CN108596084A - A kind of charging pile automatic identifying method and device - Google Patents

A kind of charging pile automatic identifying method and device Download PDF

Info

Publication number
CN108596084A
CN108596084A CN201810367073.2A CN201810367073A CN108596084A CN 108596084 A CN108596084 A CN 108596084A CN 201810367073 A CN201810367073 A CN 201810367073A CN 108596084 A CN108596084 A CN 108596084A
Authority
CN
China
Prior art keywords
profile
charging pile
alternative
scoring
module
Prior art date
Legal status (The legal status 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 status listed.)
Withdrawn
Application number
CN201810367073.2A
Other languages
Chinese (zh)
Inventor
马新军
杨洪杰
王建宽
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
NINGBO GQY VIDEO IT CO Ltd
Original Assignee
NINGBO GQY VIDEO IT CO Ltd
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 NINGBO GQY VIDEO IT CO Ltd filed Critical NINGBO GQY VIDEO IT CO Ltd
Priority to CN201810367073.2A priority Critical patent/CN108596084A/en
Publication of CN108596084A publication Critical patent/CN108596084A/en
Withdrawn legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/20Scenes; Scene-specific elements in augmented reality scenes
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/20Image preprocessing
    • G06V10/26Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
    • G06V10/267Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/40Extraction of image or video features
    • G06V10/44Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components

Abstract

The invention discloses a kind of charging pile automatic identifying methods, including:Surrounding enviroment are scanned, the profile information of each object in surrounding enviroment is obtained;According to the profile information of the charging pile to prestore, alternative profile is selected from the profile information of each object;According to the characteristic factor in the charging pile profile information to prestore, comprehensive score is carried out for each alternative profile;According to the sequence of comprehensive score from high to low, being moved to corresponding alternative profile advance row successively, further scoring confirms up to finding charging pile.In addition, the invention also discloses a kind of charging pile automatic identification equipments, including:Profile scan module, memory module, profile choose module, and score identification module and mobile confirmation module.Quickly and accurately automatic identification charging pile can be realized through the invention, to complete automatic charging;And the solution of the present invention reduces cost, accuracy of identification is high.

