WO2019174084A1 - 扫地方法、装置和扫地机器人 - Google Patents

扫地方法、装置和扫地机器人 Download PDF

Info

Publication number
WO2019174084A1
WO2019174084A1 PCT/CN2018/082043 CN2018082043W WO2019174084A1 WO 2019174084 A1 WO2019174084 A1 WO 2019174084A1 CN 2018082043 W CN2018082043 W CN 2018082043W WO 2019174084 A1 WO2019174084 A1 WO 2019174084A1
Authority
WO
WIPO (PCT)
Prior art keywords
carpet
image
sweeping
area
satisfied
Prior art date
Application number
PCT/CN2018/082043
Other languages
English (en)
French (fr)
Inventor
王声平
周毕兴
Original Assignee
深圳市沃特沃德股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳市沃特沃德股份有限公司 filed Critical 深圳市沃特沃德股份有限公司
Publication of WO2019174084A1 publication Critical patent/WO2019174084A1/zh

Links

Classifications

    • AHUMAN NECESSITIES
    • A47FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
    • A47LDOMESTIC WASHING OR CLEANING; SUCTION CLEANERS IN GENERAL
    • A47L11/00Machines for cleaning floors, carpets, furniture, walls, or wall coverings
    • A47L11/24Floor-sweeping machines, motor-driven
    • AHUMAN NECESSITIES
    • A47FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
    • A47LDOMESTIC WASHING OR CLEANING; SUCTION CLEANERS IN GENERAL
    • A47L11/00Machines for cleaning floors, carpets, furniture, walls, or wall coverings
    • A47L11/40Parts or details of machines not provided for in groups A47L11/02 - A47L11/38, or not restricted to one of these groups, e.g. handles, arrangements of switches, skirts, buffers, levers
    • AHUMAN NECESSITIES
    • A47FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
    • A47LDOMESTIC WASHING OR CLEANING; SUCTION CLEANERS IN GENERAL
    • A47L11/00Machines for cleaning floors, carpets, furniture, walls, or wall coverings
    • A47L11/40Parts or details of machines not provided for in groups A47L11/02 - A47L11/38, or not restricted to one of these groups, e.g. handles, arrangements of switches, skirts, buffers, levers
    • A47L11/4011Regulation of the cleaning machine by electric means; Control systems and remote control systems therefor
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/24Classification techniques
    • G06F18/241Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2413Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
    • G06F18/24133Distances to prototypes
    • G06F18/24137Distances to cluster centroïds
    • G06F18/2414Smoothing the distance, e.g. radial basis function networks [RBFN]
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/70Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/82Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/10Terrestrial scenes
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/35Categorising the entire scene, e.g. birthday party or wedding scene
    • G06V20/36Indoor scenes
    • AHUMAN NECESSITIES
    • A47FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
    • A47LDOMESTIC WASHING OR CLEANING; SUCTION CLEANERS IN GENERAL
    • A47L2201/00Robotic cleaning machines, i.e. with automatic control of the travelling movement or the cleaning operation
    • AHUMAN NECESSITIES
    • A47FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
    • A47LDOMESTIC WASHING OR CLEANING; SUCTION CLEANERS IN GENERAL
    • A47L2201/00Robotic cleaning machines, i.e. with automatic control of the travelling movement or the cleaning operation
    • A47L2201/04Automatic control of the travelling movement; Automatic obstacle detection
    • AHUMAN NECESSITIES
    • A47FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
    • A47LDOMESTIC WASHING OR CLEANING; SUCTION CLEANERS IN GENERAL
    • A47L2201/00Robotic cleaning machines, i.e. with automatic control of the travelling movement or the cleaning operation
    • A47L2201/06Control of the cleaning action for autonomous devices; Automatic detection of the surface condition before, during or after cleaning

