# US 2026/0131476 A1 · Techniques, machine learning, and mechanisms for enabling cleaning of devices

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0131476 A1
- Document ID: US20260131476A1
- Status: Published application
- Kind: A1
- Application number: 19/441,064
- Filed: 2026-01-06
- Published: 2026-05-14
- Earliest claimed: 2024-07-24
- Assignees: Ruby.AI Robotic Technologies Ltd.
- Applicants: Ruby.AI Robotic Technologies Ltd.
- Inventors: Daniel Ben Dov
- Web page: https://patentgrep.com/patent/US20260131476A1

## CPC

- B25J 9/1697: {Vision controlled systems}
- B25J 9/1679: {characterised by the tasks executed}
- B67D 7/0401: {arrangements for automatically fuelling vehicles, i.e. without human intervention}
- G06T 7/70: Determining position or orientation of objects or cameras (camera calibration G06T7/80)
- G06V 10/12: Details of acquisition arrangements; Constructional details thereof
- G06V 20/50: Context or environment of the image
- B67D 2007/0417: {Manipulator arms}
- B67D 2007/0474: {of the filling nozzle relative to the fuel tank opening, e.g. engagement between nozzle and tank opening}
- G06T 2207/10028: Range image; Depth image; 3D point clouds
- G06T 2207/20081: Training; Learning
- G06T 2207/20084: Artificial neural networks [ANN]
- G06V 10/764: using classification, e.g. of video objects
- G06V 10/774: Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
- G06V 10/82: using neural networks
- B25J: MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES ({manipulators specially adapted for use in surgery A61B34/70; manipulators used in cleaning hollow articles B08B9/04}; manipulators associated with rolling mills B21B39/20; manipulators associated with forging machines B21J13/10; {manipulators associated with picking-up and placing mechanisms B23P19/007}; means for holding wheels or parts thereof B60B30/00; {vehicles with ground-engaging propulsion means, e.g. walking members B62D57/02, B62D57/032; devices for picking-up and depositing articles or materials between conveyors B65G47/90, B65G47/91; manipulators with gripping or holding means for transferring packages B65H67/065}; cranes B66C; {manipulators used in the protection or supervision of pipe-line installations F17D5/00; walking equipment adapted for nuclear steam-generators F22B37/006}; manipulators specially adapted for, or associated with, nuclear reactors G21C; {apparatus used for handling wafers during manufacture or treatment of semiconductor H10P72/50})
- B67D: DISPENSING, DELIVERING OR TRANSFERRING LIQUIDS, NOT OTHERWISE PROVIDED FOR (cleaning pipes or tubes or systems of pipes or tubes B08B9/02; emptying or filling of bottles, jars, cans, casks, barrels, or similar containers, not otherwise provided for B67C; water supply E03; pipe systems F17D; domestic hot-water supply systems F24D; measuring volume, volume flow, mass flow or liquid level, metering by volume G01F; coin-freed or like apparatus G07F)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

Disclosed embodiments relate to systems and methods for automatically cleaning an object using a movable member comprising a camera and an engagement mechanism configured to clean the object. Techniques include positioning the movable member to a first orientation angle to capture image data using the camera; accessing a trained machine learning model associated with the object; identifying, based on the trained machine learning model and the captured image data, a portion of the object; positioning the movable member to a second orientation angle different from the first orientation angle; moving the engagement mechanism into a position according to the portion of the object; and causing cleaning of the portion of the object using the engagement mechanism.

## Claims

### Claim at position 1

1-20. (canceled)

### Claim 21

A computer-implemented method for automatically cleaning an object using a movable member comprising a camera and an engagement mechanism configured to clean the object, comprising:
positioning the movable member to a first orientation angle to capture image data using the camera;
accessing a trained machine learning model associated with the object;
identifying, based on the trained machine learning model and the captured image data, a portion of the object, wherein the trained machine learning model is configured to segment types of objects represented in the captured image data, and wherein identifying the portion of the object includes determining:
a position of the portion of the object relative to three perpendicular axes, and
an orientation of the portion of the object relative to each of the three perpendicular axes;
based on the determined position and orientation of the portion of the object, positioning the movable member to a second orientation angle different from the first orientation angle;
moving the engagement mechanism into a position according to the portion of the object, wherein the positioning of the movable member to the second orientation angle and moving the engagement mechanism into the position are based on the image data captured by the camera while the movable member is positioned in the first orientation angle without requiring additional image data captured by the camera after the movable member is moved from the first orientation angle; and
causing cleaning of the portion of the object using the engagement mechanism.

