3-D Object Localization in Smart Homes: A Distributed Sensor and Video Mining Approach

Tracing objects in 3-D space provides information while can be used to analyze, model, predict, and recognize daily activities in the smart home environment. In this paper, we introduce a novel approach for object localization in the smart home. Our proposed method does not require the use of sensors attached to objects, so in the data collection step, the objects may move freely in the home environment. Fuzzy logic techniques are utilized to model the localization information. Specifically, the proposed framework integrates the distributed information streams obtained from multiple sensors including the visual sensors. We discuss the use of robotic assistants as part of an integrated smart home environment. Simulation results indicate that the proposed solution provides improved localization performance over the state of the art methods and introduces an intuitively pleasing robot guiding solution.