This project is developing the scientific and engineering foundations to enable automated support of user-centered activities where a computing infrastructure, which is largely transparent to the user, is used to enhance a users ability to carry out every-day tasks. Focusing on non-computer centric environments such as home environments, the project builds on the notion of task-driven computing to design, implement, and evaluate a general platform on which many activity-based applications can be easily built and integrated. Specifically, it is creating new principles, algorithms, and design frameworks to enable flexible integration, adaptation and configuration of emerging pervasive computing devices and services; new techniques for ensuring the robustness of heterogeneous, dynamically-changing configurations of pervasive services in unpredictable environments; and new ways to think about and implement systems in which computing augments and complements ordinary human activities, rather than requiring humans to map their needs into computer-centric applications.
This research has the potential for three significant broad impacts. (1) It will extend the reach of integrated computer support for activities that critically affect non-computer users in environments where computing support has not been broadly accessible (such as in the home); (2) It will help decrease the gap between what ordinary users need from computing systems to improve their daily lives, and what is now provided by current computer-centric technology; (3) It will reduce the complexity of deploying and using emerging pervasive computing technology, while at the same time improving its predictability and robustness.
