IP Library Granted Patent US 12,733,782
Granted Patent B2
US 12,733,782 · App. 18/122,280 · Granted Sep 15, 2026

Vacuum cleaner and method for controlling the same

Inventors: Hyunkoo Kang (Suwon-si, KR); Juhyuk Kim (Suwon-si, KR); Sanghwa Choi (Suwon-si, KR); Seehyun Kim (Suwon-si, KR); Sanghyuk Park (Suwon-si, KR); Jinwook Yoon (Suwon-si, KR); Dongseok Lee (Suwon-si, KR); Seongu Lee (Suwon-si, KR); Jeonghee Cho (Suwon-si, KR); Seungryong Cha (Suwon-si, KR)
Assignee: SAMSUNG ELECTRONICS CO., LTD.
A47L9/2821A47L5/30A47L9/0411A47L9/0477A47L9/2826A47L9/2831A47L9/2842A47L9/2847
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,733,782
App. No.
18/122,280
Granted
Sep 15, 2026
Kind
B2
Abstract

A vacuum cleaner including a main body; a suction head including a suction port through which debris is sucked up; a brush configured to rotate in the suction head; a brush motor configured to rotate the brush; a suction motor configured to generate suction force so that debris is sucked up through the suction port; a pressure sensor configured to measure atmospheric pressure and pressure at the suction port; and at least one processor configured to determine suction pressure based on the measured atmospheric pressure and the measured pressure at the suction port, and determine cleaner state information, including at least one of operation state information and information of a type of surface to be cleaned, based on the determined suction pressure and a load of the brush motor.

Claims (59)

1 . A vacuum cleaner comprising:

a main body;

a suction head including a suction port through which debris is sucked up;

a brush configured to rotate in the suction head;

a brush motor configured to rotate the brush;

a suction motor configured to generate suction force so that debris is sucked up through the suction port;

a pressure sensor configured to measure atmospheric pressure and pressure at the suction port;

a current sensor configured to measure current supplied to the brush motor;

a memory configured to store trained data of a support vector machine (SVM), wherein the trained data includes correlation information in form of suction pressure values stored in correspondence with brush motor load values, such that the correlation information represent a two-dimensional graph of the suction pressure values and the brush motor load values, resulting in the trained data further representing a plurality of information regions determinable by the SVM in the two-dimensional graph, an information region among the plurality of information regions corresponding to a different cleaner state among a plurality of cleaner states based upon the information region corresponding to a region of suction pressure values and a region of brush motor load values among the suction pressure values and the brush motor load values in the correlation information; and

at least one processor configured to:

determine a currently determined suction pressure at the suction port based on the atmospheric pressure and the pressure measured by the pressure sensor,

determine a currently determined load of the brush motor based on the current measured by the current sensor,

determine a currently determined cleaner state among the plurality of cleaner states, based on the currently determined suction pressure, the currently determined load of the brush motor and the correlation information stored in the trained data by identifying an information region among the plurality of information regions to which the currently determined suction pressure and the currently determined load of the brush motor belong, the currently determined cleaner state including at least one of an operation state or a type of surface to be cleaned, and

control an output of at least one motor among the brush motor and the suction motor, based on the currently determined cleaner state.

2 . The vacuum cleaner of claim 1 , wherein the pressure sensor includes a relative pressure sensor including:

a first pressure sensor configured to measure the atmospheric pressure, and

a second pressure sensor configured to measure the pressure at the suction port,

wherein to determine the currently determined suction pressure at the suction port, the relative pressure sensor is configured to output a difference between the atmospheric pressure measured by the first pressure sensor and the pressure at the suction port measured by the second pressure sensor.

3 . The vacuum cleaner of claim 1 , wherein

the pressure sensor includes an absolute pressure sensor configured to measure the pressure at the suction port, and

the at least one processor is configured to:

determine the atmospheric pressure based on a first output of the pressure sensor before operation of the suction motor, and

determine the pressure at the suction port based on a second output of the pressure sensor during operation of the suction motor.

