Self-Configuring User Interface Design Navy SBIR FY2009.1 Sol No.: Navy SBIR FY2009.1 Topic No.: N091-061 Topic Title: Self-Configuring User Interface Design Proposal No.: N091-061-0167 Firm: Iterativity, Inc. 3236 17th Ave. S. Minneapolis, Minnesota 55407 Contact: Robin Penner Phone: (612) 669-7433 Web Site: www.iterativity.com Abstract: We propose to develop a tool that integrates diverse subsystems and automatically configures and composes well-designed, unified, and consistent user interfaces. Using active object-oriented models of situations, interactions, and interface elements, this tool will dynamically generate user interfaces that are specialized to each user, have a unified and common look and feel, require minimal training, and are consistent and easy to use. Because all knowledge is model-based, evolution and diversification are simple and verifiable. In Phase I, we will investigate the applicability of this approach to the NPES functions of Virginia-class submarines, culminating in a proof of concept system for a subset of functions, users, and tasks. In addition, we will perform directed research on the mechanisms required to allow system developers, integrators, and usability specialists to gracefully evolve and configure the system, and will investigate mechanisms for self-test and system self-improvement. Benefits: We are convinced that an automated, self-configuring user interface creation capability is optimal for solving the critical problem of reducing the time to develop user interfaces and as a powerful mechanism to provide cross-system consistency. With a model-based compositional approach, the result is not simply a toolset to help test and debug human-designed and -implemented UIs, but an automated reasoning system that dynamically designs and manages good user interactions. Seamless operation of different functional elements, guaranteed application of good user interface design procedures for interactions, and the ability to automatically use multiple devices and modalities for an integrated interaction for single or multiple users are just some of the features facilitated by this approach. Users are not the only class to benefit, however; developers are free to concentrate on basic questions of good design and appropriate modeling of tasks and roles, and the requirements for training and support are significantly decreased by using a model-based automated reasoner to automatically design well-formed user interfaces. Current military and commercial systems of all types use hard-coded user interfaces as the mechanism to communicate with users. An adaptive, responsive, dynamic, real-time user interface generation system that provides seamless integration within service-oriented architectures would be of inestimable benefit in many current and planned military systems. An automated interaction design capability will also be beneficial in a number of civilian applications, including search and rescue, security, transportation, medical, education, and systems management. Return