patent.fyi
Applications  /  US 2026/0354640
United States Patent Application Publication

US 2026/0354640 A1

Published
Nov 12, 2026
Filed
Jul 21, 2026
Earliest date on the face
Nov 17, 2017
Classification
A47L 9/281
First published here
Nov 19, 2026

Generated automatically. Not reviewed by a person. Not legal advice or an opinion.

Section 1 · machine-generated

What if?

Questions about how individual parts of claim 1 might be changed. A question may describe something different from what the claim language recites, or a version of it; no view is offered on which.

1.2Ceiling camera measures debris

What ifWhat if a fixed ceiling camera measured visible debris and sent its locations to the robot?
Claim language“dirt sensor arranged on the robot”

The camera processor compares floor images with a clean reference image and counts debris pixels in each floor grid cell. A wireless link sends the cell coordinates and debris counts to the robot.

1.2Operator enters soiling levels

What ifWhat if a person inspected the floor and entered the soiling level for each zone?
Claim language“a dirt sensor signal”

A wall panel displays a floor map and accepts one of three soiling levels for each zone. The robot retrieves the stored entries and selects a treatment setting when it enters each zone.

1.3Floor material selects suction

What ifWhat if the robot changed suction according to the floor material assigned to each zone?
Claim language“in response to the dirt sensor signal”

A stored map labels zones as hard floor or carpet, and wheel encoders and a range sensor locate the robot on that map. The controller sets a lower blower speed on hard floor and a higher speed on carpet.

1.3Inspection precedes treatment

What ifWhat if the robot set the treatment parameters from an inspection pass before floor treatment began?
Claim language“during the treatment of the floor surface”

A downward-facing camera records visible debris by grid cell while the blower and brush remain off. The controller sets a suction level for each cell before it starts the treatment pass and applies the stored levels along the route.

Section 2 · machine-generated

Variations

Further arrangements of the same subject matter, described here as of the publication date at the foot of this page.

Other field of use

Residue removal on greenhouse aisles

An autonomous mobile scrubber cleans sealed greenhouse aisles with a rotary brush, a water spray bar, and a trailing recovery squeegee. A downward-facing color camera mounted ahead of the spray bar uses diffuse illumination to measure the fraction of the brush path covered by spilled growing medium. Wheel encoder readings trigger an increase in water flow when the measured patch reaches the brush and restore baseline flow when the patch clears the brush.

Component substitution

Piezoelectric sensing in the intake duct

A piezoelectric impact sensor replaces the optical dirt sensor in a robotic vacuum. The sensor bonds to the dry side of a metal plate that forms the outer wall of an intake duct bend, where entrained particles strike the plate. A charge amplifier and band-pass filter feed a pulse counter that rejects low-frequency chassis vibration. The controller raises suction while the impact count per sampling interval exceeds a calibrated threshold.

Parameter or scale variation

Three independently loaded scrubbing heads

An autonomous scrubber carries three independently loaded 150 mm scrubbing heads across a 450 mm treatment swath. A downward-facing reflectance sensor sits 120 mm ahead of each head and samples its lane at 50 Hz. The controller uses wheel encoder distance to align each reading with the corresponding head position. A linear actuator raises the head load from 15 N to 30 N over patches above the calibrated soil threshold and restores the 15 N load after the head passes.

Combination with existing systems

Scrubber with an automatic service dock

An autonomous scrubber connects to a service dock with clean-water and detergent refill couplings and a separate wastewater drain. The robot carries a forward-facing reflectance sensor, separate water and detergent tanks, and a metering pump that injects detergent into the spray bar feed. During scrubbing, the controller increases detergent concentration as the measured soil level rises within the configured dilution range. The controller tracks dispensed volumes and sends the robot to the dock when either supply reaches its refill threshold.

Automated or data-driven variant

Image classification for treatment selection

A downward-facing camera and a shielded light source sit ahead of a treatment module with a dry roller, a spray bar, and a wet recovery head. An onboard classifier trained on labeled images assigns each floor patch to clean floor, loose particles, or adhered residue. The controller uses encoder distance to register each patch with the treatment module as the robot advances. It selects dry pickup for loose particles and enables spray and wet recovery for adhered residue.