Description

A kind of charging pile automatic identifying method and device
Technical field
The present invention relates to automatic charging field more particularly to a kind of charging pile automatic identifying methods and device.
Background technology
Currently, the intelligent navigations equipment such as robot or sweeper mainly uses infrared distance measurement during automatic recharge Or the mode of laser radar identification charger 2D or 3D tag recognition positions charging pile.
Specifically, in the prior art, Fig. 1 is to identify tagged manner using laser radar, and effect is relatively preferable, but in reality In, needs to process label, increase extra cost.Also, most of product needs for beauty by label transparent material Material blocks, and when transparency reduces transparent material due to polluting or is other, can seriously affect recognition effect.Fig. 2 is infrared Sense telemetry identification charging pile principle (multiple infrared sensors may be used in actual use).Which needs are charging Infrared sensor R-T unit is installed, and often detection range is too short for infrared sensor in stake, limits application space.
Invention content
In order to overcome above-mentioned technological deficiency, a kind of charging pile automatic identifying method of present invention offer and device, particular technique Scheme is as follows:
On the one hand, the invention discloses a kind of charging pile automatic identifying methods, including:
S100 scans surrounding enviroment, obtains the profile information of each object in surrounding enviroment;
S200 selects alternative wheel according to the profile information of the charging pile to prestore from the profile information of each object It is wide;
S300 carries out comprehensive score according to the characteristic factor in the charging pile profile information to prestore, for each alternative profile;
Sequences of the S400 according to comprehensive score from high to low is moved to corresponding alternative profile advance row further scoring successively Confirm until finding the charging pile.
Preferably, the step S100 includes:
S110 acquires the data scanning point of surrounding enviroment by laser radar scanning surrounding enviroment;
The data scanning point of the surrounding enviroment is carried out region segmentation by S120;
S130 extracts characteristic information from each region after segmentation, obtains the profile letter of the corresponding object in the region Breath.
Preferably, the step S300 includes:
The shape feature for the charging pile profile information that S310 prestores according to matches each alternative profile, according to described each The matching degree of alternative profile is that each alternative profile carries out shape scoring;
S320 carries out position according to the position relationship of the charging pile and periphery object that prestore to each alternative profile Scoring;
S330 is obtained described each according to pre-set shape feature factor scores and the weights of position relationship factor scores The comprehensive score of alternative profile.
Preferably, the step S400 includes:
Sequences of the S410 according to comprehensive score from high in the end is moved to the front of the object where corresponding alternative profile;
The angle and distance of S420 adjustment scannings, again identifies that the shape of the object corresponding to the alternative profile;
S430 is according to the feature in the charging pile profile information to prestore;Again it is the object corresponding to the alternative profile Carry out shape scoring;
S440 judges whether the shape scoring is more than preset threshold value, if so, entering step S450;Otherwise, into step Rapid S460;
S450 judges that the object corresponding to the alternative profile is charging pile.
S460 judges that the object corresponding to the alternative profile is not the charging pile, further judges all alternative wheels Whether exterior feature has had confirmed that, if it is not, return to step S410 is moved to the object front where the alternative profile of next sequence;
Preferably, further include step after the step S460:S465, which works as, has judged all alternative profiles When having confirmed, starts the device that reports an error and report an error.
On the other hand, the invention also includes a kind of charging pile automatic identification equipments, including:Profile scan module, for sweeping Surrounding enviroment are retouched, the profile information of each object in surrounding enviroment is obtained;Memory module, for the charging pile that prestores profile information and Its characteristic factor;Profile chooses module, the profile information of the charging pile for prestoring according to the memory module, from the scanning Alternative profile is selected in the profile information for each object that module scans go out;Score identification module, for according to the storage mould Characteristic factor in the charging pile profile information that block prestores carries out synthesis for each alternative profile that the selection module is chosen and comments Point;Mobile confirmation module is moved to corresponding alternative successively for the sequence according to the scoring identification module comprehensive score Before profile, and further scoring is carried out by the scoring identification module and is confirmed until finding the charging pile.
Preferably, the profile scan module includes:Submodule is acquired, for passing through laser radar scanning surrounding enviroment, Acquire the data scanning point of surrounding enviroment;Region segmentation submodule, the surrounding enviroment for being used to obtain the acquisition submodule Data scanning point carries out region segmentation;Feature extraction submodule is obtained for extracting characteristic information from each region after segmentation Obtain the profile information of the corresponding object in the region.
Preferably, the scoring identification module includes:Shape scoring submodule, for what is prestored according to the memory module The feature of charging pile profile information matches each alternative profile, according to the matching degree of each alternative profile be each alternative profile into Row shape scores;Location score submodule, the position of the charging pile and periphery object for being prestored according to the memory module Relationship is set, location score is carried out to each alternative profile;Comprehensive score submodule, for according to pre-set shape feature The weights of factor scores and position relationship factor scores obtain the comprehensive score of each alternative profile.
