By Urban Marovt, 1Home Solutions
The global home automation market is going crazy at the moment, to the point where it seems that soon, every device we bring into our homes will be smart. Over the past three years, the IoT has gained a lot of traction due to its user-friendliness and the fact that it’s great at bridging the gap between smart home AV, security and domestic appliances.
But the development of these new devices with different communication protocols calls for a kind of interconnectivity that so far hasn’t been present in the smart home industry – at least not in a friendly way. What the market needs is a simpler way to add new devices to an existing smart home and control them from a single interface.
This need is clearly illustrated by multiple companies approaching the same problem from different angles. Specific examples of good practice and ready-made solutions can be found in the full research using the link at the end of this article. As you will see, there is no universal solution. Some manufacturers ‘lock up’ a rather good approach inside their ecosystem of devices, whilst others address the problem through installer practices that have never really been embraced in this line of business.
1Home looked at this and thought that, even though a KNX installation does not store a human-readable configuration, but instead, functions that are grouped into channels with added text description fields, there is still room for improvement through the use of modern practices, such as software and machine learning.
The 1Home solution
We approached the challenge from a different angle, and are proud to announce that we have built an auto-detection algorithm that guarantees 80% accuracy in reconstructing a smart home configuration out of an existing ETS project file.
The solution is based on KNX catalogues that are used to define device functions in the ETS. After setting up this database, we grouped these functions into more than 80 function classes that are required for visualisation and voice control. In addition, we developed a machine learning model using methods of natural language processing and classification algorithms.
What does this mean for the KNX professional?
We are already using the new algorithm to offer the simplest way for KNX smart home installers and owners to retrofit voice control and visualisation apps in existing smart homes. We basically let them integrate smart assistant platforms such as Amazon Alexa, Google Home and Apple HomeKit, in a matter of minutes.
Installers and users are also able to integrate all IoT devices supported by those platforms (over 100k devices in total) with their core home base. This enables them to build interconnected automation over a superior interface, with almost no labour cost. Speech, touch, sensors, location, time – any of those can be a trigger for device actions to happen within two different communication protocols, namely IoT and KNX (wireless and wired).
At 1Home we envisage lights, blinds, heating and ventilation systems remaining wired for the sake of reliability, whilst there will be even more cooking, cleaning and entertainment appliances that are connected in a different way. We want to truly define the term ‘smart home’ by merging these worlds into one seamless experience. With the 1Home solution, we are well on our way, and we look forward to implementing our algorithm to give smart home installers and users the greatest benefit. We welcome feedback and suggestions at firstname.lastname@example.org.
Urban Marovt is the Head of Product for 1Home Solutions GmbH, developer of the 1Home auto-detection algorithm that allows voice control and visualisation apps to be retrofitted in existing smart homes.