Invention content
The main object of the present invention is to provide one kind to sweep the floor method, apparatus and sweeping robot, it is intended to which solution is swept the floor process
The impaired technical problem of middle carpet or robot, improves the intelligence degree of robot.
To achieve these objectives, the embodiment of the present invention proposes that one kind is swept the floor method, the described method comprises the following steps:
Detect whether region to be cleaned has carpet;
When there is carpet in region to be cleaned, judges whether to meet and avoid condition;
When satisfaction avoids condition, the carpet region is avoided.
Optionally, described to judge whether that meeting the step of avoiding condition includes:
Judge whether current operating mode is the pattern of mopping floor;
When for mop floor pattern when, judgement meet avoids condition.
Optionally, described to judge whether current operating mode is the step of mopping floor pattern and further includes later:
When not being to mop floor pattern, judge whether the carpet is long based carpet;
When the carpet is long based carpet, judgement, which meets, avoids condition.
Optionally, described to judge whether that meeting the step of avoiding condition includes:
Judge whether the carpet is long based carpet;
When the carpet is long based carpet, judgement, which meets, avoids condition.
Optionally, described to judge further include after the step of whether carpet is long based carpet:
When the carpet is not long based carpet, judge whether current operating mode is the pattern of mopping floor;
When for mop floor pattern when, judgement meet avoids condition.
Optionally, whether detection region to be cleaned has the step of carpet to include:
Acquire the image in region to be cleaned;
Analyzing processing is carried out to described image, obtains characteristic information;
The characteristic information is inputted into grader, whether has ground by region to be cleaned described in the grader analysis and judgment
Blanket;
Obtain the evaluation result of the grader output.
Optionally, described the step of carrying out analyzing processing to described image, obtaining characteristic information, includes:
Convolutional neural networks processing is carried out to described image, obtains convolution characteristic pattern;
Convolution characteristic pattern input candidate region is generated network to handle, obtains the characteristic information of candidate region.
Optionally, further include before described the step of carrying out convolutional neural networks processing to described image:Filter out the figure
Noise as in.
Optionally, the step of noise filtered out in described image includes:Using median filtering method to described image into
Row is filtered, to filter out the noise in described image.
Optionally, further include after the step of evaluation result for obtaining the grader output:
When the evaluation result, which is the region to be cleaned, carpet, the position area of carpet is marked on positioning map
Domain.
The embodiment of the present invention proposes that a kind of Floor-sweeping device, described device include simultaneously:
Detection module, for detecting whether region to be cleaned has carpet;
Judgment module avoids condition for when there is carpet in region to be cleaned, judging whether to meet;
Hide module, for when satisfaction avoids condition, avoiding the carpet region.
Optionally, the judgment module includes:
First state judging unit, for judging whether current operating mode is the pattern of mopping floor;
First judging unit, for when for mop floor pattern when, judgement meet avoids condition.
Optionally, the judgment module further includes:
First carpet judging unit, for when not being to mop floor pattern, judging whether the carpet is long based carpet;
Second judging unit, for when the carpet is long based carpet, judgement satisfaction to avoid condition.
Optionally, the judgment module includes:
Second carpet judging unit, for judging whether the carpet is long based carpet;
Third judging unit, for when the carpet is long based carpet, judgement satisfaction to avoid condition.
Optionally, the judgment module further includes:
Second condition adjudgement unit, for when the carpet is not long based carpet, whether judging current operating mode
For the pattern of mopping floor;
4th judging unit, for when for mop floor pattern when, judgement meet avoids condition.
Optionally, the detection module includes:
Image acquisition units, the image for acquiring region to be cleaned;
Analysis and processing unit obtains characteristic information for carrying out analyzing processing to described image;
Analysis and judgment unit, for the characteristic information to be inputted grader, described in the grader analysis and judgment
Whether region to be cleaned has carpet;
As a result acquiring unit, the evaluation result for obtaining the grader output.
