Abstract FR:

The rapid development of information and communication technologies is opening a large variety of new services and aids not only for independent people, but also for dependent persons due to physical/cognitive restriction or the elderly. Combining technical aids and mobile technology allows those people to benefit from an independent living through ambient intelligent environments. The interaction with pervasive environment represents a difficult task mainly when targeted people have special needs (mental or physical impairment). Indeed, the interaction should be done through a user‐friendly interface by making it easier to access to the environment and benefit from assistance. Consequently, the User Interface should support not only user preferences but also user capabilities. At the same time, different surrounding computing platform are considered as an additional source of heterogeneity. The design of dynamic user interface represents an open issue which should be considered especially when focusing on pervasive environment. The term “plasticity” is inspired from the property of materials that expand and contract under natural constraints without breaking, thus preserving continuous usage. Applied to HCI, plasticity is the capacity of an interactive system to withstand variations of context of use while preserving usability. This research work focuses on developing mechanism of plasticity and adaptation of user interfaces in Ambient Intelligent space based on service provision architecture. This dissertation presents a set of requirement to handle plasticity like Context awareness, service modeling and user modeling. We describe a set of approaches and mechanism for user interface development and present how they inspired us to propose our model for the user interface plasticity in pervasive environment. In parallel, we developed an ontology for the description of adaptation mechanism including service presentation, user profile and devices. This ontology makes it possible to describe the adaptation semantically and finally to generate adapted user interfaces based on ontological rules elaborated throw users investigations. Developed software framework was integrated in a global system deployed within pilot sites (residence of dependant people) in order to validate its impact when involving real end‐users in ecological situations.