Preferably, the mobile confirmation module includes:Mobile submodule, for being commented according to scoring identification module synthesis Divide sequence from high in the end, is moved to the front of the object where corresponding alternative profile;Submodule is adjusted, for adjusting scanning Angle and distance again identifies that the shape of the object corresponding to the alternative profile convenient for the profile scan module;Institute's commentary Point identification module further includes:The shape scoring submodule, for according to the feature in the charging pile profile information to prestore;Again For the carry out shape scoring of the object corresponding to the alternative profile;Identification submodule is judged, for judging the shape scoring Whether the shape scoring that submodule obtains is more than preset threshold value;It is less than or equal in preset threshold when judging that the shape scores When value, judge that the object corresponding to the alternative profile is charging pile;When judging that the shape scoring is more than preset threshold value, Judge that the object corresponding to the alternative profile is not the charging pile, further judges all alternative profiles whether really It has been recognized that, if all alternative profiles have not confirmed all, next sequence is moved to by the mobile confirmation module In front of object where alternative profile.
Preferably, the charging pile automatic identification equipment further includes:Report an error module, for identifying submodule when the judgement When judging that all alternative profiles have had confirmed that, the processing that reports an error is carried out.
The present invention at least has with the next item down advantageous effect:
(1) profile information of the invention by scanning periphery object, further according to the profile information of charging pile, to what is scanned The profile information of each object is screened, and carries out marking matching to the alternative profile chosen, to identify charging pile.It should Scheme simple possible, it is not necessary that special marking or installation special device are arranged on charging pile, according only to the profile information of charging pile The automatic identification that can realize charging pile reduces the cost of machine, and recognition efficiency is high.
(2) present invention employs laser radars to carry out scanning recognition surrounding enviroment, the 2-D data obtained according to laser radar Corresponding profile information is obtained to analyze, the treating capacity of data is reduced, reduces the difficulty of identification.And according to laser radar Characteristic, the profile information of the charging pile to prestore is general, and relatively simple (charging pile profile information is obtained again by laser radar scanning Take, shape feature is generally line segment), then profile information simple simon says, also reduces the matching marking of subsequent alternative profile Workload, reduce the requirement of the hardware and software of equipment, reduce cost.
(3) present invention to alternative profile when carrying out comprehensive score, first respectively to the shape feature factor of alternative profile It is given a mark with position feature factor, the comprehensive score of the alternative profile is then obtained according to the weight computing of each factor.It should Scheme is not the shape simply according to profile to evaluate identification charging pile, and is combined with positional factor to carry out with reference to commenting Point so that the height of comprehensive score more can concentrated expression go out the degree of approximation of alternative profile and charging pile, scoring is more objective, Accuracy also higher.
(4) present invention is true successively according to the sequence of comprehensive score after the comprehensive score for getting each alternative profile Recognize, accelerate the efficiency of identification, sequence it is more forward be more likely to be charging pile, and without to after if recognizing charging pile The alternative profile of continuous sequence carries out confirmation operation, therefore can greatly save the time for judging identification.And this programme is not Directly judge whether the alternative profile is charging pile to identify according to the comprehensive score obtained before, but needs to be moved to alternative The scanning scoring that profile comes to carry out again confirms, so that the accuracy higher of identification.
(5) when identification is less than charging pile, the present invention can also carry out the processing that reports an error, on the one hand, it is found in time convenient for user, And help to find real charging pile, the charging of machine is not influenced;On the other hand, a kind of pleasure of other style has also been let user experiencing Interest, machine are not omnipotent yet, and the machine that occasional is made a mistake also embodies its lovely place from another angle, also increases The sense of participation of user.
Description of the drawings
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment Attached drawing is briefly introduced, it should be apparent that, drawings in the following description are only some embodiments of the invention, for this For the those of ordinary skill in field, without having to pay creative labor, it can also be obtained according to these attached drawings His attached drawing.
Fig. 1 is that existing infrared sensing telemetry identifies charging pile principle schematic;
Fig. 2 is the existing schematic diagram positioned to charging pile by way of identifying charger label;
Fig. 3 is a kind of flow chart of charging pile automatic identifying method embodiment of the present invention;
Fig. 4 is a kind of flow chart of another embodiment of charging pile automatic identifying method of the present invention;
Fig. 5 is a kind of flow chart of another embodiment of charging pile automatic identifying method of the present invention;
Fig. 6 is a kind of flow chart of another embodiment of charging pile automatic identifying method of the present invention;
Fig. 7 a are surrounding enviroment schematic diagram in a kind of another embodiment of charging pile automatic identifying method of the present invention;
Fig. 7 b are the outline drawing that acquisition is scanned in a kind of another embodiment of charging pile automatic identifying method of the present invention;
Fig. 8 is a kind of block diagram of charging pile automatic identification equipment embodiment of the present invention;
Fig. 9 is a kind of block diagram of another embodiment of charging pile automatic identification equipment of the present invention.
Reference numeral:
10-- laser radars;11-- infrared sensors;20-- charging pile labels;The charge point of 30-- charging piles;01-- hinders Hinder object;02-- walls;03-- charging piles;04-- walls;05-- barriers;06-- barriers;The profile of 01a-- barriers; The profile of 02a-- walls;The profile of 03a-- charging piles;The profile of 04a- walls;The profile of 05a-- barriers;06a-- obstacles The profile of object;100-- profile scan modules;200-- memory modules;300-- profiles choose module;400-- scoring identification modules; 500-- moves confirmation module;600-- reports an error module;110-- acquires submodule;120 region segmentation submodules;130-- features carry Take submodule;410-- shapes scoring submodule;420-- location score submodules;430-- comprehensive score submodules;440-- sentences Disconnected identification submodule;510-- mobile submodules;520-- adjusts submodule.
