Dishwasher with personalized utensil detection and scanning aids therefor

Patent metadata

Field	Value
Publication number	US-12539015-B2
Application number	US-202318461182-A
Country	US
Kind code	B2
Filing date	Sep 5, 2023
Priority date	Sep 5, 2023
Publication date	Feb 3, 2026
Grant date	Feb 3, 2026

Abstract

A dishwasher may utilize personalized utensil detection to detect the utensils regularly washed by the dishwasher, in part based upon training of a personalized machine learning model used in personalized utensil detection. One or more scanning aids may also be used to provide a common frame of reference when capturing images of utensils used to train the personalized machine learning model.

First claim

What is claimed is:

1 . A dishwasher, comprising:
a wash tub; an upper rack and a lower rack supported in the wash tub, each rack configured to support a plurality of utensils and movable between a loading position and a washing position, wherein each rack is disposed in the washing position during a wash cycle; a door providing external access to the wash tub; a scanning aid disposed on an inner surface of the door and configured to support a utensil when the door is in an open position and the lower rack is in the washing position, and wherein the scanning aid is sized and positioned on the inner surface of the door to directly support the utensil when the lower rack is in the washing position; at least one spray device disposed in the wash tub; a fluid supply configured to supply wash fluid to the at least one spray device during the wash cycle that washes a load, the fluid supply including at least one pump; and a controller coupled to the at least one pump, the controller configured to control at least one parameter of the wash cycle based at least in part on detecting the utensil in the load, wherein the utensil detected in the load is of a personalized utensil type, and wherein the utensil is detected in the load by processing image data from an image captured of at least a portion of the load with a personalized machine learning model that is trained to detect the personalized utensil type using one or more training images captured of the utensil subsequent to installation of the dishwasher; wherein the controller is further configured to receive user input from a user to enter a training mode, initiate the training mode in response to the user input from the user, and during the training mode, guide the user to capture the one or more training images of the utensil from different viewpoints while the utensil is directly supported on the scanning aid and the lower rack is disposed in the washing position, including prompting the user to move or reorient the utensil after capturing a first training image of the one or more training images and before capturing a second training image of the one or more training images.

2 . The dishwasher of claim 1 , wherein the personalized machine learning model is disposed in the dishwasher, and the controller is configured to detect the utensil in the load by executing the personalized machine learning model.

3 . The dishwasher of claim 1 , wherein the personalized machine learning model is disposed in a user device in communication with the dishwasher, and the controller is configured to detect the utensil in the load by communicating the image data to the user device and receiving the personalized utensil type or the at least one parameter from the user device.

4 . The dishwasher of claim 1 , wherein the personalized machine learning model is disposed in a remote service in communication with the dishwasher, and the controller is configured to detect the utensil in the load by communicating the image data to the remote service and receiving the personalized utensil type or the at least one parameter from the remote service.

5 . The dishwasher of claim 1 , further comprising an image sensor, wherein the controller is configured to capture the image using the image sensor.

6 . The dishwasher of claim 1 , wherein the controller is in communication with a user device including an image sensor configured to capture the image, and the controller is configured to receive the image data from the user device.

7 . The dishwasher of claim 1 , wherein the scanning aid includes a plurality of radial segments from which an angular orientation of the utensil may be determined during training of the personalized machine learning model.

8 . The dishwasher of claim 7 , wherein the scanning aid includes a plurality of concentric circles from which a size of the utensil may be determined during training of the personalized machine learning model.

9 . The dishwasher of claim 7 , wherein the plurality of radial segments have differing length.

10 . The dishwasher of claim 7 , wherein at least a portion of the scanning aid is supported on a rotatable support, and the rotatable support is motorized or is manually rotatable.

11 . The dishwasher of claim 7 , further comprising a mobile device holder mounted to the dishwasher and configured to support a mobile device to capture images of the utensil during training of the personalized machine learning model.

12 . The dishwasher of claim 1 , wherein the at least one parameter includes a wash temperature, an operation duration, a number of operations, a spray pattern, a fluid pressure, a soak time, a spray isolation, or a control parameter for one or more controllably-movable sprayers.

13 . The dishwasher of claim 1 , wherein the controller is further configured to train the personalized machine learning model to detect the personalized utensil type for the utensil detected in the load using the one or more training images captured subsequent to installation of the dishwasher.

14 . The dishwasher of claim 13 , further comprising an image sensor disposed on the dishwasher and configured to capture the one or more training images.

15 . The dishwasher of claim 13 , wherein the controller is configured to detect the utensil in the load using the personalized machine learning model prior to the wash cycle being performed in the dishwasher.

16 . The dishwasher of claim 13 , wherein the controller is further configured to detect an unknown utensil in the image captured of the at least a portion of the load and prompt the user to train the personalized machine learning model to detect the unknown utensil.

17 . The dishwasher of claim 16 , wherein the personalized machine learning model is trained by personalizing a generalized machine learning model utilized with the dishwasher prior to training for any user-specific utensils.

18 . The dishwasher of claim 13 , wherein the controller is configured to detect that a second utensil is already recognized during training of the personalized machine learning model.

19 . The dishwasher of claim 18 , wherein the controller is configured to, in response to detecting that the second utensil is already recognized, train the personalized machine learning model using stored training data associated with the second utensil.

20 . The dishwasher of claim 18 , wherein the controller is configured to, in response to detecting that the second utensil is already recognized, notify the user that the second utensil is already recognized.

21 . A dishwasher, comprising:
a wash tub; an upper rack and a lower rack supported in the wash tub, each rack configured to support a plurality of utensils and movable between a loading position and a washing position, wherein each rack is disposed in the washing position during a wash cycle; a door providing external access to the wash tub; a scanning aid disposed on an inner surface of the door and configured to support a utensil when the door is in an open position and the lower rack is in the washing position, wherein the utensil is of a personalized utensil type, and wherein the scanning aid is sized and positioned on the inner surface of the door to directly support the utensil, and the scanning aid includes a plurality of concentric circles and a plurality of radial segments that provide a frame of reference from which a size and an angular orientation of the utensil may be determined; at least one spray device disposed in the was

Assignees

• Midea Group Co Ltd

Inventors

• Boyer Joel

Classifications

• A47L2501/30

  Regulation of machine operational steps within the washing process, e.g. performing an additional rinsing phase, shortening or stopping of the drying phase, washing at decreased noise operation conditions · CPC title

• A47L2401/04

  Crockery or tableware details, e.g. material, quantity, condition · CPC title

• A47L2401/00

  Automatic detection in controlling methods of washing or rinsing machines for crockery or tableware, e.g. information provided by sensors entered into controlling devices · CPC title

• A47L15/46

  Devices for the automatic control of the different phases of cleaning {; Controlling devices (A47L15/449 takes precedence)} · CPC title

• A47L15/0018

  Controlling processes, i.e. processes to control the operation of the machine characterised by the purpose or target of the control (for control of water softening or water softener regeneration A47L15/4229; for control of water level A47L15/4244; for control of bad smells or odours A47L15/4276; for control of water temperature A47L15/4287; for control of water pressure A47L15/4289; for control of condition of crockery or tableware A47L15/4295; for control of condition of washing water A47L15/4297) · CPC title