I-Corps: Passive Infrared Sensor Technology Solution for Advanced Occupancy Sensing

Texas A&M University, College Station TX

Investigators

Abstract

The broader impact/commercial potential of this I-Corps project is the potential development of artificial intelligence (AI)-enabled solid-state lens integrated with passive infrared (PIR) sensors to detect occupancy. Today, buildings are equipped with PIR sensors for occupancy detection, owing to their low cost, low energy consumption, wide field of view, and high reliability. Despite these advantages, PIR sensors only detect motion - not stationary occupancy. By differentially perceiving infrared energy received by standard PIR sensors, the proposed technology may detect moving and stationary occupants. This may provide a low-cost plug-and-play solution to achieve personalized heating and cooling, identify underutilized spaces to more accurately forecast usage, improve workspaces and reduce operating expense costs. In addition, the proposed technology provides a privacy-reserving solution to monitor sleep quality and fall detection. The proposed applications include security systems, smart home appliances, traffic management, consumer psychology, and public health. This I-Corps project is based on the potential development of a solid-state electronic lens composed of liquid crystal treated with nanoparticles to enable stationary occupancy detection with current passive infrared (PIR) sensors. The proposed technology is designed to actively perceive and differentiate infrared energy by modulating the transmission rate. Though the occupants are stationary, their thermal signatures, such as body temperature, body shape, and gestures may be perceived differentially by tuning the apertures of the lens. Using a well-tuned transmission ratio in the long-wave infrared range (8-12µm) where human skin radiates the most energy, the standard PIR sensor may now intelligently perceive infrared thermal signatures of warm subjects in the field of view. In addition, using an online AI algorithm, the lens is capable of differentiating human from non-human warm subjects by classifying human identification and activities. Localized environmental information and training data that is collected can reduce computing complexity thereby elevating detection accuracy to high accuracy. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.