Occupancy Visualization towards Activity Recognition

Abstract

We present a sensor visualization system that integrates data streams from individual custom sensor arrays together with Building Automation System (BAS) data. To help bridge the gap between actual building usage by the occupants, and the aggregate assumed usage by the control system, we have developed several sensor processing subsystems moving toward automated human activity recognition without the need for directly instrumenting the occupants. By having a system with a detailed understanding of occupancy behavior and needs, we believe buildings could be much more efficient thereby reducing energy consumption, working toward sustainability of the built environment.