Optionally, the analysis and processing unit includes:
First processing units obtain convolution characteristic pattern for carrying out convolutional neural networks processing to described image;
Second processing unit is handled for convolution characteristic pattern input candidate region to be generated network, is waited
The characteristic information of favored area.
Optionally, the detection module further includes noise filtering unit, and the noise filtering unit is used for:To the figure
As before carrying out analyzing processing, filtering out the noise in described image.
Optionally, the noise filtering unit is used for:Described image is filtered using median filtering method, with filter
Except the noise in described image.
Optionally, the detection module further includes map label unit, and the map label unit is used for:When the judge
When as a result having carpet for the region to be cleaned, the band of position of carpet is marked on positioning map.
The embodiment of the present invention also proposes a kind of sweeping robot comprising memory, processor and at least one is stored
In the memory and it is configured as the application program executed by the processor, the application program is configurable for holding
The aforementioned method of sweeping the floor of row.
One kind that the embodiment of the present invention is provided is swept the floor method, and the detection and knowledge of carpet are carried out by treating purging zone
Not, when having detected carpet, then specially treated is carried out, for example avoids carpet region, to avoid the occurrence of damage carpet
Or the situation of robot, the intelligence degree of sweeping robot is improved, the user experience is improved.
Specific implementation mode
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
The embodiment of the present invention is described below in detail, examples of the embodiments are shown in the accompanying drawings, wherein from beginning to end
Same or similar label indicates same or similar element or element with the same or similar functions.Below with reference to attached
The embodiment of figure description is exemplary, and is only used for explaining the present invention, and is not construed as limiting the claims.
Those skilled in the art of the present technique are appreciated that unless expressly stated, singulative " one " used herein, " one
It is a ", " described " and "the" may also comprise plural form.It is to be further understood that is used in the specification of the present invention arranges
It refers to there are the feature, integer, step, operation, element and/or component, but it is not excluded that presence or addition to take leave " comprising "
Other one or more features, integer, step, operation, element, component and/or their group.It should be understood that when we claim member
Part is " connected " or when " coupled " to another element, it can be directly connected or coupled to other elements, or there may also be
Intermediary element.In addition, " connection " used herein or " coupling " may include being wirelessly connected or wirelessly coupling.It is used herein to arrange
Diction "and/or" includes that the whole of one or more associated list items or any cell are combined with whole.
Those skilled in the art of the present technique are appreciated that unless otherwise defined, all terms used herein (including technology art
Language and scientific terminology), there is meaning identical with the general understanding of the those of ordinary skill in fields of the present invention.Should also
Understand, those terms such as defined in the general dictionary, it should be understood that have in the context of the prior art
The consistent meaning of meaning, and unless by specific definitions as here, the meaning of idealization or too formal otherwise will not be used
To explain.
Referring to Fig.1, propose that one embodiment of method of sweeping the floor of the present invention, the method are applied to sweeping robot, the side
Method includes the following steps:
Whether S11, detection region to be cleaned have carpet.When there is carpet in region to be cleaned, into next step S12.
S12, judge whether to meet and avoid condition.When satisfaction avoids condition, S13 is entered step;When being unsatisfactory for avoiding item
When part, terminate flow, sweeping robot continues normal clean.
S13, carpet region is avoided.
In step S11, during sweeping the floor, in real time or the scanning region to be cleaned of timing, detection wait for clear sweeping robot
Sweep whether region has carpet.Carpet described in the embodiment of the present invention shall be understood in a broad sense, and refer to all tools cushion villous, clothing
Object and other items.
In the embodiment of the present invention, sweeping robot may be used image processing techniques and carry out carpet detection, such as can be with base
Carpet identification is carried out in convolutional neural networks (Convolutional Neural Network, CNN).In the specific implementation,
Sweeping robot can carry out image procossing, detection identification carpet using RCNN, Fast RCNN scheduling algorithms.