4 . The vacuum cleaner of claim 1 , further comprising:

a voltage sensor configured to measure a voltage of a battery which supplies power to the brush motor,

wherein the at least one processor is configured to:

determine consumption power of the brush motor based on the current supplied to the brush motor and the voltage of the battery, and

determine the currently determined load of the brush motor in proportion to the consumption power of the brush motor.

5 . The vacuum cleaner of claim 1 , wherein

the plurality of information regions are delineated by a plurality of hyperplanes determined by the SVM based on the correlation information, each hyperplane among the plurality of hyperplanes being a boundary information plane in the trained data that separates the information region from among the plurality of information regions indicative of respective different cleaner states among the plurality of cleaner states differentiated by different suction pressure values among the suction pressure values and different brush motor load values among the brush motor load values stored in the correlation information,

the memory is further configured to store the currently determined suction pressure, the currently determined load of the brush motor, and the currently determined cleaner state as a previously determined suction pressure, a previously determined load of the brush motor, and a previously determined cleaner state, and

the at least one processor is configured to determine the plurality of hyperplanes by inputting the previously determined suction pressure and the previously determined load of the brush motor stored in the memory as training data to the SVM.

6 . The vacuum cleaner of claim 5 , wherein the plurality of hyperplanes are linear kernels.

7 . The vacuum cleaner of claim 5 , wherein

each hyperplane among the plurality of hyperplanes is a hyperplane having a maximum separation distance in the trained data with the suction pressure values and the brush motor load values as attributes.

8 . The vacuum cleaner of claim 1 , wherein

the at least one processor is configured to, when the currently determined cleaner state indicates a lift state in which the brush is separated from a surface to be cleaned and the currently determined suction pressure is equal to or less than a set value, determine a subsequent cleaner state by further considering current rotations per minute (rpm) of the suction motor in addition to the currently determined suction pressure at the suction port and the currently determined load of the brush motor.

9 . A method of controlling a vacuum cleaner including a main body, a suction head including a suction port through which debris is sucked up, a brush configured to rotate in the suction head, a brush motor configured to rotate the brush, a suction motor configured to generate suction force so that debris is sucked up through the suction port, a pressure sensor configured to measure atmospheric pressure and pressure at the suction port, a current sensor configured to measure current supplied to the brush motor, at least one processor and a memory, the method comprising:

measuring, by the pressure sensor, atmospheric pressure and pressure at the suction port;

measuring, by the current sensor, a current supplied to the brush motor;

determining a currently determined suction pressure at the suction port based on the atmospheric pressure and the pressure measured by the pressure sensor;

determining a currently determined load of the brush motor based on the current measured by the current sensor;

determining a currently determined cleaner state among a plurality of cleaner states, based on the currently determined suction pressure, the currently determined load of the brush motor and trained data of a support vector machine (SVM) stored in the memory,

the trained data includes correlation information in form of suction pressure values stored in correspondence with brush motor load values, such that the correlation information represent a two-dimensional graph of the suction pressure values and the brush motor load values, resulting in the trained data further representing a plurality of information regions determinable by the SVM in the two-dimensional graph, an information region among the plurality of information regions corresponding to a different cleaner state among a plurality of cleaner states based upon the information region corresponding to a region of suction pressure values and a region of brush motor load values among the suction pressure values and the brush motor load values in the correlation information, such that the currently determined cleaner state is determined by identifying an information region among the plurality of information regions to which the currently determined suction pressure and the currently determined load of the brush motor belong, the currently determined cleaner state including at least one of an operation state or a type of surface to be cleaned; and

controlling an output of at least one motor among the brush motor and the suction motor based on the currently determined cleaner state.

10 . The method of claim 9 , wherein

the pressure sensor includes a relative pressure sensor including:

a first pressure sensor configured to measure the atmospheric pressure, and

a second pressure sensor configured to measure the pressure at the suction port,

wherein to determine the currently determined suction pressure at the suction port, the relative pressure sensor is configured to output a difference between the atmospheric pressure measured by the first pressure sensor and the pressure at the suction port measured by the second pressure sensor.

