Back to results
US 2020/0122176 A1Published application
System and method for controlling and monitoring bathroom water flow
AssigneeAbstract Engineering, Inc.
InventorsGreg FLOYD, Ian HOWARD, Rahul VERMA, Cameron MEZIERE, Emily HOOD, Michael MAYES
Life of this patent
Estimating the term…
Abstract
The disclosure relates to a device that saves water, energy, and money and may record the savings with a software analytics dashboard. The device allows cold water to flow out when the shower is first turned on and slows or shuts water flow once the water is heated and the shower is unoccupied. After the presence of the user is detected, the shower flow may resume.
Claims30
A method comprising:
- providing, to a first processing device, a plurality of first reception signals received by a first occupancy sensor, the first reception signals corresponding to one or more features associated with one or more first sources of flowing water associated with a first property;
- training, by the first processing device, a neural network with the one or more features, wherein the trained neural network is configured to provide one or more categorizations from a set of categorizations associated with reception signals;
- optimizing, by the first processing device, the trained neural network for a lower-powered processing device, wherein the first processing device is configured to operate at a higher power than the lower-powered processing device;
- and
- deploying, by the first processing device, the optimized neural network on a remote device having a second processing device and a second occupancy sensor of a same type as the first occupancy sensor, wherein the second processing device is a same type as the lower-power processing device, wherein the deployed optimized neural network causes the remote device to determine, via the second processing device, a first categorization from the set of categorizations of second reception signals received by the second occupancy sensor from one or more second sources of flowing water associated with a second property.
2 more elements
A system for training a neural network, comprising:
- a first processing device;
- a receiving device;
- and
- a transmitting device, wherein:
- the receiving device is configured to provide to the first processing device a plurality of first reception signals received by a first occupancy sensor, the first reception signals corresponding to one or more features associated with one or more first sources of flowing water associated with a first property;
- the first processing device is configured to optimize the trained neural network for a lower-powered processing device, wherein the first processing device is configured to operate at a higher power than the lower-powered processing device;
- and
- the first processing device is configured to deploy, via the transmitting device, the optimized neural network on a remote device having a second processing device and a second occupancy sensor of a same type as the first occupancy sensor, wherein the second processing device is a same type as the lower-power processing device, wherein the deployed optimized neural network causes the remote device to determine, via the second processing device, a first categorization from the set of categorizations of second reception signals received by the second occupancy sensor from one or more second sources of flowing water associated with a second property.
5 more elements
Dependencies are inferred from the claim text.