Infrastructure-free positioning and motion modelling for independent and supported housing of the elderly


Knowledge of the position and motion of the elderly will provide invaluable information about the state of health and need for help when living independently or in supported housing. Satellite positioning is the most used method providing global accurate and reliable position information. However, satellite positioning is not available indoors and therefore other technologies have to be used. Signals-of-opportunity, such as WLAN and Bluetooth, are used in most commercial indoor positioning applications, but they require massive infrastructure and preparation and are therefore not suitable to be used in elderly housing. Different sensors worn by the user provide motion information which can be further transformed into position information. In addition to not requiring infrastructure, sensors provide measurements revealing user motion models (e.g. crawling, lying, walking) when processed with machine learning methods. Motion models and position information may be used for remotely monitoring the behavior and health condition of the elderly for e.g. assessing the need for immediate help.


Full name: Laura Ruotsalainen
Filiation: University of Helsinki


Dr. Laura Ruotsalainen is an Associate Professor of Spatiotemporal Data Analysis at the Department of Computer Science at the University of Helsinki, Finland. Until August 2018 she led a research group on Sensors and Indoor Navigation at the Finnish Geospatial Research Institute, Finland. She received her master’s degree from the Department of Computer Science, University of Helsinki in 2003 and doctoral degree in 2013 from the Department of Pervasive Computing, Tampere University of Technology. Her research interests include use of computer vision in navigation, development of sophisticated Global Navigation Satellite Systems (GNSS) and sensor fusion methods for robust navigation and situational awareness especially in urban areas, indoors and for GNSS interference mitigation. Her interests include also development of machine learning algorithms using spatiotemporal data.