Specific implementation mode
To make the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with attached drawing to the present invention make into It is described in detail to one step, it is clear that the described embodiments are only some of the embodiments of the present invention, rather than whole implementation Example.Based on the embodiments of the present invention, obtained by those of ordinary skill in the art without making creative efforts All other embodiment, shall fall within the protection scope of the present invention.
The invention discloses a kind of charging pile automatic identifying methods, and embodiment is as shown in figure 3, include:
S100 scans surrounding enviroment, obtains the profile information of each object in surrounding enviroment;
S200 selects alternative wheel according to the profile information of the charging pile to prestore from the profile information of each object It is wide;
S300 carries out comprehensive score according to the characteristic factor in the charging pile profile information to prestore, for each alternative profile;
Sequences of the S400 according to comprehensive score from high to low is moved to corresponding alternative profile advance row further scoring successively Confirm until finding the charging pile.
In the present embodiment, the profile information for storing charging pile is needed, so as to the profile information of the object obtained as scanning Comparison standard, and scan there are many means of surrounding enviroment, for example, can be scanned by laser radar, obtain scanning 2D profile informations on face;It can also be scanned by high-definition camera/binocular camera, obtain each object in surrounding enviroment 3D profile informations etc..Certainly, if the laser radar scanning used, being also required to for prestoring is that the laser radar is swept before Retouch the profile information of the charging pile of charging pile acquisition.It scans into each environment in periphery after the profile information of each object, further according to pre- The profile information deposited selects alternative profile from each contour of object that scanning obtains, that is to say, that weeds out and is obviously unlikely to be Then the profile of charging pile carries out comprehensive score to selecting the alternative profile come, certainly, the standard foundation of scoring still wants root Shape feature factor, position feature factor according to the characteristic factor in charging pile profile information, such as in its profile information etc.. Finally, we are moved to the further marking of corresponding alternatively profile advance row really successively according to the sequence of comprehensive score Recognize, until finding charging pile.That is be first moved in front of the corresponding object of alternative profile of comprehensive score first, then into The scanning scoring of one step, judges whether it is the charging pile to be found, if it is, can be carried out identification charging pile Automatic charging operation afterwards, if it is not, then needing the alternative profile for being moved to comprehensive score sequence second again corresponding In front of object, the scanning scoring for then equally carrying out front judges operation;If the alternative profile is not equally charging pile, Before being so moved to the corresponding object of alternative profile of next sequence again, until finding target charging pile.
Another embodiment of the method for the present invention, as shown in figure 4, on the basis of the above embodiments, specifically describing each object The obtaining step of body profile information, specifically, the method for the automatic identification charging pile of the program includes:
S110 acquires the data scanning point of surrounding enviroment by laser radar scanning surrounding enviroment;
The data scanning point of the surrounding enviroment is carried out region segmentation by S120;
S130 extracts characteristic information from each region after segmentation, obtains the profile letter of the corresponding object in the region Breath;
S200 selects alternative wheel according to the profile information of the charging pile to prestore from the profile information of each object It is wide;
S300 carries out comprehensive score according to the characteristic factor in the charging pile profile information to prestore, for each alternative profile;
Sequences of the S400 according to comprehensive score from high to low is moved to corresponding alternative profile advance row further scoring successively Confirm until finding the charging pile.
In the present embodiment, surrounding enviroment are scanned using laser radar, in general, we can be according to the installation of charging pile The height of laser radar is rationally arranged so that includes in the 2D profile informations of the laser radar scanning height in the height of position The profile information of charging pile.The detectable ambient enviroment of laser radar is general to swash relative to its range information and azimuth information The data that optical radar scanning obtains are 2-D datas, and for the 2-D data of a frame surrounding enviroment of scanning, laser is swept first Described point is divided into different blocks.Specifically, if the distance of continuous two scanning elements is less than a threshold value, the two scanning elements Belong to the same block.Scanning element in the same block tentatively can be determined that from the same object.If continuous two are swept As soon as the distance of described point is more than a threshold value, data frame is separated from this place.After region segmentation, we are further according to the every of segmentation 2-D data scanning element in a region carries out feature extraction, to obtain the profile information that the region corresponds to object.Laser thunder Up in the 2-D data of scanning, several important features:Tear point (breakPoint), angle point (Corner), straight line, circular arc Deng.The tearing point in data actually just has been found in region segmentation.After feature extraction, we can tentatively get The profile information of the corresponding object in region after each segmentation, for example have plenty of camber line, have plenty of broken line, have plenty of straight line etc.. Then, we can according to the profile information of charging pile, therefrom weed out be clearly not charging pile profile information, then will pick Remaining profile after removing alternately profile.Comprehensive marking and further confirmation identification are carried out to each alternative profile again, from And realize the automatic identification function of charging pile.
Another embodiment of the method for the present invention, based on any of the above embodiments, to the comprehensive score of alternative profile The step for propose a kind of specific marking mode so that the comprehensive score more can comprehensively reflect the profile pair The similarity of the object and charging pile answered.Specifically, as shown in figure 5, including:
