After heating, ventilation, and air conditioning (HVAC), lighting is the most dominant component of energy consumption in workspaces and commercial buildings. The technology offer presents an occupancy aware smart lighting system that supports dramatic reduction of such lighting energy consumption by (a) non-intrusively identifying the specific areas within a floor that are occupied by humans, occupants, (b) estimating the current luminance levels in different areas within the floor and (c) automatically adjusting the intensity levels of controllable LED lights. Using the available CCTV surveillance cameras in the environment (buildings), the system takes camera images as input, and applies AI technologies to detect occupant locations and the current level of lighting at 2-3 meter granularity. Utilizing these inputs, the solution then optimally and continually adjusts the intensity level of different LED luminaires, via standard BMS (Building Management System) interfaces, to minimize energy overheads while assuring occupant comfort.
The smartlight system does not require any specific equipment, but uses existing CCTV surveillance feeds together with AI-based technologies, to continually adapt the lighting levels of multiple luminaires collectively. Features: Real-time estimation of luminance intensity and daylight levels in a given area, without requiring instrumentation of additional sensors. Real-time, fine-grained (2-3 meters error i.e., at table-level granularity) sensing of human occupancy. Automated and joint intensity tuning of multiple LED lights to maximize energy savings without sacrificing users' comfort. Lower lighting energy consumption of workspaces/buildings, compared to alternatives based on use of motion sensors OR time schedules. As mentioned before, there is no extra cost involved in the purchase of new equipment and the technology can use the surveillance cameras available in any public space.
Urban Solutions and Smart Buildings
Promotes sustainable operations of built environments and indoor spaces, by applying AI techniques on camera-captured images with no additional infrastructure investment cost: