Multi-Sensor Data Fusion for Smart Home Reliable Pedestrian Localization

Today's elderly face physical problems that reduce their independence to perform daily activities. The integration of environmental and body sensors in smart homes is a pertinent solution to provide a better quality of life in safer conditions. People localization and tracking in indoor environments are among the main services to be developed to follow them at home and allow them to assess their physical states. In this paper, we propose to localize and track the center of pressure of people (one or more) in a smart home using a load sensing floor and wearable sensors. Data fusion is performed using an informational filter where an inverted pendulum biomechanical model is introduced. The results obtained are very promising and have been validated using a motion tracking system and force plates.