Technology
Bridging the gap between theory and the real world
True battery intelligence, powerful and applied
Born out of a decade's experience developing batteries for the toughest of applications, we bring together this practical approach with the latest in electrochemical modelling from academia, fused with the best of artificial intelligence and probabalistic statistical methods
Physics-informed battery modelling, turning the best of academia into robust state estimation
We select the most important elements of physics-based models for interpretability at the electrode level, but with a grounded pragmatism that ensures observability from real-world data. We understand the cell as a system of systems, allowing us to elegantly break-down degradation into multiple dimensions for lifetime insight, without getting bogged down in micro-scale electrochemistry.
Fusing machine learning with deep battery domain understanding
We believe in the fusion of probabilistic machine learning with a deep understanding of the battery domain. Our virtual sensors are derived from fundamental electrochemical principles for best-in-class life forecasting that minimises reliance on extensive training data.
Designed for the rigours of real-world data
Having worked on battery systems for the last decade, deployed into the toughest of environments, we understand the limitations of lab-based data when applied to real-world problems. Our Embedded algorithms are designed to run on standard automotive-grade hardware platforms, and our Cloud Platform features prognostics designed to detect real-world failure mechanisms at module and pack-level.