As shown in Fig. 2, sweeping robot detects whether region to be cleaned has the detailed process of carpet as follows:
S111, the image for acquiring region to be cleaned.
In the embodiment of the present invention, sweeping robot, can acquisition real-time by camera or timing during sweeping the floor
The image in front region to be cleaned.
S112, analyzing processing is carried out to image, obtains characteristic information.
In the embodiment of the present invention, sweeping robot carries out convolutional neural networks processing to the image of acquisition first, is rolled up
Product characteristic pattern (feature map);Then convolution characteristic pattern input candidate region is generated into network (Region Proposal
Network, RPN) it is handled, obtain the characteristic information of candidate region.
Further, when acquiring image, due to being influenced by camera and external environment, noise can be introduced in image,
To improve the accuracy of subsequent algorithm processing, before step S112, sweeping robot can also filter out the noise in image.It can
Selection of land can be filtered the image of acquisition using median filtering method to filter out the noise in image, median filtering method
It is that can obviously inhibit noise and the non-linear filtering method on boundary can be protected.
S113, characteristic information is inputted into grader, whether has carpet by grader analysis and judgment region to be cleaned.
In the embodiment of the present invention, the characteristic information of candidate region is input to advance trained grader by sweeping robot
In, grader carries out similarity mode to the characteristic information of input and the characteristic information of carpet, when similarity reaches threshold value, then
Judgement candidate region belongs to carpet, to judge that purging zone has carpet.
Further, when candidate region belongs to carpet, the villus that grader can also further judge the carpet is long
Villus or down then assert that the carpet is long based carpet when being long fine hair, then assert that the carpet is when being down
Low level loop pile carpet.
S114, the evaluation result for obtaining grader output.
Sweeping robot obtains the evaluation result of grader output, learns whether region to be cleaned has carpet, and be long
Based carpet or low level loop pile carpet.
Further, sweeping robot also further adjusts its position to the candidate region for belonging to carpet with device is returned, will
Carpet position and size are converted by coordinate, are converted into the position of earth axes, and carpet is marked on positioning map
The shapes and sizes of carpet are sketched the contours of in the band of position.
In step S12, when there is carpet in region to be cleaned, sweeping robot, which then judges whether to meet, avoids condition.It sweeps the floor
Robot can avoid condition, operating mode packet according to the type of current operating mode and/or carpet to determine whether meeting
The pattern of mopping floor and cleaning pattern are included, the type of carpet includes long based carpet and low level loop pile carpet.
As shown in figure 3, in certain embodiments, sweeping robot judges whether that satisfaction avoids the detailed process of condition such as
Under:
S101, judge whether current operating mode is the pattern of mopping floor.When mopping floor pattern when not being, but sweeping pattern,
Then enter step S102;When for mop floor pattern when, enter step S103.
S102, judge whether carpet is long based carpet.When for long based carpet, S103 is entered step;When not for long wool
When blanket, S104 is entered step.
S103, judgement, which meet, avoids condition.
S104, judgement are unsatisfactory for avoiding condition.
When for mop floor pattern when, since sweeping robot carries water tank, carpet can be got wet, therefore no matter the carpet is long
Based carpet or low level loop pile carpet, sweeping robot will avoid carpet.When to sweep pattern, sweeping robot can be swept short
Based carpet, but long based carpet is avoided, because the villus of long based carpet is longer, villus can be wound up and be made by sweeping robot
It is directly stuck on carpet to also result in sweeping robot for the villus shedding of carpet.
In certain embodiments, step S102 can also be omitted, when not being to mop floor pattern, directly judgement is unsatisfactory for avoiding
Condition (is suitble in family the only situation of the not long based carpet of low level loop pile carpet).
As shown in figure 4, in further embodiments, sweeping robot judges whether that satisfaction avoids the detailed process of condition such as
Under:
S201, judge whether carpet is long based carpet.When not being long based carpet, S202 is entered step;When for long wool
When blanket, S203 is entered step.