Definitions

  • the present invention relates to the field of robot technology, and in particular to a method, a device and a cleaning robot.
  • the main object of the present invention is to provide a method, a device and a sweeping robot for sweeping the ground, aiming at solving the technical problem that the carpet or the robot is damaged during the sweeping process, and improving the intelligence of the robot.
  • the embodiment of the present invention provides a method for sweeping the ground, and the method includes the following steps:
  • the step of determining whether the avoidance condition is met includes:
  • the method further includes:
  • the step of determining whether the avoidance condition is met includes:
  • the step of determining whether the carpet is a plush carpet further comprises:
  • the current working mode is a mopping mode
  • the step of detecting whether the area to be cleaned has a carpet comprises:
  • the step of performing analysis processing on the image to obtain feature information includes:
  • the convolution feature map is input to the candidate region generation network for processing, and the feature information of the candidate region is obtained.
  • the step of performing convolutional neural network processing on the image further comprises: filtering out noise in the image.
  • the step of filtering out noise in the image comprises: filtering the image using a median filtering method to filter out noise in the image.
  • the step of acquiring the evaluation result output by the classifier further includes:
  • the location area of the carpet is marked on the positioning map.
  • the embodiment of the invention simultaneously provides a cleaning device, the device comprising:
  • a detecting module for detecting whether there is a carpet in the area to be cleaned
  • a judging module configured to determine whether the avoidance condition is satisfied when there is a carpet in the area to be cleaned
  • An evasion module for avoiding the area in which the carpet is located when the avoidance condition is met is met.
  • the determining module includes:
  • a first state determining unit configured to determine whether the current working mode is a mopping mode
  • the first determining unit is configured to determine that the avoidance condition is satisfied when in the mopping mode.
  • the determining module further includes:
  • a first carpet judging unit configured to determine whether the carpet is a long-staple carpet when not in the mopping mode
  • the second determining unit is configured to determine that the avoidance condition is satisfied when the carpet is a long-staple carpet.
  • the determining module includes:
  • a second carpet determining unit configured to determine whether the carpet is a long-staple carpet
  • the third determining unit is configured to determine that the avoidance condition is satisfied when the carpet is a long-staple carpet.
  • the determining module further includes:
  • a second state determining unit configured to determine whether the current working mode is a mopping mode when the carpet is not a long-staple carpet
  • the fourth determining unit is configured to determine that the avoidance condition is satisfied when it is in the mopping mode.
  • the detecting module includes:
  • An image acquisition unit configured to collect an image of the area to be cleaned
  • An analysis processing unit configured to perform analysis processing on the image to acquire feature information
  • An analysis judging unit configured to input the feature information into a classifier, and use the classifier to analyze whether the carpet to be cleaned has a carpet;
  • a result obtaining unit configured to obtain a judging result of the output of the classifier.
  • the analysis processing unit includes:
  • a first processing unit configured to perform convolutional neural network processing on the image to obtain a convolution feature map
  • a second processing unit configured to input the convolution feature map into the candidate region generation network for processing, to obtain feature information of the candidate region.
  • the detecting module further includes a noise filtering unit, wherein the noise filtering unit is configured to filter out noise in the image before performing analysis processing on the image.
  • the noise filtering unit is configured to: perform filtering processing on the image by using a median filtering method to filter out noise in the image.
  • the detecting module further includes a map marking unit, wherein the map marking unit is configured to: when the evaluation result is that the carpet to be cleaned has a carpet, mark a location area of the carpet on the positioning map.
  • Embodiments of the present invention also provide a cleaning robot including a memory, a processor, and at least one application stored in the memory and configured to be executed by the processor, the application being configured to be used for Perform the aforementioned sweeping method.
  • a sweeping method provided by an embodiment of the present invention detects and identifies a carpet by using a cleaning area.
  • special treatment is performed, such as avoiding the area where the carpet is located, thereby avoiding damage to the carpet or the robot.
  • the intelligence of the sweeping robot is improved and the user experience is improved.
  • FIG. 1 is a flow chart of an embodiment of a method of sweeping the present invention
  • FIG. 2 is a specific flow chart of the steps of detecting whether there is a carpet in the area to be cleaned in the embodiment of the present invention
  • FIG. 3 is a specific flowchart of a step of determining whether a avoidance condition is satisfied in an embodiment of the present invention
  • Figure 5 is a block diagram showing an embodiment of a sweeping device of the present invention.
  • Figure 6 is a block diagram of the detection module of Figure 5;
  • FIG. 7 is a block diagram of the analysis processing module of Figure 6;
  • FIG. 8 is another block diagram of the detection module of FIG. 5;
  • FIG. 9 is another block diagram of the detection module of FIG. 5;
  • FIG. 10 is a block diagram of the determination module of FIG. 5;
  • FIG. 11 is a block diagram of the determination module of FIG. 5.
  • step S12 Determine whether the avoidance condition is satisfied. When the avoidance condition is satisfied, the process proceeds to step S13; when the avoidance condition is not satisfied, the process is terminated and the sweeping robot continues to clean normally.
  • step S11 the sweeping robot scans the area to be cleaned in real time or periodically during the sweeping process, and detects whether there is a carpet in the area to be cleaned.
  • the carpet according to the embodiment of the present invention should be understood in a broad sense, and generally refers to all the mats, clothes and the like having fluff.
  • the cleaning robot can perform image detection using image processing technology.
  • the carpet recognition can be performed based on a Convolutional Neural Network (CNN).
  • CNN Convolutional Neural Network
  • the sweeping robot can use the RCNN, Fast RCNN and other algorithms for image processing to detect and identify the carpet.
  • S111 Collect an image of the area to be cleaned.
  • the sweeping robot can collect the image of the area to be cleaned in front of the real-time or timing by the camera during the sweeping process.
  • the cleaning robot first performs convolutional neural network processing on the acquired image to obtain a convolution feature map; then, the convolution feature map is input into the candidate region generation network (Region)
  • the Proposal Network, RPN performs processing to obtain feature information of the candidate region.
  • the cleaning robot can filter out the noise in the image before step S112.
  • the collected image may be filtered by a median filtering method to filter out noise in the image.
  • the median filtering method is a nonlinear filtering method capable of significantly suppressing noise and protecting a boundary.
  • S113 Enter the feature information into the classifier, and use the classifier to evaluate whether the area to be cleaned has a carpet.
  • the cleaning robot inputs the feature information of the candidate region into the pre-trained classifier, and the classifier performs similarity matching on the input feature information and the feature information of the carpet, and when the similarity reaches the threshold, it is determined.
  • the candidate area belongs to the carpet, thereby determining that the cleaning area has a carpet.
  • the classifier may further judge whether the fluff of the carpet is long fluff or short fluff, and when it is long fluff, the carpet is determined to be a long pile carpet, and when it is short fluff, the carpet is determined For short velvet carpets.
  • the sweeping robot obtains the evaluation result of the classifier output, and knows whether there is a carpet in the area to be cleaned, and whether it is a long-staple carpet or a short-staple carpet.
  • the cleaning robot further adjusts the position of the candidate area belonging to the carpet by using a regression device, converts the position and size of the carpet into coordinates by the coordinate conversion, and converts the position of the ground coordinate system, and marks the position area of the carpet on the positioning map, and outlines The shape and size of the carpet.
  • step S12 when there is a carpet in the area to be cleaned, the sweeping robot determines whether the avoidance condition is satisfied.
  • the sweeping robot can judge whether the avoidance condition is satisfied according to the current working mode and/or the type of the carpet.
  • the working mode includes the mopping mode and the sweeping mode, and the types of the carpet include a long pile carpet and a short pile carpet.
  • the specific process of the cleaning robot determining whether the avoidance condition is satisfied is as follows:
  • step S101 Determine whether the current working mode is a mopping mode. When it is not the mopping mode, but the cleaning mode, the process proceeds to step S102; when it is the mopping mode, the process proceeds to step S103.
  • step S102 Determine whether the carpet is a long-staple carpet. When it is a long pile carpet, it progresses to step S103; When it is not a long pile carpet, it progresses to step S104.
  • the sweeping robot In the mopping mode, since the sweeping robot carries the water tank, it will wet the carpet, so whether the carpet is a long-staple carpet or a short-staple carpet, the sweeping robot should avoid the carpet. When in the sweep mode, the sweeping robot can clean the short-staple carpet, but avoid the long-staple carpet. Because the fluff of the long-staple carpet is long, the sweeping robot will wrap the fluff and the fluff of the carpet will fall off, which will also cause the sweeping robot. Directly stuck on the carpet.
  • step S102 may also be omitted.
  • the avoidance condition is not satisfied (suitable for a case where only a short pile carpet has no long pile carpet in the home).
  • the specific process of the cleaning robot determining whether the avoidance condition is satisfied is as follows:
  • step S201 Determine whether the carpet is a long-staple carpet. When it is not a long pile carpet, it proceeds to step S202; when it is a long pile carpet, it proceeds to step S203.
  • step S202 Determine whether the current working mode is a mopping mode. When it is the mopping mode, it proceeds to step S203; when it is not the mopping mode, but the cleaning mode, it proceeds to step S204.
  • the sweeping robot should avoid the carpet regardless of whether it is in the sweep mode or the mopping mode.
  • the sweeping robot avoids the carpet if it is in the mopping mode, preventing the carpet from getting wet, and does not need to avoid if it is in the cleaning mode.
  • step S202 may also be omitted.
  • the avoidance condition is not satisfied (suitable for the case where the sweeping robot has only one working mode of the cleaning mode).
