In this paper, we present an implementation of a real-time position inference system for a research group. In a research group, knowing the other members' presence, predicting visitors, and holding real-time meetings are time-consuming but important tasks in everyday life. Thus, in this system, we realize a prediction mechanism for group members and visitors. We also develop a support mechanism for holding real-time meetings. In order to implement a real-time position inference system, we need to detect location information of group members. In our school building, EIRIS (ELPAS InfraRed Identification and Search System) has been installed, and we utilize EIRIS to detect members' positions. However, there exists a trade-off between the value and cost of detecting exact positions. Thus, we first try to complement EIRIS's detection ability by employing a stochastic method, probabilistic reasoning, and heuristic rules. Then, we implement a visitor prediction mechanism and a real-time meeting support mechanism. Our experiments demonstrate that our position inference mechanism is more effective than a simple prediction mechanism that employs only stochastic data.
CITATION STYLE
Ito, T., Oguri, K., & Matsuo, T. (2004). A location information system based on real-time probabilistic position inference. In Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science) (Vol. 3029, pp. 797–806). Springer Verlag. https://doi.org/10.1007/978-3-540-24677-0_82
Mendeley helps you to discover research relevant for your work.