S100 scans surrounding enviroment, obtains the profile information of each object in surrounding enviroment;
S200 selects alternative wheel according to the profile information of the charging pile to prestore from the profile information of each object It is wide;
The shape feature for the charging pile profile information that S310 prestores according to matches each alternative profile, according to described each The matching degree of alternative profile is that each alternative profile carries out shape scoring;
S320 carries out position according to the position relationship of the charging pile and periphery object that prestore to each alternative profile Scoring;
S330 is obtained described each according to pre-set shape feature factor scores and the weights of position relationship factor scores The comprehensive score of alternative profile;
Sequences of the S400 according to comprehensive score from high to low is moved to corresponding alternative profile advance row further scoring successively Confirm until finding the charging pile.
In the present embodiment, the shape feature of charging pile profile information, to match each alternative profile, according to each alternative profile with The shape matching degree of the profile of the charging pile is given a mark.For example, charging pile is generally a cuboid or square, correspond to The shape feature of profile information be a line segment, line segment length range:More than m centimetres, it is less than n centimetres.So we It is given a mark to each alternative profile according to this feature, line segment is contained in alternative profile, and calculated line segment length meets charging The marking of line segment length range in device profile information is also higher, and the straightness of the line segment for being fitted in alternative profile compared with Difference or calculate the line segment length range that line segment length is not inconsistent charger alternative profile, marking will be low.Specifically, can A set of shape feature standards of grading are arranged, each alternative profile is given to carry out shape scoring accordingly.The possibility scanned in view of us 2D profile informations, then only carry out shape scoring if, it may not be possible to according to shape scoring quickly find out charging pile come, because This, we can also give a mark according to other characteristic factors, to obtain the comprehensive score of each alternative profile, specifically, We can according to the position feature factor of charging pile, such as charging pile we generally wall is leaned on to be arranged, then can be according to this Characteristic factor gives a mark to the positional factor of each alternative profile.If it is by laser radar scanning surrounding enviroment, then When the 2D profile informations of the periphery object of laser radar scanning height, the 2D profile informations of scanning to wall are just also line segment. Therefore, we can with the presence or absence of wall line segment profile, come preliminary judgement, whether the alternative profile may according to around alternative profile By wall, further given a mark to each alternative profile by this positional factor.Certainly, if charging pile is not placed against the wall It sets, and is proximate to other objects, or be provided separately within certain area, we also can be according to the position of this charging pile and periphery object Relationship etc. carries out the scoring of positional factor.After the shape scoring for getting each alternative profile and location score, due to each spy Sign factor is different to the significance level of identification, and therefore, we can be that each characteristic factor be arranged different weights, such as by shape The weights of scoring are set as 70%;The weights of location score are set as 30%;To obtain the comprehensive score of each alternative profile:Shape Shape scoring * 70%+ location scores * 30%.Certainly, if shape and position are of equal importance to the identification of charging pile, 0.5 weights can also be respectively set.Preferably, other than shape feature factor and position feature factor, we can be with root It gives a mark according to the other feature factor in the profile information of charging pile, then assigns each characteristic factor weights respectively, and then obtain Obtain the comprehensive score of each alternative profile.
According to the shape of charging pile in the present embodiment, it is line that can get shape feature in the corresponding profile information of charging pile Section only need to carry out marking matching according to the line segment and line segment length to alternative profile, be different from complicated 3D rendering and identify skill Art, program calculation amount smaller, the requirement to hardware is relatively low, realizes more simple, efficient.
Next it is according to comprehensive score after the comprehensive score for getting each alternative profile in any of the above-described embodiment Sequence scored be confirmed again successively, specifically, another embodiment of the method for the present invention, as shown in fig. 6, packet It includes:
S100 scans surrounding enviroment, obtains the profile information of each object in surrounding enviroment;
S200 selects alternative wheel according to the profile information of the charging pile to prestore from the profile information of each object It is wide;
S300 carries out comprehensive score according to the characteristic factor in the charging pile profile information to prestore, for each alternative profile;
Sequences of the S410 according to comprehensive score from high in the end is moved to the front of the object where corresponding alternative profile;
The angle and distance of S420 adjustment scannings, again identifies that the shape of the object corresponding to the alternative profile;
S430 is according to the feature in the charging pile profile information to prestore;Again it is the object corresponding to the alternative profile Carry out shape scoring;
S440 judges whether the shape scoring is more than preset threshold value, if so, entering step S450;Otherwise, into step Rapid S460;
S450 judges that the object corresponding to the alternative profile is charging pile;
S460 judges that the object corresponding to the alternative profile is not the charging pile, further judges all alternative wheels Whether exterior feature has had confirmed that, if it is not, return to step S410 is moved to the object front where the alternative profile of next sequence;
S465 starts the device that reports an error and reports an error when judging that all alternative profiles have had confirmed that.
