Ordonnancement dans l'habitat intelligent
In Revue Ouverte d'Intelligence Artificielle 4(1). pages 53-76. 2023.
Alexandre Demeure, Sybille Caffiau
Résumé
The text is about the problem of the scheduling of the actions applied to the actuators of a smart home. These actions can be triggered either by inhabitants or by programs encoding automatisms. We show that this is a complex problem that cannot be solved a priori. On the contrary, it depends on the context. We defend the idea that this problem should be tackle from the angle of an operating system which scheduling engine would be based on CCBL (Cascading Contexts Based Language). CCBL is an end-user programming language for the smart home that enable inhabitants to programs automatisms based on devices and services. We provide several examples of scheduling strategies programmed with CCBL. We show using CCBL to program such strategies is not fundamentally different than programming mere automatisms. Hence, the skills acquired in one of the tasks will be reusable in the other.