission-aware approach could provide B2B operators with a critical edge: using predictive diagnostics to reduce unplanned downtime, manage energy resources more efficiently, and ensure safer operations across a wide range of applications.