  • step S13 when the avoidance condition is satisfied, the sweeping robot avoids the area where the carpet is located.
  • the sweeping robot defines the location area of the carpet marked on the map as a forbidden zone, does not enter the restricted zone during the sweeping process, and automatically bypasses the restricted zone.
  • the method for sweeping the earth in the embodiment of the present invention detects and identifies the carpet by using the cleaning area.
  • special treatment is performed, such as avoiding the area where the carpet is located, thereby avoiding damage to the carpet or the robot, and improving the situation.
  • the intelligence of the sweeping robot enhances the user experience.
  • an embodiment of the present invention includes a detection module 10, a determination module 20, and an avoidance module 30, wherein: the detection module 10 is configured to detect whether there is a carpet in the area to be cleaned; the determination module 20, It is used to determine whether the avoidance condition is satisfied when the carpet to be cleaned has a carpet; the avoidance module 30 is configured to avoid the area where the carpet is located when the avoidance condition is satisfied.
  • the detecting module 10 scans the area to be cleaned in real time or periodically, and detects whether there is a carpet in the area to be cleaned.
  • the detecting module 10 may perform image detection using image processing technology, for example, based on a Convolutional Neural Network (CNN) for carpet recognition.
  • CNN Convolutional Neural Network
  • the detection module 10 can utilize RCNN, Fast Algorithms such as RCNN perform image processing to detect and identify carpets.
  • the detection module 10 includes an image acquisition unit 11, an analysis processing unit 12, an analysis evaluation unit 13, and a result acquisition unit 14, wherein: the image acquisition unit 11 is configured to collect an image of the area to be cleaned; the analysis processing unit 12 And the analyzing unit 13 is configured to input the feature information into the classifier, and the classifier analyzes whether the carpet to be cleaned has a carpet; the result obtaining unit 14 is configured to obtain the classification. The result of the evaluation of the output of the device.
  • the image capturing unit 11 can collect the image of the area to be cleaned in front of the camera through the camera in real time or at a time.
  • the analysis processing unit 12 includes a first processing unit 121 and a second processing unit 122, as shown in FIG. 7, wherein: the first processing unit 121 is configured to perform convolutional neural network processing on the image to obtain a volume.
  • the second processing unit 122 is configured to input the convolution feature map into the candidate region generation network to obtain feature information of the candidate region.
  • the analysis judging unit 13 inputs the feature information of the candidate region into the pre-trained classifier, and the classifier performs similarity matching on the input feature information and the feature information of the carpet. When the similarity reaches the threshold, Then, it is determined that the candidate area belongs to the carpet, thereby determining that the cleaning area has a carpet.
  • the classifier may further judge whether the fluff of the carpet is long fluff or short fluff, and when it is long fluff, the carpet is determined to be a long pile carpet, and when it is short fluff, the carpet is determined For short velvet carpets.
  • the result obtaining unit 14 obtains the evaluation result of the classifier output, and knows whether there is a carpet in the area to be cleaned, and whether it is a long pile carpet or a short pile carpet.
  • the detecting module 10 may further include a noise filtering unit 15 for filtering out noise in the image before performing analysis processing on the image.
  • the noise filtering unit 15 may filter the collected image by using a median filtering method to filter out noise in the image, and the median filtering method can significantly suppress noise and protect the boundary. Nonlinear filtering method.
  • the detecting module 10 may further include a map marking unit 16 for marking the location area of the carpet on the positioning map when the evaluation result is that the carpet to be cleaned has a carpet. .
  • the map marking unit 16 can further adjust the position of the candidate area belonging to the carpet by the regression device, convert the position and size of the carpet into coordinates by the coordinates, convert the position into the ground coordinate system, and mark the position area of the carpet on the positioning map to outline the position.
  • the shape and size of the carpet can be adjusted to adjust the position of the candidate area belonging to the carpet by the regression device, convert the position and size of the carpet into coordinates by the coordinates, convert the position into the ground coordinate system, and mark the position area of the carpet on the positioning map to outline the position. The shape and size of the carpet.
  • the determining module 20 determines whether the avoidance condition is satisfied.
  • the judging module 20 can judge whether the avoidance condition is satisfied according to the current working mode and/or the kind of the carpet.
  • the working mode includes a mopping mode and a sweeping mode, and the type of the carpet includes a long pile carpet and a short pile carpet.
  • the determining module 20 includes a first state determining unit 21, a first determining unit 22, a first carpet determining unit 23, and a second determining unit 24, wherein: the first state determining unit 21 uses Determining whether the current working mode is the mopping mode; the first determining unit 22 is configured to determine that the avoiding condition is satisfied when the mopping mode is used; and the first carpet judging unit 23 is configured to: when not in the mopping mode, It is judged whether the carpet is a long-staple carpet; and the second determining unit 24 is configured to determine that the avoidance condition is satisfied when the carpet is a long-staple carpet.
  • the sweeping robot In the mopping mode, since the sweeping robot carries the water tank, it will wet the carpet, so whether the carpet is a long-staple carpet or a short-staple carpet, the sweeping robot should avoid the carpet. When in the sweep mode, the sweeping robot can clean the short-staple carpet, but avoid the long-staple carpet. Because the fluff of the long-staple carpet is long, the sweeping robot will wrap the fluff and the fluff of the carpet will fall off, which will also cause the sweeping robot. Directly stuck on the carpet.
  • the first carpet determining unit 23 and the second determining unit 24 may also be omitted.
  • the first determining unit 22 directly determines that the avoiding condition is not satisfied (suitable for only the short pile in the home) There is no carpet in the carpet.)
  • the determining module 20 includes a second carpet determining unit 27, a third determining unit 26, a second state determining unit 25, and a fourth determining unit 28, wherein: the second carpet determining unit 27 uses For determining whether the carpet is a long-staple carpet; the third determining unit 26 is configured to determine that the avoidance condition is satisfied when the carpet is a long-staple carpet; and the second state determining unit 25 is configured to determine the current when the carpet is not a long-staple carpet. Whether the working mode is the mopping mode; the fourth determining unit 28 is configured to determine that the avoiding condition is satisfied when it is the mopping mode.
  • the sweeping robot should avoid the carpet regardless of whether it is in the sweep mode or the mopping mode.
  • the sweeping robot avoids the carpet if it is in the mopping mode, preventing the carpet from getting wet, and does not need to avoid if it is in the cleaning mode.
  • the second state determining unit 25 and the fourth determining unit 28 may also be omitted.
  • the third determining unit 26 directly determines that the avoiding condition is not satisfied (suitable for the sweeping robot only the sweeping mode) The situation of this kind of working mode).
  • the avoidance module 30 when the avoidance condition is satisfied, avoids the area where the carpet is located.
  • the avoidance module 30 defines a location area of the carpet marked on the map as a forbidden zone, and controls the sweeping robot to not enter the forbidden zone during the sweeping process, and automatically bypasses the restricted zone.
  • the cleaning device of the embodiment of the present invention performs the detection and recognition of the carpet by the cleaning area.
  • special treatment is performed, such as avoiding the area where the carpet is located, thereby avoiding damage to the carpet or the robot, and improving the situation.
  • the intelligence of the sweeping robot enhances the user experience.
  • the invention also proposes a cleaning robot comprising a memory, a processor and at least one application stored in the memory and configured to be executed by the processor, the application being configured to perform a sweeping method.
  • the method for sweeping the ground includes the following steps: detecting whether there is a carpet in the area to be cleaned; determining whether the avoiding condition is satisfied when there is a carpet in the area to be cleaned; and avoiding the area where the carpet is located when the avoiding condition is satisfied.
  • the method of sweeping in the present embodiment is the method of sweeping the ground in the above embodiments of the present invention, and details are not described herein again.
  • the present invention includes apparatus that is directed to performing one or more of the operations described herein. These devices may be specially designed and manufactured for the required purposes, or may also include known devices in a general purpose computer. These devices have computer programs stored therein that are selectively activated or reconfigured.
  • Such computer programs may be stored in a device (eg, computer) readable medium or in any type of medium suitable for storing electronic instructions and coupled to a bus, respectively, including but not limited to any Types of disks (including floppy disks, hard disks, optical disks, CD-ROMs, and magneto-optical disks), ROM (Read-Only Memory), RAM (Random Access Memory), EPROM (Erasable Programmable) Read-Only Memory, EEPROM (Electrically Erasable) Programmable Read-Only Memory, flash memory, magnetic card or light card.
  • a readable medium includes any medium that is stored or transmitted by a device (eg, a computer) in a readable form.
  • each block of the block diagrams and/or block diagrams and/or flow diagrams and combinations of blocks in the block diagrams and/or block diagrams and/or flow diagrams can be implemented by computer program instructions. .
  • these computer program instructions can be implemented by a general purpose computer, a professional computer, or a processor of other programmable data processing methods, such that the processor is executed by a computer or other programmable data processing method.
  • steps, measures, and solutions in the various operations, methods, and processes that have been discussed in the present invention may be alternated, changed, combined, or deleted. Further, other steps, measures, and schemes of the various operations, methods, and processes that have been discussed in the present invention may be alternated, modified, rearranged, decomposed, combined, or deleted. Further, the steps, measures, and solutions in the prior art having various operations, methods, and processes disclosed in the present invention may also be alternated, changed, rearranged, decomposed, combined, or deleted.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Databases & Information Systems (AREA)
  • Computing Systems (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Electric Vacuum Cleaner (AREA)

