Occupancy distribution estimation for smart light delivery with perturbation-modulated light sensing

Abstract

The advent of modern light-emitting diode (LED) techniques enables us to develop novel lighting systems with numerous previously unavailable features. Specifically, by using the fixtures for both illumination and to interrogate the space, source-to-sensor communication becomes possible at very low cost. In this paper, we present a novel framework to estimate the occupancy distribution in an indoor space using color-controllable LED fixtures (the same fixtures providing the illumination, simultaneously) and sparsely distributed non-imaging color sensors. By modulating randomly generated perturbation patterns onto the drive signals of the LED fixtures and measuring the changes in the color sensor responses, we are able to recover a light transport model for the room. Two approaches are proposed to estimate the spatial distribution of the occupancy, based on a light blockage model and a light reflection model, respectively. These two approaches, which can be combined, can faithfully reveal the occupancy scenario of the indoor space, while preserving the privacy of its occupants. An occupancy-sensitive lighting system can be designed based on this technique.

The Energy Saving Potential of Occupancy-Based Lighting Control Strategies in Open-Plan Offices: The Influence of Occupancy Patterns

“…In addition to new lighting sources, automatic lighting control strategies are being applied to provide lighting based on the real-time needs of offices [4]. The main variables determining these needs are (1) the availability of daylight, and (2) the presence of occupants in an office space.…”

Short Paper:

“…For example, the ToF sensors provide object distance information and are complementary to the installed color sensor array that provides spectral distribution of reflected light patterns. Previous work [7] demonstrated the feasibility of using imperceptibly-modulated light in combination with sparse color sensors to estimate configurations of objects in a smart room. We also plan to integrate finer-level classification of human gestures for smart lighting applications, using machine-learning algorithms applied to very-low-pixel-count sensors [3].…”

Exploiting Color Sensors to Provide Optimal Lighting and Anonymous Tracking in Stores

“…Tunable White LED. Tunable white lighting technology allows users to adjust the color temperature of luminaries [15], which can be used to improve the productivity of people in offices or the well-being of patients in healthcare facilities [18] No No High decimeter [19] No No None meter [20] No Yes None zone (several mtrs) [21] Yes No High zone [22] Yes Yes None zone [23] Yes Yes None zone Lux-Track Yes Yes Medium zone [16]. We are not aware of any study that provides automatic tunable white lighting.…”

“…Active Source, Passive Object. Other systems modify light bulbs to detect occupancy without requiring users to carry any electronic device [23], [20], [24]. Those systems place simple photosensors in the environment to measure perturbations in light reflections caused by people.…”