Residual soil feedback for brush control

An autonomous dry sweeper carries downward-facing reflectance sensors ahead of the brush and behind the pickup slot. The controller uses wheel encoder readings to pair measurements of the same floor patch before and after treatment. It computes the fraction of measured soil remaining for patches whose initial soil reading exceeds a minimum threshold. A bounded feedback loop adjusts brush speed for upcoming patches using the rolling average of that fraction.

Specific version of a part

Dock mounted suction blower

A blower and dust separator in the dock draw air through a flexible hose connected to the nozzle. A motorized reel pays out the hose as the robot travels and takes up slack as it returns.

Powered roller and suction

A motor rotates a bristled roller that lifts debris from the floor into a suction inlet. A centrifugal blower draws the debris into a removable bin and passes the exhaust through a pleated filter.

Optical particle counting sensor

An infrared emitter and a photodiode face each other across the intake duct. The controller counts brief drops in received light over one-second intervals and uses the count as a measure of floor soiling.

Travel speed tracks debris

The controller reduces travel speed from 0.30 to 0.15 meters per second when the debris count exceeds an upper threshold, increasing the time the suction inlet spends over each part of the floor. It restores the higher speed after the count stays below a lower threshold for two seconds.

Section 3 · quoted from USPTO records

Other work

Passages reproduced word for word from two earlier-published U.S. patent documents, placed beside the language of claim 1. No characterization is offered. Each passage is marked A or B for the document it comes from; ¶ numbers are the paragraph numbers in the USPTO’s published full text of that document.

Document AUS 2011/0138558 A1
Published
Jun 16, 2011
Filed
Jul 24, 2009
Class
A46B 9/02
Document BUS 8,871,030 B2
Published
Oct 28, 2014
Filed
Jul 18, 2011
Class
G05D 1/0274
Claim 1
Document A
Document B
1.1controlling the robot in a treatment mode for treating a floor surface with a floor treatment module of the robot;
The automatic cleaning and washing, for example of a motor vehicle body, are carried out by rotatively driving a plurality of brush assemblies of the above mentioned type, so that the flexible strip elements, held in a present configuration thereof under the centrifugal force generated by the rotary bush assemblies, impact on the motor vehicle body, with an interposition of water and/or suitable washing solutions or emulsions.
A · US 2011/0138558 A1 · ¶ 4
1.2detecting a dirt sensor signal representing the level of soiling of the floor surface by means of a dirt sensor arranged on the robot;
The detection module 130 is used for detecting a flow of the dirt cleared away by the cleaning module 120 to obtain a dirt level of each grid.
B · US 8,871,030 B2 · ¶ 22
1.3adjusting the treatment mode in response to the dirt sensor signal during the treatment of the floor surface.
In a mode-switch guidance mechanism integrated with AI, the multiple cleaning path modes is still built in the automatic cleaning device, but a dirt detection mechanism is further set in the automatic cleaning device to detect the dirt distribution and the dirt level through various sensors.
B · US 8,871,030 B2 · ¶ 8

An empty cell means only that no passage was reproduced there.

Statements in these documents
However, this mode of operation negatively affects a proper duration of paint materials coating the motor vehicle body.
A · US 2011/0138558 A1 · ¶ 7
In fact, the foamed bristles, independently from the shapes thereof (of a star, cross, flat or round type), leave signs or marks of the operation and passage thereof, since they are rotatively driven in contact with the Vehicle painted bodies.
A · US 2011/0138558 A1 · ¶ 8
In this case, the automatic cleaning device may rarely or even never pass through a specific local block, thus resulting in poor cleaning in the specific local block.
B · US 8,871,030 B2 · ¶ 9
The automatic cleaning device plans an optimal cleaning strategy in the area to be cleaned in combination with the dirt detection mechanism, so as to ensure that each corner in the area to be cleaned can really be cleaned.
B · US 8,871,030 B2 · ¶ 10
Cite aspatent.fyi, US 2026/0354640 A1, claim 1, first published 2026-11-19T01:26:01-05:00, https://patent.fyi/US20260354640/, SHA-256 a7e3ea6120abefcee3f3ea54d74d77e91b241e585724ca03a8a96178195231dd
First published
Permanent addresshttps://patent.fyi/US20260354640/
Content fingerprint (SHA-256)a7e3ea6120abefcee3f3ea54d74d77e91b241e585724ca03a8a96178195231dd
Methodv2026-11-18.2 · automated, not reviewed by a person
Correctionscorrections@patent.fyi