Abstract

一种扫地方法、装置和扫地机器人,所述方法包括以下步骤:检测待清扫区域是否有地毯(S11);当待清扫区域有地毯时,判断是否满足避开条件(S12);当满足避开条件时,避开地毯所在区域(S13)。所提供的一种扫地方法,通过对待清扫区域进行地毯的检测和识别,当检测到有地毯时,则进行特殊处理,比如避开地毯所在区域,从而避免出现损坏地毯或机器人的情形,提高了扫地机器人的智能化程度,提升了用户体验。

Description

扫地方法、装置和扫地机器人 技术领域
本发明涉及机器人技术领域,特别是涉及到一种扫地方法、装置和扫地机器人。
背景技术
随着扫地机器人的智能化程度逐渐提高,其清洁效率得到了极大的改善,越来越多的家庭开始使用扫地机器人。同时,家庭环境的复杂性的提升,也使扫地机器人面临着越来越多的挑战。
目前,越来越多的家庭铺设了地毯。当扫地机器携带水箱拖地时,仍然会爬上地毯,将地毯弄湿,导致地毯被损坏。而有些地毯是长绒地毯,当扫地机器人在打扫模式下爬上长绒地毯时,还会将地毯的绒毛缠绕起来而使地毯的绒毛脱落,同时绒毛也会导致机器人被卡死。
因此,如何避免扫地过程中地毯或机器人被损坏,是当前亟需解决的技术问题。
技术问题
本发明的主要目的为提供一种扫地方法、装置和扫地机器人,旨在解决扫地过程中地毯或机器人被损坏的技术问题,提高机器人的智能化程度。
技术解决方案
为达以上目的,本发明实施例提出一种扫地方法,所述方法包括以下步骤:
检测待清扫区域是否有地毯;
当待清扫区域有地毯时,判断是否满足避开条件;
当满足避开条件时,避开所述地毯所在区域。
可选地,所述判断是否满足避开条件的步骤包括:
判断当前的工作模式是否为拖地模式;
当为拖地模式时,判定满足避开条件。
可选地,所述判断当前的工作模式是否为拖地模式的步骤之后还包括:
当不为拖地模式时,判断所述地毯是否为长绒地毯;
当所述地毯为长绒地毯时,判定满足避开条件。
可选地,所述判断是否满足避开条件的步骤包括:
判断所述地毯是否为长绒地毯;
当所述地毯为长绒地毯时,判定满足避开条件。
可选地,所述判断所述地毯是否为长绒地毯的步骤之后还包括:
当所述地毯不是长绒地毯时,判断当前的工作模式是否为拖地模式;
当为拖地模式时,判定满足避开条件。
可选地,所述检测待清扫区域是否有地毯的步骤包括:
采集待清扫区域的图像;
对所述图像进行分析处理,获取特征信息;
将所述特征信息输入分类器,通过所述分类器分析评判所述待清扫区域是否有地毯;
获取所述分类器输出的评判结果。
可选地,所述对所述图像进行分析处理,获取特征信息的步骤包括:
对所述图像进行卷积神经网络处理,得到卷积特征图;
将所述卷积特征图输入候选区域生成网络进行处理,得到候选区域的特征信息。
可选地,所述对所述图像进行卷积神经网络处理的步骤之前还包括:滤除所述图像中的噪声。
可选地,所述滤除所述图像中的噪声的步骤包括:利用中值滤波法对所述图像进行滤波处理,以滤除所述图像中的噪声。
可选地,所述获取所述分类器输出的评判结果的步骤之后还包括:
当所述评判结果为所述待清扫区域有地毯时,在定位地图上标记出地毯的位置区域。
本发明实施例同时提出一种扫地装置,所述装置包括:
检测模块,用于检测待清扫区域是否有地毯;
判断模块,用于当待清扫区域有地毯时,判断是否满足避开条件;
躲避模块,用于当满足避开条件时,避开所述地毯所在区域。
可选地,所述判断模块包括:
第一状态判断单元,用于判断当前的工作模式是否为拖地模式;
第一判定单元,用于当为拖地模式时,判定满足避开条件。
可选地,所述判断模块还包括:
第一地毯判断单元,用于当不为拖地模式时,判断所述地毯是否为长绒地毯;
第二判定单元,用于当所述地毯为长绒地毯时,判定满足避开条件。
可选地,所述判断模块包括:
第二地毯判断单元,用于判断所述地毯是否为长绒地毯;
第三判定单元,用于当所述地毯为长绒地毯时,判定满足避开条件。
可选地,所述判断模块还包括:
第二状态判断单元,用于当所述地毯不是长绒地毯时,判断当前的工作模式是否为拖地模式;
第四判定单元,用于当为拖地模式时,判定满足避开条件。
可选地,所述检测模块包括:
图像采集单元,用于采集待清扫区域的图像;
分析处理单元,用于对所述图像进行分析处理,获取特征信息;
分析评判单元,用于将所述特征信息输入分类器,通过所述分类器分析评判所述待清扫区域是否有地毯;
结果获取单元,用于获取所述分类器输出的评判结果。
可选地,所述分析处理单元包括:
第一处理单元,用于对所述图像进行卷积神经网络处理,得到卷积特征图;
第二处理单元,用于将所述卷积特征图输入候选区域生成网络进行处理,得到候选区域的特征信息。
可选地,所述检测模块还包括噪声滤除单元,所述噪声滤除单元用于:在对所述图像进行分析处理之前,滤除所述图像中的噪声。
可选地,所述噪声滤除单元用于:利用中值滤波法对所述图像进行滤波处理,以滤除所述图像中的噪声。
可选地,所述检测模块还包括地图标记单元,所述地图标记单元用于:当所述评判结果为所述待清扫区域有地毯时,在定位地图上标记出地毯的位置区域。
本发明实施例还提出一种扫地机器人,其包括存储器、处理器和至少一个被存储在所述存储器中并被配置为由所述处理器执行的应用程序,所述应用程序被配置为用于执行前述扫地方法。
有益效果
本发明实施例所提供的一种扫地方法,通过对待清扫区域进行地毯的检测和识别,当检测到有地毯时,则进行特殊处理,比如避开地毯所在区域,从而避免出现损坏地毯或机器人的情形,提高了扫地机器人的智能化程度,提升了用户体验。
附图说明
图1是本发明的扫地方法一实施例的流程图;
图2是本发明实施例中检测待清扫区域是否有地毯的步骤的具体流程图;
图3是本发明实施例中判断是否满足避开条件的步骤的具体流程图;
图4是本发明实施例中判断是否满足避开条件的步骤的又一具体流程图;
图5是本发明的扫地装置一实施例的模块示意图;
图6是图5中的检测模块的模块示意图;
图7是图6中的分析处理模块的模块示意图;
图8是图5中的检测模块的又一模块示意图;
图9是图5中的检测模块的又一模块示意图;
图10是图5中的判断模块的模块示意图;
图11是图5中的判断模块的模块示意图。
本发明目的的实现、功能特点及优点将结合实施例,参照附图做进一步说明。
本发明的最佳实施方式
应当理解,此处所描述的具体实施例仅仅用以解释本发明,并不用于限定本发明。
下面详细描述本发明的实施例,所述实施例的示例在附图中示出,其中自始至终相同或类似的标号表示相同或类似的元件或具有相同或类似功能的元件。下面通过参考附图描述的实施例是示例性的,仅用于解释本发明,而不能解释为对本发明的限制。
本技术领域技术人员可以理解,除非特意声明,这里使用的单数形式“一”、“一个”、“所述”和“该”也可包括复数形式。应该进一步理解的是,本发明的说明书中使用的措辞“包括”是指存在所述特征、整数、步骤、操作、元件和/或组件,但是并不排除存在或添加一个或多个其他特征、整数、步骤、操作、元件、组件和/或它们的组。应该理解,当我们称元件被“连接”或“耦接”到另一元件时,它可以直接连接或耦接到其他元件,或者也可以存在中间元件。此外,这里使用的“连接”或“耦接”可以包括无线连接或无线耦接。这里使用的措辞“和/或”包括一个或更多个相关联的列出项的全部或任一单元和全部组合。