S202, judge whether current operating mode is the pattern of mopping floor.When for mop floor pattern when, enter step S203;When
It is not the pattern that mops floor, but when cleaning pattern, then enter step S204.
S203, judgement, which meet, avoids condition.
S204, judgement are unsatisfactory for avoiding condition.
When carpet is long based carpet, no matter it is in cleaning pattern or mops floor pattern, sweeping robot will avoid ground
Blanket.When carpet is low level loop pile carpet, sweeping robot avoids carpet if in the pattern that mops floor, and prevents from getting wet carpet, such as
Fruit is in cleaning pattern and is then not necessarily to avoid.
In certain embodiments, step S202 can also be omitted, when not being long based carpet, directly judgement is unsatisfactory for avoiding
Condition (be suitble to sweeping robot only have cleaning pattern a kind of this situation of operating mode).
In step S13, when satisfaction avoids condition, sweeping robot then avoids carpet region.For example, machine of sweeping the floor
The band of position of the carpet marked on positioning map is defined as forbidden zone by people, does not enter forbidden zone during sweeping the floor, automatically around lifting a ban
Area.
The method of sweeping the floor of the embodiment of the present invention carries out the detection and identification of carpet by treating purging zone, when detecting
When having carpet, then specially treated is carried out, for example avoid carpet region, to avoid the occurrence of the feelings of damage carpet or robot
Shape improves the intelligence degree of sweeping robot, and the user experience is improved.
With reference to Fig. 5, one embodiment of Floor-sweeping device of the present invention is proposed, described device includes detection module 10, judgment module
20 and hide module 30, wherein:Detection module 10, for detecting whether region to be cleaned has carpet;Judgment module 20, for working as
When there is carpet in region to be cleaned, judges whether to meet and avoid condition;Hide module 30, for when satisfaction avoids condition, avoiding
Carpet region.
During sweeping robot is swept the floor, detection module 10 is in real time or the scanning region to be cleaned of timing, detection wait for clear
Sweep whether region has carpet.In the embodiment of the present invention, detection module 10 may be used image processing techniques and carry out carpet detection, than
Such as convolutional neural networks (Convolutional Neural Network, CNN) can be based on to carry out carpet identification.Specific
When implementation, detection module 10 can carry out image procossing, detection identification carpet using RCNN, Fast RCNN scheduling algorithms.
As shown in fig. 6, detection module 10 includes image acquisition units 11, analysis and processing unit 12, analysis and judgment unit 13
With result acquiring unit 14, wherein:Image acquisition units 11, the image for acquiring region to be cleaned;Analysis and processing unit 12,
Analyzing processing is carried out for the image to acquisition, obtains characteristic information;Analysis and judgment unit 13, for dividing characteristic information input
Whether class device has carpet by grader analysis and judgment region to be cleaned;As a result acquiring unit 14, for obtaining grader output
Evaluation result.
In the embodiment of the present invention, during sweeping robot is swept the floor, image acquisition units 11 can pass through camera reality
When or timing acquisition in front of region to be cleaned image.
In the embodiment of the present invention, analysis and processing unit 12 is as shown in fig. 7, comprises first processing units 121 and second processing
Unit 122, wherein:First processing units 121 obtain convolution characteristic pattern for carrying out convolutional neural networks processing to image;The
Two processing units 122 handle for convolution characteristic pattern input candidate region to be generated network, obtain the feature of candidate region
Information.
In the embodiment of the present invention, the characteristic information of candidate region is input to trained point in advance by analysis and judgment unit 13
In class device, grader carries out similarity mode to the characteristic information of input and the characteristic information of carpet, when similarity reaches threshold value
When, then judge that candidate region belongs to carpet, to judge that purging zone has carpet.