11 . The method of claim 9 , wherein

the pressure sensor includes an absolute pressure sensor configured to measure the pressure at the suction port, and

the determining of the currently determined suction pressure includes:

determining the atmospheric pressure based on first output of the pressure sensor before operation of the suction motor, and

determining the pressure at the suction port based on second output of the pressure sensor during operation of the suction motor.

12 . The method of claim 9 , wherein

the plurality of information regions are delineated by a plurality of hyperplanes determined by the SVM based on the correlation information, each hyperplane among the plurality of hyperplanes being a boundary information plane in the trained data that separates the information region from among the plurality of information regions indicative of respective different cleaner states among the plurality of cleaner states differentiated by different suction pressure values among the suction pressure values and different brush motor load values among the brush motor load values stored in the correlation information, the method further comprising:

storing, in the memory, the currently determined suction pressure, the currently determined load of the brush motor, and the currently determined cleaner state as a previously determined suction pressure, a previously determined load of the brush motor, and a previously determined cleaner state; and

determining, by the at least one processor, the plurality of hyperplanes by inputting the previously determined suction pressure and the previously determined load of the brush motor stored in the memory as training data to the SVM.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 16, 2023
From: KANG, HYUNKOO; KIM, JUHYUK; CHOI, SANGHWA; KIM, SEEHYUN; PARK, SANGHYUK; YOON, JINWOOK; LEE, DONGSEOK; LEE, SEONGU; CHO, JEONGHEE; CHA, SEUNGRYONG
To: SAMSUNG ELECTRONICS CO., LTD.
Reel/Frame 063001/0726 →
Priority Claims (2)
KR 10-2022-0059807 · May 16, 2022 · national
KR 10-2022-0179727 · Dec 20, 2022 · national
Continuity (2)
Continuation PCTKR2023003163 · Mar 8, 2023
Related Publication 20230363601A1 · Nov 16, 2023
References Cited (34)
US 4733431A · Martin · 1988 [cited by examiner]
US 5255409A · Fujiwara · 1993 [cited by examiner]
US 5722109A · Delmas · 1998 [cited by examiner]
US 5881430A · Driessen · 1999 [cited by examiner]
US 9763551B2 · Murchie et al. · 2017 [cited by applicant]
US 10178933B2 · Lee et al. · 2019 [cited by applicant]
US 12440078B2 · Kim · 2025 [cited by examiner]
US 12564301B2 · Choi · 2026 [cited by examiner]
US 20060085095A1 · Reindle · 2006 [cited by examiner]
US 20080034532A1 · Yang · 2008 [cited by applicant]
US 20130201316A1 · Binder · 2013 [cited by examiner]
US 20190387942A1 · Truitt et al. · 2019 [cited by applicant]
US 20200219507A1 · Jun · 2020 [cited by examiner]
US 20210007569A1 · Howard et al. · 2021 [cited by applicant]
US 20210186291A1 · Nguyen · 2021 [cited by examiner]
US 20210251451A1 · Ko et al. · 2021 [cited by applicant]
US 20230329499A1 · Lee · 2023 [cited by examiner]
US 20230363301P1 · Arts · 2023 [cited by examiner]
US 20240032755A1 · Choi · 2024 [cited by examiner]
US 20250134331A1 · Cho · 2025 [cited by examiner]
CN 1891132A · 2007 [cited by applicant]
GB 2572433B · 2020 [cited by applicant]
JP 759697A · 1995 [cited by applicant]
JP 2008220812A · 2008 [cited by applicant]
JP 2015130956A · 2015 [cited by applicant]
KR 1020150126046A · 2015 [cited by applicant]
KR 1020160075450 · 2016 [cited by applicant]
KR 1020210105207 · 2021 [cited by applicant]
WO WO2018202367A1 · 2018 [cited by applicant]
WO WO2022008871A1 · 2022 [cited by applicant]
WO WO2022008873A2 · 2022 [cited by applicant]
FR 2590471 (Year: 1987). [cited by examiner]
International Search Report dated Jun. 29, 2023 for International Application No. PCT/KR2023/003163. [cited by applicant]
Extended European Search Report dated Apr. 1, 2025 for European Application No. 23807753.1. [cited by applicant]