本技术领域技术人员可以理解,除非另外定义,这里使用的所有术语(包括技术术语和科学术语),具有与本发明所属领域中的普通技术人员的一般理解相同的意义。还应该理解的是,诸如通用字典中定义的那些术语,应该被理解为具有与现有技术的上下文中的意义一致的意义,并且除非像这里一样被特定定义,否则不会用理想化或过于正式的含义来解释。
参照图1,提出本发明的扫地方法一实施例,所述方法应用于扫地机器人,所述方法包括以下步骤:
S11、检测待清扫区域是否有地毯。当待清扫区域有地毯时,进入下一步骤S12。
S12、判断是否满足避开条件。当满足避开条件时,进入步骤S13;当不满足避开条件时,结束流程,扫地机器人继续正常清扫。
S13、避开地毯所在区域。
步骤S11中,扫地机器人在扫地过程中,实时或定时的扫描待清扫区域,检测待清扫区域是否有地毯。本发明实施例所述的地毯应做广义理解,泛指所有具有绒毛的垫子、衣物等物品。
本发明实施例中,扫地机器人可以采用图像处理技术进行地毯检测,比如可以基于卷积神经网络(Convolutional Neural Network,CNN)来进行地毯识别。在具体实施时,扫地机器人可以利用RCNN、Fast RCNN等算法进行图像处理,检测识别地毯。
如图2所示,扫地机器人检测待清扫区域是否有地毯的具体流程如下:
S111、采集待清扫区域的图像。
本发明实施例中,扫地机器人在扫地过程中,可以通过摄像头实时或定时的采集前方待清扫区域的图像。
S112、对图像进行分析处理,获取特征信息。
本发明实施例中,扫地机器人首先对采集的图像进行卷积神经网络处理,得到卷积特征图(feature map);然后将卷积特征图输入候选区域生成网络(Region Proposal Network, RPN)进行处理,得到候选区域的特征信息。
进一步地,在采集图像时,由于受摄像头及外部环境的影响,图像中会引入噪声,为提高后续算法处理的准确度,在步骤S112之前,扫地机器人还可以滤除图像中的噪声。可选地,可以利用中值滤波法对采集的图像进行滤波处理来滤除图像中的噪声,中值滤波法是能够明显抑制噪声且能保护边界的非线性滤波方法。
S113、将特征信息输入分类器,通过分类器分析评判待清扫区域是否有地毯。
本发明实施例中,扫地机器人将候选区域的特征信息输入到预先训练好的分类器中,分类器对输入的特征信息与地毯的特征信息进行相似度匹配,当相似度达到阈值时,则判定候选区域属于地毯,从而判定清扫区域有地毯。
进一步地,当候选区域属于地毯时,分类器还可以进一步评判该地毯的绒毛是长绒毛还是短绒毛,当是长绒毛时则认定该地毯为长绒地毯,当是短绒毛时则认定该地毯为短绒地毯。
S114、获取分类器输出的评判结果。
扫地机器人获取分类器输出的评判结果,得知待清扫区域是否有地毯,以及是长绒地毯还是短绒地毯。
进一步地,扫地机器人还对属于地毯的候选区域用回归器进一步调整其位置,将地毯位置及大小通过坐标转换,换算成地面坐标系的位置,并在定位地图上标记出地毯的位置区域,勾勒出地毯的形状和大小。
步骤S12中,当待清扫区域有地毯时,扫地机器人则判断是否满足避开条件。扫地机器人可以根据当前的工作模式和/或地毯的种类来判断是否满足避开条件,工作模式包括拖地模式和打扫模式,地毯的种类包括长绒地毯和短绒地毯。
如图3所示,在某些实施例中,扫地机器人判断是否满足避开条件的具体流程如下:
S101、判断当前的工作模式是否为拖地模式。当不为拖地模式,而是打扫模式时,则进入步骤S102;当为拖地模式时,进入步骤S103。
S102、判断地毯是否为长绒地毯。当为长绒地毯时,进入步骤S103;当不为长绒地毯时,进入步骤S104。
S103、判定满足避开条件。
S104、判定不满足避开条件。
当为拖地模式时,由于扫地机器人携带了水箱,会弄湿地毯,因此无论该地毯是长绒地毯还是短绒地毯,扫地机器人都要避开地毯。当为打扫模式时,扫地机器人可以打扫短绒地毯,但要避开长绒地毯,因为长绒地毯的绒毛较长,扫地机器人会将绒毛缠绕起来而使地毯的绒毛脱落,还会导致扫地机器人直接卡死在地毯上。
在某些实施例中,也可以省略步骤S102,当不为拖地模式时,直接判定不满足避开条件(适合家庭中只有短绒地毯没有长绒地毯的情形)。
如图4所示,在另一些实施例中,扫地机器人判断是否满足避开条件的具体流程如下:
S201、判断地毯是否为长绒地毯。当不为长绒地毯时,进入步骤S202;当为长绒地毯时,进入步骤S203。
S202、判断当前的工作模式是否为拖地模式。当为拖地模式时,进入步骤S203;当不为拖地模式,而是打扫模式时,则进入步骤S204。
S203、判定满足避开条件。
S204、判定不满足避开条件。
当地毯为长绒地毯时,无论处于打扫模式还是拖地模式,扫地机器人都要避开地毯。当地毯为短绒地毯时,扫地机器人如果处于拖地模式则避开地毯,防止将地毯弄湿,如果处于打扫模式则无需避开。
在某些实施例中,也可以省略步骤S202,当不为长绒地毯时,直接判定不满足避开条件(适合扫地机器人只有打扫模式这一种工作模式的情形)。
步骤S13中,当满足避开条件时,扫地机器人则避开地毯所在区域。例如,扫地机器人将定位地图上标记的地毯的位置区域定义为禁区,扫地过程中不进入禁区,自动绕开禁区。
本发明实施例的扫地方法,通过对待清扫区域进行地毯的检测和识别,当检测到有地毯时,则进行特殊处理,比如避开地毯所在区域,从而避免出现损坏地毯或机器人的情形,提高了扫地机器人的智能化程度,提升了用户体验。
参照图5,提出本发明的扫地装置一实施例,所述装置包括检测模块10、判断模块20和躲避模块30,其中:检测模块10,用于检测待清扫区域是否有地毯;判断模块20,用于当待清扫区域有地毯时,判断是否满足避开条件;躲避模块30,用于当满足避开条件时,避开地毯所在区域。
在扫地机器人扫地过程中,检测模块10实时或定时的扫描待清扫区域,检测待清扫区域是否有地毯。本发明实施例中,检测模块10可以采用图像处理技术进行地毯检测,比如可以基于卷积神经网络(Convolutional Neural Network,CNN)来进行地毯识别。在具体实施时,检测模块10可以利用RCNN、Fast RCNN等算法进行图像处理,检测识别地毯。
如图6所示,检测模块10包括图像采集单元11、分析处理单元12、分析评判单元13和结果获取单元14,其中:图像采集单元11,用于采集待清扫区域的图像;分析处理单元12,用于对采集的图像进行分析处理,获取特征信息;分析评判单元13,用于将特征信息输入分类器,通过分类器分析评判待清扫区域是否有地毯;结果获取单元14,用于获取分类器输出的评判结果。