Further, when candidate region belongs to carpet, the villus that grader can also further judge the carpet is long
Villus or down then assert that the carpet is long based carpet when being long fine hair, then assert that the carpet is when being down
Low level loop pile carpet.
As a result acquiring unit 14 obtains the evaluation result of grader output, learns whether region to be cleaned has carpet, and
It is long based carpet or low level loop pile carpet.
Further, as shown in figure 8, detection module 10 can also include noise filtering unit 15, the noise filtering unit
15 are used for:Before carrying out analyzing processing to image, the noise in image is filtered out.
When acquiring image, due to being influenced by camera and external environment, noise can be introduced in image.Therefore, to carry
The accuracy of high subsequent algorithm processing, noise filtering unit 15 can be filtered place using median filtering method to the image of acquisition
It manages to filter out the noise in image, median filtering method is can obviously to inhibit noise and can protect the nonlinear filtering side on boundary
Method.
Further, as shown in figure 9, detection module 10 can also include map label unit 16, the map label list
Member 16 is used for:When evaluation result, which is region to be cleaned, carpet, the band of position of carpet is marked on positioning map.
Map label unit 16 can further adjust its position to the candidate region for belonging to carpet with device is returned, by carpet
Position and size are converted by coordinate, are converted into the position of earth axes, and the position of carpet is marked on positioning map
The shapes and sizes of carpet are sketched the contours of in region.
In the embodiment of the present invention, when there is carpet in region to be cleaned, judgment module 20, which then judges whether to meet, avoids condition.
Judgment module 20 can avoid condition, Working mould according to the type of current operating mode and/or carpet to determine whether meeting
Formula includes pattern and the cleaning pattern of mopping floor, and the type of carpet includes long based carpet and low level loop pile carpet.
Optionally, as shown in Figure 10, judgment module 20 includes first state judging unit 21, the first judging unit 22, the
One carpet judging unit 23 and the second judging unit 24, wherein:First state judging unit 21, for judging current Working mould
Whether formula is the pattern of mopping floor;First judging unit 22, for when for mop floor pattern when, judgement meet avoids condition;First carpet
Judging unit 23, for when not being to mop floor pattern, judging whether carpet is long based carpet;Second judging unit 24, for working as
When carpet is long based carpet, judgement, which meets, avoids condition.
When for mop floor pattern when, since sweeping robot carries water tank, carpet can be got wet, therefore no matter the carpet is long
Based carpet or low level loop pile carpet, sweeping robot will avoid carpet.When to sweep pattern, sweeping robot can be swept short
Based carpet, but long based carpet is avoided, because the villus of long based carpet is longer, villus can be wound up and be made by sweeping robot
It is directly stuck on carpet to also result in sweeping robot for the villus shedding of carpet.
In certain embodiments, the first carpet judging unit 23 and the second judging unit 24 can also be omitted, when not be drag
When ground mode, the first judging unit 22 directly judges that being unsatisfactory for avoiding condition (is suitble to only low level loop pile carpet in family not have long wool
The situation of carpet).
Optionally, as shown in figure 11, judgment module 20 includes the second carpet judging unit 27, third judging unit 26, the
Two-state judging unit 25 and the 4th judging unit 28, wherein:Second carpet judging unit 27, for judging whether carpet is long
Based carpet;Third judging unit 26, for when carpet is long based carpet, judgement satisfaction to avoid condition;Second condition adjudgement list
Member 25, for when carpet is not long based carpet, judging whether current operating mode is the pattern of mopping floor;4th judging unit
28, for when for mop floor pattern when, judgement meet avoids condition.
When carpet is long based carpet, no matter it is in cleaning pattern or mops floor pattern, sweeping robot will avoid ground
Blanket.When carpet is low level loop pile carpet, sweeping robot avoids carpet if in the pattern that mops floor, and prevents from getting wet carpet, such as
Fruit is in cleaning pattern and is then not necessarily to avoid.
