Concerning the inhabitant behavior the system trains on the one hand the usual presence samples and uses these for different tasks of automation. On the other hand the inhabitant can communicate desires to the system and adapt the automation rules thereby. This is done not via a programming system with a special aid, but completely simply, as the usual control elements (e.g. a wall switch for the light) are used, in order to override (unwanted) the pre-setting of the system. The system seizes this information, in order to adapt later, with a similar situation, the control advoice. This adjustment takes place not immediately to 100% - the user made an over-regulation by hand perhaps only because of an exceptional case - to separate gradually, with consistent, several times same behavior.

This kind of learning from the inhabitant to applies to all building services components, i.e. for light (also dimmbar), blinds, heating, ventilation, etc. the usual control elements become thereby thus quasi programming tools the adjustment of the automation rules to the desires of the inhabitants.

These self learning mechanisms were internationally patented and examined in detail and optimized in a 9-month practice test.