本发明实施例中,在扫地机器人扫地过程中,图像采集单元11可以通过摄像头实时或定时的采集前方待清扫区域的图像。
本发明实施例中,分析处理单元12如图7所示,包括第一处理单元121和第二处理单元122,其中:第一处理单元121,用于对图像进行卷积神经网络处理,得到卷积特征图;第二处理单元122,用于将卷积特征图输入候选区域生成网络进行处理,得到候选区域的特征信息。
本发明实施例中,分析评判单元13将候选区域的特征信息输入到预先训练好的分类器中,分类器对输入的特征信息与地毯的特征信息进行相似度匹配,当相似度达到阈值时,则判定候选区域属于地毯,从而判定清扫区域有地毯。
进一步地,当候选区域属于地毯时,分类器还可以进一步评判该地毯的绒毛是长绒毛还是短绒毛,当是长绒毛时则认定该地毯为长绒地毯,当是短绒毛时则认定该地毯为短绒地毯。
结果获取单元14获取分类器输出的评判结果,得知待清扫区域是否有地毯,以及是长绒地毯还是短绒地毯。
进一步地,如图8所示,检测模块10还可以包括噪声滤除单元15,该噪声滤除单元15用于:在对图像进行分析处理之前,滤除图像中的噪声。
在采集图像时,由于受摄像头及外部环境的影响,图像中会引入噪声。因此,为提高后续算法处理的准确度,噪声滤除单元15可以利用中值滤波法对采集的图像进行滤波处理来滤除图像中的噪声,中值滤波法是能够明显抑制噪声且能保护边界的非线性滤波方法。
更进一步地,如图9所示,检测模块10还可以包括地图标记单元16,该地图标记单元16用于:当评判结果为待清扫区域有地毯时,在定位地图上标记出地毯的位置区域。
地图标记单元16可以对属于地毯的候选区域用回归器进一步调整其位置,将地毯位置及大小通过坐标转换,换算成地面坐标系的位置,并在定位地图上标记出地毯的位置区域,勾勒出地毯的形状和大小。
本发明实施例中,当待清扫区域有地毯时,判断模块20则判断是否满足避开条件。判断模块20可以根据当前的工作模式和/或地毯的种类来判断是否满足避开条件,工作模式包括拖地模式和打扫模式,地毯的种类包括长绒地毯和短绒地毯。
可选地,如图10所示,判断模块20包括第一状态判断单元21、第一判定单元22、第一地毯判断单元23和第二判定单元24,其中:第一状态判断单元21,用于判断当前的工作模式是否为拖地模式;第一判定单元22,用于当为拖地模式时,判定满足避开条件;第一地毯判断单元23,用于当不为拖地模式时,判断地毯是否为长绒地毯;第二判定单元24,用于当地毯为长绒地毯时,判定满足避开条件。
当为拖地模式时,由于扫地机器人携带了水箱,会弄湿地毯,因此无论该地毯是长绒地毯还是短绒地毯,扫地机器人都要避开地毯。当为打扫模式时,扫地机器人可以打扫短绒地毯,但要避开长绒地毯,因为长绒地毯的绒毛较长,扫地机器人会将绒毛缠绕起来而使地毯的绒毛脱落,还会导致扫地机器人直接卡死在地毯上。
在某些实施例中,也可以省略第一地毯判断单元23和第二判定单元24,当不为拖地模式时,第一判定单元22直接判定不满足避开条件(适合家庭中只有短绒地毯没有长绒地毯的情形)。
可选地,如图11所示,判断模块20包括第二地毯判断单元27、第三判定单元26、第二状态判断单元25和第四判定单元28,其中:第二地毯判断单元27,用于判断地毯是否为长绒地毯;第三判定单元26,用于当地毯为长绒地毯时,判定满足避开条件;第二状态判断单元25,用于当地毯不是长绒地毯时,判断当前的工作模式是否为拖地模式;第四判定单元28,用于当为拖地模式时,判定满足避开条件。
当地毯为长绒地毯时,无论处于打扫模式还是拖地模式,扫地机器人都要避开地毯。当地毯为短绒地毯时,扫地机器人如果处于拖地模式则避开地毯,防止将地毯弄湿,如果处于打扫模式则无需避开。
在某些实施例中,也可以省略第二状态判断单元25和第四判定单元28,当不为长绒地毯时,第三判定单元26直接判定不满足避开条件(适合扫地机器人只有打扫模式这一种工作模式的情形)。
本发明实施例中,当满足避开条件时,躲避模块30则避开地毯所在区域。例如,躲避模块30将定位地图上标记的地毯的位置区域定义为禁区,控制扫地机器人扫地过程中不进入禁区,自动绕开禁区。
本发明实施例的扫地装置,通过对待清扫区域进行地毯的检测和识别,当检测到有地毯时,则进行特殊处理,比如避开地毯所在区域,从而避免出现损坏地毯或机器人的情形,提高了扫地机器人的智能化程度,提升了用户体验。
本发明同时提出一种扫地机器人,其包括存储器、处理器和至少一个被存储在存储器中并被配置为由处理器执行的应用程序,所述应用程序被配置为用于执行扫地方法。所述扫地方法包括以下步骤:检测待清扫区域是否有地毯;当待清扫区域有地毯时,判断是否满足避开条件;当满足避开条件时,避开地毯所在区域。本实施例中所描述的扫地方法为本发明中上述实施例所涉及的扫地方法,在此不再赘述。
本领域技术人员可以理解,本发明包括涉及用于执行本申请中所述操作中的一项或多项的设备。这些设备可以为所需的目的而专门设计和制造,或者也可以包括通用计算机中的已知设备。这些设备具有存储在其内的计算机程序,这些计算机程序选择性地激活或重构。这样的计算机程序可以被存储在设备(例如,计算机)可读介质中或者存储在适于存储电子指令并分别耦联到总线的任何类型的介质中,所述计算机可读介质包括但不限于任何类型的盘(包括软盘、硬盘、光盘、CD-ROM、和磁光盘)、ROM(Read-Only Memory,只读存储器)、RAM(Random Access Memory,随机存储器)、EPROM(Erasable Programmable Read-Only Memory,可擦写可编程只读存储器)、EEPROM(Electrically Erasable Programmable Read-Only Memory,电可擦可编程只读存储器)、闪存、磁性卡片或光线卡片。也就是,可读介质包括由设备(例如,计算机)以能够读的形式存储或传输信息的任何介质。
本技术领域技术人员可以理解,可以用计算机程序指令来实现这些结构图和/或框图和/或流图中的每个框以及这些结构图和/或框图和/或流图中的框的组合。本技术领域技术人员可以理解,可以将这些计算机程序指令提供给通用计算机、专业计算机或其他可编程数据处理方法的处理器来实现,从而通过计算机或其他可编程数据处理方法的处理器来执行本发明公开的结构图和/或框图和/或流图的框或多个框中指定的方案。
本技术领域技术人员可以理解,本发明中已经讨论过的各种操作、方法、流程中的步骤、措施、方案可以被交替、更改、组合或删除。进一步地,具有本发明中已经讨论过的各种操作、方法、流程中的其他步骤、措施、方案也可以被交替、更改、重排、分解、组合或删除。进一步地,现有技术中的具有与本发明中公开的各种操作、方法、流程中的步骤、措施、方案也可以被交替、更改、重排、分解、组合或删除。
以上所述仅为本发明的优选实施例,并非因此限制本发明的专利范围,凡是利用本发明说明书及附图内容所作的等效结构或等效流程变换,或直接或间接运用在其他相关的技术领域,均同理包括在本发明的专利保护范围内。