In certain embodiments, the second condition adjudgement unit 25 and the 4th judging unit 28 can also be omitted, when not be length
When based carpet, the directly judgement of third judging unit 26 be unsatisfactory for avoiding condition (be suitble to sweeping robot only have cleaning pattern this
The situation of kind operating mode).
In the embodiment of the present invention, when satisfaction avoids condition, hides module 30 and then avoid carpet region.For example, hiding
It keeps away module 30 and the band of position of the carpet marked on positioning map is defined as forbidden zone, control sweeping robot is swept the floor in the process not
Into forbidden zone, forbidden zone is got around automatically.
The Floor-sweeping device of the embodiment of the present invention carries out the detection and identification of carpet by treating purging zone, when detecting
When having carpet, then specially treated is carried out, for example avoid carpet region, to avoid the occurrence of the feelings of damage carpet or robot
Shape improves the intelligence degree of sweeping robot, and the user experience is improved.
The present invention proposes a kind of sweeping robot simultaneously comprising memory, processor and at least one be stored in are deposited
In reservoir and it is configured as the application program executed by processor, the application program is configurable for executing method of sweeping the floor.
The method of sweeping the floor includes the following steps:Detect whether region to be cleaned has carpet;When there is carpet in region to be cleaned, judgement is
No satisfaction avoids condition;When satisfaction avoids condition, carpet region is avoided.The method of sweeping the floor described in the present embodiment is
Method of sweeping the floor in the present invention involved by above-described embodiment, details are not described herein.
It will be understood by those skilled in the art that the present invention includes being related to for executing one in operation described herein
Or multinomial equipment.These equipment can specially be designed and be manufactured for required purpose, or can also include general-purpose computations
Known device in machine.There is these equipment the computer program being stored in it, these computer programs selectively to activate
Or reconstruct.Such computer program, which can be stored in equipment (for example, computer) readable medium or be stored in, to be suitable for
Storage e-command is simultaneously coupled in any kind of medium of bus respectively, and the computer-readable medium includes but not limited to
Any kind of disk (including floppy disk, hard disk, CD, CD-ROM and magneto-optic disk), ROM (Read-Only Memory, it is read-only to deposit
Reservoir), RAM (Random Access Memory, random access memory), EPROM (Erasable Programmable Read-
Only Memory, Erarable Programmable Read only Memory), EEPROM (Electrically Erasable Programmable
Read-Only Memory, Electrically Erasable Programmable Read-Only Memory), flash memory, magnetic card or light card.It is, readable
Medium includes by any medium of equipment (for example, computer) storage or transmission information in the form of it can read.
Those skilled in the art of the present technique be appreciated that can with computer program instructions come realize these structure charts and/or
The combination of each frame and these structure charts and/or the frame in block diagram and/or flow graph in block diagram and/or flow graph.This technology is led
Field technique personnel be appreciated that these computer program instructions can be supplied to all-purpose computer, special purpose computer or other
The processor of programmable data processing method is realized, to pass through the processing of computer or other programmable data processing methods
Device come execute structure chart and/or block diagram and/or flow graph disclosed by the invention frame or multiple frames in specify scheme.
Those skilled in the art of the present technique are appreciated that in the various operations crossed by discussion in the present invention, method, flow
Steps, measures, and schemes can be replaced, changed, combined or be deleted.Further, each with having been crossed by discussion in the present invention
Other steps, measures, and schemes in kind operation, method, flow may also be alternated, changed, rearranged, decomposed, combined or deleted.
Further, in the prior art to have and step, measure, the scheme in various operations, method, flow disclosed in the present invention
It may also be alternated, changed, rearranged, decomposed, combined or deleted.
The foregoing is merely the preferred embodiment of the present invention, are not intended to limit the scope of the invention, every utilization
Equivalent structure or equivalent flow shift made by description of the invention and accompanying drawing content is applied directly or indirectly in other correlations
Technical field, be included within the scope of the present invention.