In above-described embodiment, after the comprehensive score for getting each alternative profile, we can be according to the height of comprehensive score Confirmation is identified in sequence successively.Specifically, we can be first suitable according to from high to low by the comprehensive score of each alternative profile Sequence is ranked up, and then, recognition and verification is proceeded by from the alternative profile of sequence first.That is first it is moved to sequence first Alternative profile where object in front of, then adjust the angle and distance of scanning, so as to can the identification of more " clear " this is alternative The corresponding object of profile.Specifically, scanning means we by taking laser radar as an example, then being moved to before and after alternative profile, then adjust Then the scanning angle and distance of whole good laser radar scan the corresponding object of the alternative profile, obtain relatively sharp standard again True alternative profile, then further according to the shape feature of the profile information of the charging pile to prestore, to the alternative wheel of the reacquisition Exterior feature carries out shape scoring, then judges whether the score value of shape scoring is more than preset threshold value (threshold value to score), if so, It then can be determined that the corresponding object of the alternative profile is charging pile, operated then subsequent automatic charging can also be carried out.And If the score value of the shape scoring after scoring again is less than or equal to preset threshold value, the alternative profile is decided that not It is charging pile, then in front of the corresponding object of the alternative profile that next sequence according to the sequence of comprehensive score, will be moved to, with The previous alternative profile confirmation operation that equally carries out similarly scoring (is equally adjusted after the operations such as scanning and carries out shape again and comment Point, further according to score value and the comparison of threshold value of shape scoring come whether judge the corresponding object of the alternative profile be charging pile).Such as Identified have confirmed that not of all alternative profiles of fruit is charging pile, that is to say, that was still identified after matching all alternative profiles In the case of less than charging pile, then system will start the device that reports an error, inform that user's machine can not find charging pile, to allow User is manipulated accordingly, is moved it to before real charging pile.
Confirm successively according to the sequence of scoring, accelerate the efficiency of identification, more forward being more likely to be of sorting is filled Electric stake, and without carrying out confirmation operation to the alternative profile subsequently to sort if recognizing charging pile, therefore can be significantly Save the time for judging identification.And this programme is not that the comprehensive score obtained before directly judges the alternative wheel to identify Whether exterior feature is charging pile, but needs to be moved to the scanning scoring that alternative profile comes to carry out again and confirm, so that knowing Other accuracy higher.
When identification is less than charging pile, system can also carry out the processing that reports an error, on the one hand, find, and help in time convenient for user It helps and finds real charging pile, do not influence the charging of machine;On the other hand, a kind of enjoyment of other style has also been let user experiencing, Machine is not omnipotent yet, and the machine that occasional is made a mistake also embodies its lovely place from another angle, also increases use The sense of participation at family.
Preferably, after the device that reports an error is reported an error, before being moved to charging pile according to the manipulation of user, and reopens and sweep It retouches acquisition and learns the profile information of the charging pile;It, can be to the profile of the charging pile to prestore after a certain number of study Information is updated.
Another embodiment of the method for the present invention proposes a kind of side relying solely on charging pile outer profile identification charger position Method, basic principle are as follows:
It is assumed herein that charger body does not have any special shape feature, only a cuboid.Robot only can Identify the 2D profile informations (if the identification sensor of robot is laser radar) or charger one of the certain altitude of charger Determine the 3D outer profiles information (if the identification sensor of robot is binocular camera or Kinect) of altitude range.Both feelings Condition, outline identification principle is similar, therefore, mainly introduces 2D profile information recognition methods.As shown in Figure 7a, it is assumed that robot works In the environment of a variety of chaff interferents, robot may identify many places, similar charger 2D profile informations.Due to robot Charger profile information is pre-deposited, therefore can be filled by charger shape (being in this case a line segment) and identification candidate Electric appliance 2D contour signals are matched one by one, and matching result is given a mark.It in practical applications, can also be according to charger By wall situation, the information such as the distance apart from ideal position are further given a mark, the highest 2D profiles of final choice synthesis marking Signal is moved to robot immediately ahead of charger as true charger, is further confirmed that, if it is confirmed that success, i.e., Docking operation can be started to charge up.The embodiment specific implementation process is following (herein still by taking 2D profile information schemes as an example):
1) contour signal matches
As shown in Figure 7b, due to 2D chargers signal it is known that therefore can be believed with alternative profile by shape matching algorithm Number 01a--06a is matched, and is given a mark respectively to matching result.
2) by matching result, the larger candidate contours of difference are rejected
As shown in Figure 7b, it is evident that profile 02a, 04a can be rejected (length is too short);Profile 06a (straight lines can also be rejected Degree is too poor).
3) by whether leaning on wall, situations such as distance apart from ideal position, secondary marking is carried out.
By wall situation:03a>05a>01a does not have other objects around 01 corresponding profile 01a of apparent barrier herein, It can not possibly be placed against the wall and set, profile 03a or so has line segment, highest of giving a mark.
Apart from the marking of ideal position distance (if being ideal position at 03a):03a>05a>01a
Marking factor can also be added according to other factors.
4) comprehensive marking is calculated, highest marking is selected.
Each factor weights are set, and carry out comprehensive marking, profile 03a marking highests think that 03a is charging device wheel at this time It is wide.
5) it immediately ahead of adjustment robot to profile 03a, verifies again.
