Context Reasoning and Prediction in Smart Environments: the Home Manager case

In Smart Environments, people are immersed in time and space in an augmented reality: computing systems are ubiquitous, intelligence pervades the whole environment and exploits space and time awareness to provide a globally contextualised, adaptive user experience. Because of their socio-technical nature, the development of Smart Environments calls for skills, concepts, methodologies, technologies from the most diverse fields – AI, coordination, distributed systems, organisational sciences, etc. –, thus inherently calling for a multi-paradigm perspective. This paper explores the Smart Environment context moving from two basic bricks: Butlers for Smart Spaces, a framework for the design of smart services for users immersed and interacting with their surrounding environment; and Home Manager, a multi-paradigm, agent-based platform for the implementation of Smart Living contexts—particularly focused on the reasoning aspects, so as to anticipate the users’ needs. For concreteness, we take a Smart Kitchen as our running example: we first discuss how it can be devised using Butlers for Smart Spaces, and then deployed on the Home Manager platform—highlighting in particular the context reasoning, prediction and adaptation aspects.