Claims (17)

  1. 一种扫地方法,其特征在于,包括以下步骤:
    检测待清扫区域是否有地毯;
    当待清扫区域有地毯时,判断是否满足避开条件;
    当满足避开条件时,避开所述地毯所在区域。
  2. 根据权利要求1所述的扫地方法,其特征在于,所述判断是否满足避开条件的步骤包括:
    判断当前的工作模式是否为拖地模式;
    当为拖地模式时,判定满足避开条件;
    当不为拖地模式时,判断所述地毯是否为长绒地毯;
    当所述地毯为长绒地毯时,判定满足避开条件。
  3. 根据权利要求1所述的扫地方法,其特征在于,所述判断是否满足避开条件的步骤包括:
    判断所述地毯是否为长绒地毯;
    当所述地毯为长绒地毯时,判定满足避开条件;
    当所述地毯不是长绒地毯时,判断当前的工作模式是否为拖地模式;
    当为拖地模式时,判定满足避开条件。
  4. 根据权利要求1-3任一项所述的扫地方法,其特征在于,所述检测待清扫区域是否有地毯的步骤包括:
    采集待清扫区域的图像;
    对所述图像进行分析处理,获取特征信息;
    将所述特征信息输入分类器,通过所述分类器分析评判所述待清扫区域是否有地毯;
    获取所述分类器输出的评判结果。
  5. 根据权利要求4所述的扫地方法,其特征在于,所述对所述图像进行分析处理,获取特征信息的步骤包括:
    对所述图像进行卷积神经网络处理,得到卷积特征图;
    将所述卷积特征图输入候选区域生成网络进行处理,得到候选区域的特征信息。
  6. 根据权利要求5所述的扫地方法,其特征在于,所述对所述图像进行卷积神经网络处理的步骤之前还包括:滤除所述图像中的噪声。
  7. 根据权利要求6所述的扫地方法,其特征在于,所述滤除所述图像中的噪声的步骤包括:利用中值滤波法对所述图像进行滤波处理,以滤除所述图像中的噪声。
  8. 根据权利要求4所述的扫地方法,其特征在于,所述获取所述分类器输出的评判结果的步骤之后还包括:
    当所述评判结果为所述待清扫区域有地毯时,在定位地图上标记出地毯的位置区域。
  9. 一种扫地装置,其特征在于,包括:
    检测模块,用于检测待清扫区域是否有地毯;
    判断模块,用于当待清扫区域有地毯时,判断是否满足避开条件;
    躲避模块,用于当满足避开条件时,避开所述地毯所在区域。
  10. 根据权利要求9所述的扫地装置,其特征在于,所述判断模块包括:
    第一状态判断单元,用于判断当前的工作模式是否为拖地模式;
    第一判定单元,用于当为拖地模式时,判定满足避开条件;
    第一地毯判断单元,用于当不为拖地模式时,判断所述地毯是否为长绒地毯;
    第二判定单元,用于当所述地毯为长绒地毯时,判定满足避开条件。
  11. 根据权利要求9所述的扫地装置,其特征在于,所述判断模块包括:
    第二地毯判断单元,用于判断所述地毯是否为长绒地毯;
    第三判定单元,用于当所述地毯为长绒地毯时,判定满足避开条件;
    第二状态判断单元,用于当所述地毯不是长绒地毯时,判断当前的工作模式是否为拖地模式;
    第四判定单元,用于当为拖地模式时,判定满足避开条件。
  12. 根据权利要求9-11任一项所述的扫地装置,其特征在于,所述检测模块包括:
    图像采集单元,用于采集待清扫区域的图像;
    分析处理单元,用于对所述图像进行分析处理,获取特征信息;
    分析评判单元,用于将所述特征信息输入分类器,通过所述分类器分析评判所述待清扫区域是否有地毯;
    结果获取单元,用于获取所述分类器输出的评判结果。
  13. 根据权利要求12所述的扫地装置,其特征在于,所述分析处理单元包括:
    第一处理单元,用于对所述图像进行卷积神经网络处理,得到卷积特征图;
    第二处理单元,用于将所述卷积特征图输入候选区域生成网络进行处理,得到候选区域的特征信息。
  14. 根据权利要求13所述的扫地装置,其特征在于,所述检测模块还包括噪声滤除单元,所述噪声滤除单元用于:在对所述图像进行分析处理之前,滤除所述图像中的噪声。
  15. 根据权利要求14所述的扫地装置,其特征在于,所述噪声滤除单元用于:利用中值滤波法对所述图像进行滤波处理,以滤除所述图像中的噪声。
  16. 根据权利要求12所述的扫地装置,其特征在于,所述检测模块还包括地图标记单元,所述地图标记单元用于:当所述评判结果为所述待清扫区域有地毯时,在定位地图上标记出地毯的位置区域。
  17. 一种扫地机器人,包括存储器、处理器和至少一个被存储在所述存储器中并被配置为由所述处理器执行的应用程序,其特征在于,所述应用程序被配置为用于执行权利要求1至5任一项所述的扫地方法。
PCT/CN2018/082043 2018-03-14 2018-04-04 扫地方法、装置和扫地机器人 WO2019174084A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810210156.0 2018-03-14
CN201810210156.0A CN108514381A (zh) 2018-03-14 2018-03-14 扫地方法、装置和扫地机器人

Publications (1)

Publication Number Publication Date
WO2019174084A1 true WO2019174084A1 (zh) 2019-09-19

Family

ID=63433664

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/082043 WO2019174084A1 (zh) 2018-03-14 2018-04-04 扫地方法、装置和扫地机器人

Country Status (2)

Country Link
CN (1) CN108514381A (zh)
WO (1) WO2019174084A1 (zh)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110673597A (zh) * 2019-09-20 2020-01-10 深圳乐动机器人有限公司 一种控制机器人回充的方法及机器人
CN110710935A (zh) * 2019-11-14 2020-01-21 北京富通芯科技有限公司 扫地机器人、地面检测方法和智能家居系统
CN112847383A (zh) * 2019-11-28 2021-05-28 深圳津梁生活科技有限公司 机器人
CN114259187A (zh) * 2021-12-15 2022-04-01 华帝股份有限公司 一种清洁设备控制方法、清洁设备
WO2022117107A1 (zh) * 2020-12-04 2022-06-09 苏州宝时得电动工具有限公司 清洁机器人、清洁系统及清洁方法
CN114652235A (zh) * 2022-03-02 2022-06-24 深圳市杉川机器人有限公司 扫地机器人及其控制方法以及存储介质
CN115024659A (zh) * 2022-06-22 2022-09-09 宁波国琅机器人科技有限公司 一种扫地机器人湿扫控制方法、系统、存储介质及智能终端

