Incorporation of Commonsense Reasoning of Geographical Spaces and Advanced Computer Visualization for Voice Controlled Map Sketching

The goal of this Phase I research project is to develop an integrated system to enable multi-modal input, allowing a participant to use spoken commands to create and edit map sketches. The project will develop an architecture for multi-modal interaction,perform initial experimental validation, and develop a set of metrics and an experimental paradigm for demonstrating the strength and weaknesses of the proposed technology. The Phase I research will evaluate the performance and usability differences amongthe modes of input for different sketch creation tasks, with a particular focus on speech commands. The research will also explore the use of a speech recognition approach that exploits commonsense knowledge of geographical spaces for a particular set ofcommand tasks (e.g., via incorporation of Na¿ve Geography based reasoning). The ultimate goal is to synergistically combine this spatial reasoning with sketch understanding to support course of action's planning, editing, and monitoring within amulti-modal command post. The first targeted commercialization sector for this technology is the area of multimodal command interaction in military command posts. This technology will also be extremely applicable to all other areas where instructors orknowledge workers might wish to incorporate illustrative sketches, animations and who desire a more fluid interface to their presentation creation tool.