Project Goal
Main goal of SCALLOPS is to develop a coherent, methodological framework for the design and implementation of secure agent-based pervasive computing applications. Basic research is on innovative means for flexible agent-coordinated semantic Web service discovery and composition, and effective data and service privacy enforcement in open, large scale pervasive computing environments. For this purpose, we will draw in particular on our expertise in the areas of security, multiagent systems, and service co-ordination. Research on the pervasive computational infrastructure is outside the scope of this project, instead we will rely on existing and appropriate standards, techniques, and open source software in this area where necessary. The application domain of our research will be Personal Health Information (Health-SCALLOPS).
Project Vision
The vision of a vast landscape of heterogeneous semantic Web services deployed as encapsulated business software assets in the Internet is currently becoming a reality as part of the Semantic Web. The potential of the combination of both visions, i.e. Semantic Web and pervasive computing, is some form of ambient intelligence working on the user's behalf. The computing intelligence required for these pervasive applications can be realized in the form of pro-active agents which handle the context-aware discovery, acquisition, composition, and management of application services and data. As context-aware personal assistants they will not disturb their human users in inappropriate moments other than emergency cases. What is fundamentally new to these agents is their ability to adaptively process, plan, collaborate, and compete with other agents for complex service provision on demand. They adapt dynamically to changes in the large scale and open computing and network environment with its agent society, and services landscape. These agents will no longer be visible, but become an integral part of the pervasive computing devices and infrastructure they are using. Furthermore, open environments face the additional challenge to take the users' protection of personal data and use of services, and electronic trails into account: Confidentiality (authorized disclosure of sensitive data only), integrity (preserving accuracy and reliability of information), and availability (timely and context-aware access of information and services). In the future, ubiquitous sensing, and the invisibility of embedded and mobile computing devices will make it easier than ever for context-aware agents to collect and use information about individuals. Although considerable progress has been made, the realization of secure agent-based pervasive service provisioning goes far beyond what is available to date but will become a main characteristic of the future computing world.
