CI-ADDO-EN: Smart Home in a Box: Creating a Large Scale, Long Term Repository for Smart Environment Technologies

Washington State University, Pullman WA

Investigators

Abstract

There is considerable current interest in developing smart environment technologies. Such efforts present research challenges in, and integration of research outcomes from, diverse disciplines including artificial intelligence, pervasive computing, robotics, interfaces, middleware, and sensor networks. The lack of availability of large-scale sharable data sets from smart environments is a major stumbling block for rapid advances in this area. Against this background, this project aims to develop and deploy a data and tool repository needed by the smart environment research community. The anticipated results of this infrastructure project include 1) a streamlined, do-it-yourself smart home kit, 2) a web interface to upload, access, and annotate smart environment data, 3) meta data including functional assessment scores and energy usage, and 4) software tools to recognize, visualize, and analyze home-based behaviors. The investigators aim to assess the impact of the resulting repository (CASAS) using measures such as number and diversity of researchers utilizing the repository, number of datasets and tools contributed to the repository, research, education, and commercial advances related to the repository, and publication citations to the repository. Broader impacts of the project include (i) the do-it-yourself smart home toolkit, data sets and software tools that enable research and educational efforts by a large community of researchers in artificial intelligence, pervasive computing, robotics, interfaces, middleware, and sensor networks; (ii) enhanced opportunities for researchers in cognitive psychology, gerontology, and sociology to contribute to interdisciplinary research in smart environments; and (iii) enhanced research-based training opportunities for students from underrepresented groups. The datasets, software tools and educational materials that result from this work will be made available as part of the CASAS repository at http://ailab.wsu.edu/casas/.