Ekkono means cognition, and that's what we add to your world of connected things. We make them smart through online learning that empowers you to detect anomalies, predict future performance and failures, do forecast-based optimization, and identify what makes them perform better and more intuitively.
The unique component, and the result of six years of machine learning research at the University of Borås (Sweden), is a lightweight embedded advanced analytics engine that we license (OEM) to software and hardware developers. The small footprint enables it to run all the way out at the edge, even on small microcontroller platforms at par with Raspberry Pi/Arduino, or smartphones like iPhone and Android.
IDC: ”By 2019, 40% of digital transformation initiatives – and 100% of IoT initiatives – will be supported by AI capabilities.”
The result is tailored artificial intelligence, instant response, reduced network noise, and consistent functionality, even when there is a loss of connectivity. This translates into money as you proactively avoid unscheduled downtime, and optimizes performance. Your things and machines get more intuitive through self-learning and prediction, which makes them more fun, more useful, and less annoying. They get upgraded from connected to smart. By doing this at the edge, Ekkono uses available processing resources in a better way. Simultaneously, our cloud-base core module cross-references the findings and leverages crowdsourcing of data from the entire installed base to benefit the power of many.
Ekkono was awarded finalist in the Serendipity Challenge 2017, which makes us one of "the 50 most innovative start-ups in the Nordics". Read more about the Serendipity Challenge and all the finalists at www.theserendipitychallenge.se.
Ekkono is an embedded advanced analytics engine for Internet of Things (IoT). The unique design makes it resource efficient with an unparalleled small footprint. Still, Ekkono doesn’t compromise on functionality:
This enables proactive detection and self-healing of faults, alerts when a system deviates from normal state or behavior, scheduled rather than unscheduled maintenance, determination of what has the biggest impact on target, optimization of performance, more intuitive human-machine interaction, genuine personalization, behavioral learning, tailored recommendations, and more.
Ekkono manages the entire workflow, i.e. data aggregation, model training, and execution. All these steps are built in a modular fashion that enables one, some, or all of them to run at the edge, in the cloud, or on a connecting laptop. The choice depends on hardware configuration and conditions, like update interval and response time. The modular design also allows you to sequentially deploy functionality from the feature list above, and thereby make your product better and better, step by step.
Ekkono is platform-agnostic and runs on virtually any platform and in any environment. In combination with standard APIs for input/output of data, and integration with surrounding applications, it becomes quick and easy to try out, and to start implementing your vision of things/machines that are a little smarter.
IoT (Internet of Things) holds the promise of everything getting smarter – from homes to cars, robots, vending machines and cities. The reality is that most of them are just getting connected – when they get a connection. Many of them move around, or are in geographical or physical locations with challenged connectivity. And the smartness is capped at uploading raw data to a big-data haystack, showing historical averages for the installed base, and maybe keeping track of holidays.
But what if you could switch to being truly smart by proactively detecting an emerging issue? What if you could do this for an individual unit based on its normal state and operational environment? What if each unit got constantly better by learning more about, and by, itself? And what if it would function, autonomously, even when connectivity was lost?
Well, guess what, this is all possible. This is the kind of smarts Ekkono provides to the world of IoT. We do this as an embedded software engine that can be deployed all the way out at the edge. That’s why it’s called edge analytics, which enables intelligence to be out where issues occur, and where things interact with humans. The benefits of edge analytics are multifold:
For industrial implementations, edge analytics translates into money. Either by being one step ahead of an emerging issue, or by being one step ahead of the competition:
Ekkono works with the following industries: Energy, Telecom, Smart Cities, Manufacturing, Agriculture and Forestry, Airspace and Defense, Transportation (air, trucks, buses, shipping), Healthcare, Real-Estate, Finance and Retail.
For consumer IoT the value proposition is more in an elevated user experience. As consumers today, we are more sensitive and susceptible to the overall impression of a product. Adding Ekkono makes your things self-learning about the user’s behavior, and predicts her next step. This facilitates more relevant and applicable recommendations, and more effective communication with the end-user.
Ekkono focuses on the following consumer IoT segments: Smart Homes, Health, and Transportation.
Ekkono’s architecture and platform-independent design, enables integration at all levels in the stack. This enables Ekkono to adapt the solution and integrate it as native code into CPUs, IoT platforms, gateways, hardware platforms (e.g. Arduino, RaspberryPi, iOS, Android), smartphones, VR goggles and drones.