### Claim 22 (depends on 21)

The computer-implemented method of claim 21, wherein the movable member comprises a robotic arm.

### Claim 23 (depends on 21)

The computer-implemented method of claim 21, wherein the movable member comprises a mobile robotic device.

### Claim 24 (depends on 21)

The computer-implemented method of claim 21, wherein in the first orientation angle the engagement mechanism does not obstruct a field of view of the camera.

### Claim 25 (depends on 21)

The computer-implemented method of claim 21, wherein the camera and the engagement mechanism are positioned on surfaces of the movable member that differ by ninety degrees.

### Claim 26 (depends on 21)

The computer-implemented method of claim 21, wherein the camera and the engagement mechanism are positioned on surfaces of the movable member that differ by one-hundred and eighty degrees.

### Claim 27 (depends on 21)

The computer-implemented method of claim 21, wherein camera is configured to capture images from inside the object.

### Claim 28 (depends on 21)

The computer-implemented method of claim 21, wherein the movable member further comprises a chamber.

### Claim 29 (depends on 21)

The computer-implemented method of claim 21, wherein the chamber houses an electronic controller configured to control the camera.

### Claim 30 (depends on 21)

The computer-implemented method of claim 21, wherein the cleaning of the portion of the object using the engagement mechanism is based on the image data captured by the camera while the movable member is positioned in the first orientation angle without requiring additional image data captured by the camera after the movable member is moved from the first orientation angle.

### Claim 31

A non-transitory computer-readable medium storing instructions executable by at least one hardware processor to perform operations for automatically cleaning an object using a movable member comprising a camera and an engagement mechanism configured to clean the object, the operations comprising:
positioning the movable member to a first orientation angle to capture image data using the camera;
accessing a trained machine learning model associated with the object;
identifying, based on the trained machine learning model and the captured image data, a portion of the object, wherein the trained machine learning model is configured to segment types of objects represented in the captured image data, and wherein identifying the portion of the object includes determining:
a position of the portion of the object relative to three perpendicular axes, and
an orientation of the portion of the object relative to each of the three perpendicular axes;
based on the determined position and orientation of the portion of the object, positioning the movable member to a second orientation angle different from the first orientation angle;
moving the engagement mechanism into a position according to the portion of the object, wherein the positioning of the movable member to the second orientation angle and moving the engagement mechanism into the position are based on the image data captured by the camera while the movable member is positioned in the first orientation angle without requiring additional image data captured by the camera after the movable member is moved from the first orientation angle; and
causing cleaning of the portion of the object using the engagement mechanism.

### Claim 32 (depends on 31)

The non-transitory computer-readable medium of claim 31, wherein the camera is laser-based.

### Claim 33 (depends on 31)

The non-transitory computer-readable medium of claim 31, wherein the camera is time-of-flight based.

### Claim 34 (depends on 31)

The non-transitory computer-readable medium of claim 31, wherein the camera captures three-dimensional image data.

### Claim 35 (depends on 34)

The non-transitory computer-readable medium of claim 34, wherein the three-dimensional image data includes a point cloud.

### Claim 36 (depends on 31)

The non-transitory computer-readable medium of claim 31, wherein the engagement mechanism is pneumatically operated.

### Claim 37 (depends on 31)

The non-transitory computer-readable medium of claim 31, wherein the engagement mechanism is magnetically operated.

### Claim 38 (depends on 31)

The non-transitory computer-readable medium of claim 31, wherein the engagement mechanism is electrically operated.

### Claim 39 (depends on 31)

The non-transitory computer-readable medium of claim 31, wherein causing cleaning of the portion of the object using the engagement mechanism comprises manipulating a component of the engagement mechanism.

### Claim 40 (depends on 31)

The non-transitory computer-readable medium of claim 31, wherein causing cleaning of the portion of the object using the engagement mechanism comprises controlling a start and a stop of the cleaning of the portion of the object using the engagement mechanism.
