Crosser teams with Scania and University of Linköping for a VINNOVA research project regarding large-scale edge analytics with federated machine learning models. VINNOVA has granted funds for the project which is led by Scania with Crosser and the University of Linköping as other participants.
“In general, Industrial IoT projects with edge analytics are not too difficult to build in small scale, lab or prototype environments”, says Göran Appelquist, CTO at Crosser. “The real challenges are around enterprise grade scalability, security and efficiency. How do you cost-efficiently scale the deployments and life-cycle management of ML models and edge analytics logic if you have a very large fleet of machines in the field? We are very excited in working with Scania and University of Linköping to address these challenges”.
The goals of the project, named FAMOUS - Federated Anomaly Modelling and Orchestration for modUlar Systems, are to develop federated protocol and models for fault detection with intermittent connected vehicles that guarantees convergence of the federated models; to integrate the federated models with a hierarchical clustering based on the underlying modular system of Scania´s vehicles; and to develop a scalable and flexible vehicle edge analytics solution for efficient development, testing, and deployment of models as well as data streaming for selected time-series sensor signals.
The planned results are to develop federated anomaly detection methods optimized for edge computing and evaluated on injected faults and on a test fleet with methods that maps anomaly classes to known or undiscovered faults. The federated anomaly detection method will be developed for a diverse vehicle fleet.
Read more about the project at VINNOVA
Article: Machine learning at the edge
Article: MLOps for Edge Analytics