Skip to the content

Stockholm and Varberg, Sweden, March 31, 2019 The two leading Swedish IoT companies, Crosser and Ekkono Solutions, today announced a joint solution that enables Crosser’s customers to embed Ekkono’s edge machine learning in their Edge Analytics & Orchestration solution.

“With Ekkono’s technology we’re adding a high level of adaptive machine learning to the edge”, says Mikael Samuelsson, Director of Strategic Alliances & Partners at Crosser. “In our library of modules, we have a couple of machine learning alternatives already, but with Ekkono we’re introducing the ability to do real-time, incremental learning on streaming data to learn individual conditions and adapt to changes over time.”

Crosser and Ekkono individually solves two parts of the same problem; It’s unfeasible to upload all high-frequency raw sensor data to the cloud. Applying intelligence in the edge, close to the data source, facilitates instant actions on all sensor data. It also enables custom learning of the local conditions and circumstances. And it mitigates poor, varying or restricted connectivity by circumventing it through distributed functionality. The combination Crosser and Ekkono offers a comprehensive and best-of-breed solution for genuine, on-premise, predictive maintenance, performance optimization, automation, and self-configuration of machines, vehicles and other installations.

Technically, Ekkono’s edge machine learning library is integrated into Crosser’s library of functionality modules in the Crosser Flow Studio. This allows end-users to design, test and remotely deploy integrated calls to Ekkono’s functionality from Crosser’s streaming analytics environment. This functionality includes preprocessing of data through smart pipelines, incremental learning, predictions, generation of virtual sensor values (e.g. KPIs or health indicators), change and anomaly detection, and attribution of what has the biggest impact on the outcome. The generated output can also be sent to the cloud as enriched data for further processing and complementary functions.

“The market has matured dramatically, and enterprises today have real plans and intentions with connecting their assets to the Internet” says Rikard König, CTO at Ekkono. “We provide one – though crucial – component to the solution, and we work with state-of-the-art partners, like Crosser, to ensure comprehensive functionality and quality to customers.”

Both Crosser and Ekkono will be present at Hannover Messe this week to promote the joint announcement. Crosser is found in Hall 6, stand B52, and Ekkono at the Swedish Co-Lab Pavilion (Hall 27, stand H30) and Automatic Region (Hall 11, stand E64);

www.ekkono.ai/hannover2019.

Ekkono Solutions AB

Ekkono does machine learning for IoT (Internet of Things). The product, which is the result of 7 years of research at the University of Borås, Sweden, is a programmable embedded software that runs onboard connected devices, so called Edge Machine Learning, that makes them smart.

For further information visit www.ekkono.ai,
or contact Jon Lindén, CEO, at jon@ekkono.ai or +46-709-576006.

About the author

Try Crosser For Free

Start innovating today. How it works:

  1. Get an account
  2. Log in and start designing your flows in the sandbox
  3. Download the Crosser Container to your local test node
  4. Test with real data