The future smart home will be enhanced and driven by the recent advance of the Internet of Things (IoT), which advocates the integration of computational devices within an Internet architecture on a global scale [1, 2]. In the IoT paradigm, the smart home will be developed by interconnecting a plethora of smart objects both inside and outside the home environment [3-5]. The recent take-up of these connected devices within home environments is slowly and surely transforming traditional home living environments. Such connected and integrated home environments lead to the concept of the smart home, which has attracted significant research efforts to enhance the functionality of home environments with a wide range of novel services. The wide availability of services and devices within contemporary smart home environments make their management a challenging and rewarding task. The trend whereby the development of smart home services is decoupled from that of smart home devices increases the complexity of this task. As such, it is desirable that smart home services are developed and deployed independently, rather than pre-bundled with specific devices, although it must be recognised that this is not always practical. Moreover, systems need to facilitate the deployment process and cope with any changes in the target environment after deployment. Maintaining complex smart home systems throughout their lifecycle entails considerable resources and effort. These challenges have stimulated the need for dynamic auto-configurable services amongst such distributed systems. Although significant research has been directed towards achieving auto-configuration, none of the existing solutions is sufficient to achieve auto-configuration within smart home environments. All such solutions are considered incomplete, as they lack the ability to meet all smart home requirements efficiently. These requirements include the ability to adapt flexibly to new and dynamic home environments without direct user intervention. Fulfilling these requirements would enhance the performance of smart home systems and help to address cost-effectiveness, considering the financial implications of the manual configuration of smart home environments. Current configuration approaches fail to meet one or more of the requirements of smart homes. If one of these approaches meets the flexibility criterion, the configuration is either not executed online without affecting the system or requires direct user intervention. In other words, there is no adequate solution to allow smart home systems to adapt dynamically to changing circumstances, hence to enable the correct interconnections among its components without direct user intervention and the interruption of the whole system. Therefore, it is necessary to develop an efficient, adaptive, agile and flexible system that adapts dynamically to each new requirement of the smart home environment. This research aims to devise methods to automate the activities associated with customised service delivery for dynamic home environments by exploiting recent advances in the field of ubiquitous robotics and Semantic Web technologies. It introduces a novel approach called the Knowledge-based Auto-configuration Software Robot (Sobot) for Smart Home Environments, which utilises the Sobot to achieve auto-configuration of the system. The research work was conducted under the Distributed Integrated Care Services and Systems (iCARE) project, which was designed to accomplish and deliver integrated distributed ecosystems with a homecare focus. The auto-configuration Sobot which is the focus of this thesis is a key component of the iCARE project. It will become one of the key enabling technologies for generic smart home environments. It has a profound impact on designing and implementing a high quality system. Its main role is to generate a feasible configuration that meets the given requirements using the knowledgebase of the smart home environment as a core component. The knowledgebase plays a pivotal role in helping the Sobot to automatically select the most appropriate resources in a given context-aware system via semantic searching and matching. Ontology as a technique of knowledgebase representation generally helps to design and develop a specific domain. It is also a key technology for the Semantic Web, which enables a common understanding amongst software agents and people, clarifies the domain assumptions and facilitates the reuse and analysis of its knowledge. The main advantages of the Sobot over traditional applications is its awareness of the changing digital and physical environments and its ability to interpret these changes, extract the relevant contextual data and merge any new information or knowledge. The Sobot is capable of creating new or alternative feasible configurations to meet the system's goal by utilising inferred facts based on the smart home ontological model, so that the system can adapt to the changed environment. Furthermore, the Sobot has the capability to execute the generated reconfiguration plan without interrupting the running of the system. A proof-of-concept testbed has been designed and implemented. The case studies carried out have shown the potential of the proposed approach to achieve flexible and reliable auto-configuration of the smart home system, with promising directions for future research.

Avec l'apparition des paradigmes des systèmes ubiquitaires ou omniprésents et de l'intelligence ambiante, on assiste à l'émergence d'un nouveau domaine de recherche visant à créer des environnements ou écosystèmes intelligents pouvant offrir une multitude de services permettant d'améliorer la qualité de vie, l'état physique et mental, et le bien-être social des usagers. Dans cette thèse, nous nous focalisons sur la problématique de la représentation sémantique des connaissances et du raisonnement dans le cadre des systèmes à intelligence ambiante et des robots ubiquitaires. Nous proposons deux modèles sémantiques permettant d'améliorer les fonctions cognitives de ces systèmes en termes de gestion du contexte. Au premier modèle, de type ontologique, sont associés un langage de règles et un raisonnement réactif pour la sensibilité au contexte. Pour prendre en compte le caractère dynamique du contexte et assurer une prise de décision cohérente, le mode de raisonnement retenu garantit deux propriétés essentielles : la décidabilité et la non-monotonie. Le deuxième modèle, également de type ontologique, complète le modèle précédent en termes d'expressivité pour la représentation de contextes non-triviaux et/ou liés au temps. Il s'appuie sur des relations n-aires et une représentation narrative des événements pour inférer des causalités entre événements et reconnaitre des contextes complexes non-observables à partir d'événements passés et courants. Les modèles proposés ont été mis en oeuvre et validés sur la plateforme ubiquitaire d'expérimentation du LISSI à partir de trois scenarii d'assistance cognitive et de reconnaissance de contexte / With the appearance of the paradigms of ubiquitous systems and ambient intelligence, a new domain of research is emerging with the aim of creating intelligent environments and ecosystems, that can provide multiple services that can improve quality of life, the physical and mental status and the social wellness of the users. In this thesis, we address the problem of semantic knowledge representation and reasoning, in the context of ambient intelligent systems and ubiquitous robots. We propose two semantic models that improve the cognitive functions of these systems, in terms of context recognition, and context adaptation. The first one is an ontology-based model, which is associated with a rule language to model reactive reasoning process on contextual knowledge. To take into account the dynamicity of context and insure coherent decision-making, this process guarantees two essential properties: decidability and non-monotonic reasoning. The second model is also an ontology-based model that completes the previous model in terms of expressiveness for semantic representation of non-trivial contexts with temporal dimension It is based on n-ary relations and a narrative representation of events for inferring causalities between events, and therefore to build the chronological context of a situation as from past and current events. The proposed models have been implemented on the ubiquitous experimental platform of LISSI, and validated through three scenarios for cognitive assistance and context recognition