Distributed product usage analytics, asset management of off-road and special-purpose vehicles in mobile operation.
Many companies operate large groups of complex, geographically-distributed assets, such as turbines, photo-voltaic plants, special-purpose vehicles, industrial machines, etc. These are ever more instrumented and connected and their usage and performance data is captured continuously. Current applications exploiting that data do not consider the inherent distributed aspects and the associated challenges it provokes. In particular, continuous connectivity is required for gathering and analyzing usage and performance data in the cloud, but is in many cases expensive or just not feasible. Edge intelligence, a mix of edge computing and data intelligence where more elaborate computations are done on the device itself, is a promising principle in this context. It allows assets to exploit the data in (semi) real-time tocompute statistics and train models, while only sending a selected amount of data to the cloud when needed or when near a collection infrastructure.
Within STUNT, we are researching edge intelligence solutions addressing the following challenges::
- Intelligent data reduction > how to address limited on-asset storage, low bandwidth and high communication costs?
- Opportunistic data transfer > how to tackle the absence of continuous connectivity and short offloading opportunities?
- Distributed intelligence > how to train models at the edge and propagate knowledge?
The solutions we realize will be demonstrated through industrial prototypes in the context of asset management of off-road and special-purpose vehicles in mobile operations, in close collaboration with the industrial partner ILIAS Solutions.
Start date: 01/04/2019
End date: 31/03/2022
Projet subsidié par la Région de Bruxelles-Capital - Innoviris (Project subsidized by the Brussels Capital Region - Innoviris):