Purpose in front of robot to profile 03a mainly adjusts laser radar angle and distance, can more " clear " Ground contour identification 03a carries out shape matching again, if marking is higher than certain threshold value, illustrates that selection is correct, otherwise adjusts machine People position is to 04a) in marking second profile in front of, 5 are repeated the above steps, until finding charger, or the comprehensive marking of selection Less than the candidate contours of certain threshold value, report an error.
Based on the same technical idea, the present invention also provides a kind of charging pile automatic identification equipment, which can be used The charging pile automatic identifying method of the present invention, specifically, the charging pile automatic identification equipment is as shown in figure 8, include:Profile scan Module 100 obtains the profile information of each object in surrounding enviroment for scanning surrounding enviroment;Memory module 200, for prestoring The profile information and its characteristic factor of charging pile;Profile chooses module 300, is filled for what is prestored according to the memory module 200 The profile information of electric stake selects alternative profile in the profile information of each object scanned from the scan module;Scoring is known Other module 400, the characteristic factor in charging pile profile information for being prestored according to the memory module 200 are the selection Each alternative profile that module is chosen carries out comprehensive score;Mobile confirmation module 500, for according to the scoring identification module 400 The sequence of comprehensive score, before being moved to corresponding alternative profile successively, and by the scoring identification module 400 into traveling one Step scoring confirms until finding the charging pile.
The present apparatus by profile scan module 100 scan periphery object profile information, profile choose module 300 further according to The profile information for the charging pile that memory module 200 prestores screens the profile information of each object scanned, obtains alternative Profile;Scoring identification module 400 chooses the alternative profile that module 300 is chosen to profile and carries out marking matching, to identify Charging pile.Program simple possible, it is not necessary that special marking or installation special device are arranged on charging pile, according only to charging pile Profile information can realize the automatic identification of charging pile, reduce the cost of machine, and recognition efficiency is high
Another embodiment of apparatus of the present invention, on the basis of above-mentioned apparatus embodiment, as shown in figure 8, the profile is swept Retouching module 100 includes:Submodule 110 is acquired, for by laser radar scanning surrounding enviroment, the data for acquiring surrounding enviroment to be swept Described point;The data scanning point of region segmentation submodule 120, the surrounding enviroment for obtaining the acquisition submodule 110 carries out Region segmentation;Feature extraction submodule 130 obtains the region for extracting characteristic information from each region after segmentation The profile information of corresponding object.
On the basis of any of the above-described device, as shown in figure 9, the scoring identification module 400 includes:Shape scoring submodule Block 410, the feature of the charging pile profile information for being prestored according to the memory module 200 match each alternative profile, according to institute The matching degree for stating each alternative profile is that each alternative profile carries out shape scoring;Location score submodule 420, for being deposited according to The position relationship for the charging pile and periphery object that storage module 200 prestores carries out location score to each alternative profile;It is comprehensive Scoring submodule 430 is closed to obtain for the weights according to pre-set shape feature factor scores and position relationship factor scores Take the comprehensive score of each alternative profile.
On the basis of any of the above-described device, the mobile confirmation module 500 includes:Mobile submodule 510, for according to The sequence of 400 comprehensive score of the scoring identification module from high in the end, is moved to the front of the object where corresponding alternative profile; Submodule 520 is adjusted, the angle and distance for adjusting scanning again identifies that described standby convenient for the profile scan module 100 Select the shape of the object corresponding to profile;The scoring identification module 400 further includes:The shape scoring submodule 410, is used for According to the feature in the charging pile profile information to prestore;Again the carry out shape for the object corresponding to the alternative profile is commented Point;Identification submodule 440 is judged, for judging it is default whether the shape scoring of the shape scoring acquisition of submodule 410 is more than Threshold value;When judging that the shape scoring is less than or equal in preset threshold value, the object corresponding to the alternative profile is judged Body is charging pile;When judging that the shape scoring is more than preset threshold value, the object corresponding to the judgement alternative profile is not It is the charging pile, further judges whether all alternative profiles have had confirmed that, if all alternative profiles is not complete Portion has confirmed, then the object front where the alternative profile of next sequence is moved to by the mobile confirmation module 500.
Preferably, the charging pile automatic identification equipment further includes:The module that reports an error 600, for when judgement identification When module 440 judges that all alternative profiles have had confirmed that, the processing that reports an error is carried out.
When identification is less than charging pile, the present apparatus can also carry out the processing that reports an error, consequently facilitating user has found in time, and help Real charging pile is found, does not influence the charging of machine.
The device of the invention embodiment is corresponding with the inventive embodiments of the present invention, and the technology of embodiment of the method for the invention is thin Section is equally applicable to the device of the invention embodiment, repeats, repeats no more to reduce.
Although preferred embodiments of the present invention have been described, it is created once a person skilled in the art knows basic Property concept, then additional changes and modifications may be made to these embodiments.So it includes excellent that the following claims are intended to be interpreted as It selects embodiment and falls into all change and modification of the scope of the invention.
Obviously, various changes and modifications can be made to the invention without departing from essence of the invention by those skilled in the art God and range.In this way, if these modifications and changes of the present invention belongs to the range of the claims in the present invention and its equivalent technologies Within, then the present invention is also intended to include these modifications and variations.

