Uncertainties in Structural Behavior for Model-Based Occupant Localization Using Floor Vibrations

In sensed buildings, information related to occupant movement helps optimize functions such as security, energy management, and caregiving. Due privacy needs, non-intrusive sensing approaches for tracking occupants inside vibration sensors, are often preferred over intrusive strategies that involve use of cameras wearable devices. Current sensor-based occupant-localization data-driven techniques do not account structural behavior limited slabs on grade. Varying-rigidity floors inherent variability in walking gaits lead ambiguous interpretations floor vibrations when performing model-free localization. this paper, an extensive analysis induced by a range is described. Firstly, the need structural-behavior model localization assessed using two full-scale case studies. Structural found significantly influence footstep impacts. Since simple relationship between distances from footstep-impact sensor locations cannot be assured, physics-based models necessary accurate Secondly, measured data interpreted uncertainties multiple sources. There types uncertainties: modelling measurement uncertainties, including gaits. Error-domain falsification (EDMF) residual minimization (RM) model-based interpretation. Unlike RM, EDMF explicitly accounts presence systematic errors parameters overall bias. carried out RM study. By accounting behavior, EDMF, unlike accurately reveals possible slabs.