Families Citing this family (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111166238A (zh) * 2018-11-09 2020-05-19 北京奇虎科技有限公司 清洁禁区的处理方法、装置、设备及存储介质
CN109846427A (zh) * 2019-01-16 2019-06-07 深圳乐动机器人有限公司 一种清洁机器人的控制方法及清洁机器人
CN111568306A (zh) * 2019-02-19 2020-08-25 北京奇虎科技有限公司 基于清洁机器人的清洁方法、装置、电子设备及存储介质
CN110018239B (zh) * 2019-04-04 2022-07-08 珠海一微半导体股份有限公司 一种地毯检测方法
CN110547727A (zh) * 2019-08-14 2019-12-10 深圳市银星智能科技股份有限公司 清洁机器人的流体施加方法及清洁机器人
CN112493924B (zh) * 2019-08-26 2023-03-10 苏州宝时得电动工具有限公司 清洁机器人及其控制方法
CN110522360A (zh) * 2019-09-05 2019-12-03 深圳市杉川机器人有限公司 地毯检测方法、装置、扫地机器人和计算机存储介质
CN110870721A (zh) * 2019-11-26 2020-03-10 上海高仙自动化科技发展有限公司 清洁机器人的控制方法、装置、清洁机器人及存储介质
CN111035327B (zh) * 2019-12-31 2024-01-30 上海飞科电器股份有限公司 清洁机器人、地毯检测方法及计算机可读存储介质
CN111275112A (zh) * 2020-01-20 2020-06-12 上海高仙自动化科技发展有限公司 机器人控制方法、机器人和可读存储介质
CN112022003A (zh) * 2020-08-21 2020-12-04 苏州三六零机器人科技有限公司 扫地机器人及其控制方法、装置、计算机可读存储介质
CN111973079B (zh) * 2020-08-21 2022-06-10 苏州三六零机器人科技有限公司 拖地绕行区域设置方法、扫地机器人、设备及计算机可读介质
CN112353295A (zh) * 2020-11-04 2021-02-12 深圳市普森斯科技有限公司 基于地毯识别的扫地机清扫方法、电子装置及存储介质
CN114587204A (zh) * 2021-07-05 2022-06-07 北京石头世纪科技股份有限公司 清洁方法及清洁控制装置
CN114587210B (zh) * 2021-11-16 2023-06-20 北京石头创新科技有限公司 清洁机器人控制方法及控制装置
CN114652217B (zh) * 2022-03-02 2023-10-27 美智纵横科技有限责任公司 控制方法、清洁机器人和存储介质
CN114711671B (zh) * 2022-06-08 2022-09-02 深圳市瑞亿科技电子有限公司 一种可对动作行为进行判断的机器人防护系统
CN118203264A (zh) * 2022-12-16 2024-06-18 珠海一微半导体股份有限公司 一种机器人检测地毯的方法、避障方法、机器人和芯片

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20160090567A (ko) * 2015-01-22 2016-08-01 주식회사 파인로보틱스 로봇 청소기 및 로봇 청소기 제어 방법
CN106264357A (zh) * 2016-08-30 2017-01-04 宁波菜鸟智能科技有限公司 扫地机器人的地毯判定方法及系统
CN206403708U (zh) * 2016-08-31 2017-08-15 科沃斯机器人股份有限公司 清洁机器人
CN107049155A (zh) * 2017-04-25 2017-08-18 北京小米移动软件有限公司 扫地机器人的控制方法和装置
CN107569181A (zh) * 2016-07-04 2018-01-12 九阳股份有限公司 一种智能清洁机器人及清扫方法
CN107788913A (zh) * 2016-08-31 2018-03-13 科沃斯机器人股份有限公司 清洁机器人及其控制方法

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102613944A (zh) * 2012-03-27 2012-08-01 复旦大学 清洁机器人脏物识别系统及清洁方法
JP6680453B2 (ja) * 2012-12-05 2020-04-15 フォルヴェルク・ウント・ツェーオー、インターホールディング・ゲーエムベーハーVorwerk & Compagnie Interholding Gesellshaft Mit Beschrankter Haftung 走行式掃除機並びにこのような装置の運転方法
CN105511478B (zh) * 2016-02-23 2019-11-26 百度在线网络技术(北京)有限公司 应用于扫地机器人的控制方法、扫地机器人及终端

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20160090567A (ko) * 2015-01-22 2016-08-01 주식회사 파인로보틱스 로봇 청소기 및 로봇 청소기 제어 방법
CN107569181A (zh) * 2016-07-04 2018-01-12 九阳股份有限公司 一种智能清洁机器人及清扫方法
CN106264357A (zh) * 2016-08-30 2017-01-04 宁波菜鸟智能科技有限公司 扫地机器人的地毯判定方法及系统
CN206403708U (zh) * 2016-08-31 2017-08-15 科沃斯机器人股份有限公司 清洁机器人
CN107788913A (zh) * 2016-08-31 2018-03-13 科沃斯机器人股份有限公司 清洁机器人及其控制方法
CN107049155A (zh) * 2017-04-25 2017-08-18 北京小米移动软件有限公司 扫地机器人的控制方法和装置

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110673597A (zh) * 2019-09-20 2020-01-10 深圳乐动机器人有限公司 一种控制机器人回充的方法及机器人
CN110710935A (zh) * 2019-11-14 2020-01-21 北京富通芯科技有限公司 扫地机器人、地面检测方法和智能家居系统
CN112847383A (zh) * 2019-11-28 2021-05-28 深圳津梁生活科技有限公司 机器人
WO2022117107A1 (zh) * 2020-12-04 2022-06-09 苏州宝时得电动工具有限公司 清洁机器人、清洁系统及清洁方法
CN114259187A (zh) * 2021-12-15 2022-04-01 华帝股份有限公司 一种清洁设备控制方法、清洁设备
CN114652235A (zh) * 2022-03-02 2022-06-24 深圳市杉川机器人有限公司 扫地机器人及其控制方法以及存储介质
CN115024659A (zh) * 2022-06-22 2022-09-09 宁波国琅机器人科技有限公司 一种扫地机器人湿扫控制方法、系统、存储介质及智能终端
CN115024659B (zh) * 2022-06-22 2024-03-29 宁波国琅机器人科技有限公司 一种扫地机器人湿扫控制方法、系统、存储介质及智能终端

Also Published As

Publication number Publication date
CN108514381A (zh) 2018-09-11

Similar Documents

Publication Publication Date Title
WO2019174084A1 (zh) 扫地方法、装置和扫地机器人
US11042760B2 (en) Mobile robot, control method and control system thereof
CN110025268B (zh) 一种顽固污渍的清扫方法
CN107569181B (zh) 一种智能清洁机器人及清扫方法
CN106894481B (zh) 一种基于智能马桶自动检测并清洁污垢方法及系统
US8924019B2 (en) Cleaning robot, dirt recognition device thereof and cleaning method of robot
CN112716376A (zh) 基于扫地机的清洁方法、装置、可读存储介质及电子设备
CN112274065B (zh) 清洁机器人及其控制方法
CN111643014A (zh) 智能清洁方法、装置、智能清洁设备和存储介质
US9877628B2 (en) Vacuum cleaner, control method thereof, and computer readable recording medium
WO2019174080A1 (zh) 扫地机器人及其扫地方法和装置
CN105467985B (zh) 自移动表面行走机器人及其图像处理方法
US11222207B2 (en) Intelligent sweeping robot
CN112056991A (zh) 机器人的主动清洁方法、装置、机器人和存储介质
CN113633221A (zh) 自动清洁设备漏扫区域处理方法、装置和系统
CN109363581A (zh) 一种清洁机器人的吸尘控制方法、芯片及清洁机器人
KR100820740B1 (ko) 자동 주행 청소기 및 그 제어방법
WO2024109590A1 (zh) 清洁设备控制方法及装置、介质及电子设备
KR20210007360A (ko) 이동 로봇 및 그 제어방법
CN111166238A (zh) 清洁禁区的处理方法、装置、设备及存储介质
CN108338748A (zh) 一种基于视频监控的自动清洁方法、装置以及清洁机器人
KR102504411B1 (ko) 오염물 인식장치
CN114451816A (zh) 清扫策略生成方法、装置、计算机设备和存储介质
CN114639003A (zh) 基于人工智能的工地车辆清洁度判断方法、设备及介质
WO2021184665A1 (zh) 自动工作系统、自动行走设备及其控制方法及计算机可读存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18910183

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18910183

Country of ref document: EP

Kind code of ref document: A1