Claims (10)

1. a kind of charging pile automatic identifying method, which is characterized in that including:
S100 scans surrounding enviroment, obtains the profile information of each object in surrounding enviroment;
S200 selects alternative profile according to the profile information of the charging pile to prestore from the profile information of each object;
S300 carries out comprehensive score according to the characteristic factor in the charging pile profile information to prestore, for each alternative profile;
Sequences of the S400 according to comprehensive score from high to low, being moved to corresponding alternative profile advance row successively, further scoring confirms Until finding the charging pile.
2. a kind of charging pile automatic identifying method according to claim 1, which is characterized in that the step S100 includes:
S110 acquires the data scanning point of surrounding enviroment by laser radar scanning surrounding enviroment;
The data scanning point of the surrounding enviroment is carried out region segmentation by S120;
S130 extracts characteristic information from each region after segmentation, obtains the profile information of the corresponding object in the region.
3. a kind of charging pile automatic identifying method according to claim 1, which is characterized in that the step S300 includes:
The shape feature for the charging pile profile information that S310 prestores according to matches each alternative profile, according to described each alternative The matching degree of profile is that each alternative profile carries out shape scoring;
S320 carries out position to each alternative profile and comments according to the position relationship of the charging pile and periphery object that prestore Point;
S330 is obtained described each alternative according to pre-set shape feature factor scores and the weights of position relationship factor scores The comprehensive score of profile.
4. according to a kind of charging pile automatic identifying method of claim 1-3 any one of them, which is characterized in that the step S400 includes:
Sequences of the S410 according to comprehensive score from high in the end is moved to the front of the object where corresponding alternative profile;
The angle and distance of S420 adjustment scannings, again identifies that the shape of the object corresponding to the alternative profile;
S430 is according to the feature in the charging pile profile information to prestore;Again it is the progress of the object corresponding to the alternative profile Shape scores;
S440 judges whether the shape scoring is more than preset threshold value, if so, entering step S450;Otherwise, it enters step S460;
S450 judges that the object corresponding to the alternative profile is the charging pile;
S460 judges that the object corresponding to the alternative profile is not the charging pile, further judges that all alternative profiles are It is no to have had confirmed that, if it is not, return to step S410 is moved to the object front where the alternative profile of next sequence.
5. a kind of charging pile automatic identifying method according to claim 4, which is characterized in that after the step S460 It further include step:
S465 starts the device that reports an error and reports an error when judging that all alternative profiles have had confirmed that.
6. a kind of charging pile automatic identification equipment, which is characterized in that including:
Profile scan module obtains the profile information of each object in surrounding enviroment for scanning surrounding enviroment;
Memory module, profile information and its characteristic factor for the charging pile that prestores;
Profile chooses module, and the profile information of the charging pile for being prestored according to the memory module is swept from the scan module Alternative profile is selected in the profile information of each object described;
Score identification module, and the characteristic factor in charging pile profile information for being prestored according to the memory module, is described It chooses each alternative profile that module is chosen and carries out comprehensive score;
It is standby to be moved to correspondence for the sequence according to the scoring identification module comprehensive score successively for mobile confirmation module Before selecting profile, and further scoring is carried out by the scoring identification module and is confirmed until finding the charging pile.
7. a kind of charging pile automatic identification equipment according to claim 6, which is characterized in that the profile scan module packet It includes:
Submodule is acquired, for by laser radar scanning surrounding enviroment, acquiring the data scanning point of surrounding enviroment;
The data scanning point of region segmentation submodule, the surrounding enviroment for obtaining the acquisition submodule carries out region minute It cuts;
Feature extraction submodule obtains the corresponding object in the region for extracting characteristic information from each region after segmentation The profile information of body.
8. a kind of charging pile automatic identification equipment according to claim 6, which is characterized in that the scoring identification module packet It includes:
Shape scoring submodule, the feature of the charging pile profile information for being prestored according to the memory module match each alternative Profile is that each alternative profile carries out shape scoring according to the matching degree of each alternative profile;
Location score submodule, the position relationship of the charging pile and periphery object for being prestored according to the memory module, Location score is carried out to each alternative profile;
Comprehensive score submodule, for the power according to pre-set shape feature factor scores and position relationship factor scores Value obtains the comprehensive score of each alternative profile.
9. according to a kind of charging pile automatic identification equipment of claim 6-8 any one of them, which is characterized in that the movement is true Recognizing module includes:
Mobile submodule is moved to corresponding alternative for the sequence according to the scoring identification module comprehensive score from high in the end The front of object where profile;
Submodule is adjusted, the angle and distance for adjusting scanning again identifies that described alternative convenient for the profile scan module The shape of object corresponding to profile;
The scoring identification module further includes:
The shape scoring submodule, for according to the feature in the charging pile profile information to prestore;Again it is the alternative wheel The carry out shape scoring of object corresponding to exterior feature;
Identification submodule is judged, for judging whether the shape scoring that the shape scoring submodule obtains is more than preset threshold Value;When judging that the shape scoring is less than or equal in preset threshold value, judge that the object corresponding to the alternative profile is The charging pile;When judging that the shape scoring is more than preset threshold value, the object corresponding to the judgement alternative profile is not It is the charging pile, further judges whether all alternative profiles have had confirmed that, if all alternative profiles is not complete Portion has confirmed, then the object front where the alternative profile of next sequence is moved to by the mobile confirmation module.
10. a kind of charging pile automatic identification equipment according to claim 9, which is characterized in that further include:
Report an error module, for when judgement identification submodule judges that all alternative profiles have had confirmed that, carrying out Report an error processing.
CN201810367073.2A 2018-04-23 2018-04-23 A kind of charging pile automatic identifying method and device Withdrawn CN108596084A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810367073.2A CN108596084A (en) 2018-04-23 2018-04-23 A kind of charging pile automatic identifying method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810367073.2A CN108596084A (en) 2018-04-23 2018-04-23 A kind of charging pile automatic identifying method and device

Publications (1)

Publication Number Publication Date
CN108596084A true CN108596084A (en) 2018-09-28

Family

ID=63613992

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810367073.2A Withdrawn CN108596084A (en) 2018-04-23 2018-04-23 A kind of charging pile automatic identifying method and device

Country Status (1)

Country Link
CN (1) CN108596084A (en)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109394220A (en) * 2018-11-28 2019-03-01 广东永衡良品科技有限公司 A kind of intelligence body height measuring device and its measurement method
CN109782770A (en) * 2019-02-02 2019-05-21 南京航空航天大学 A kind of method of grass trimmer recharging
CN110109450A (en) * 2019-04-01 2019-08-09 深圳市女娲机器人科技有限公司 A kind of automatic recharging method of robot, device and terminal device
CN110378285A (en) * 2019-07-18 2019-10-25 北京小狗智能机器人技术有限公司 A kind of recognition methods of cradle, device, robot and storage medium
CN111027540A (en) * 2019-11-08 2020-04-17 深兰科技(上海)有限公司 Method and equipment for searching target object
CN111986221A (en) * 2020-09-07 2020-11-24 北京凌云光技术集团有限责任公司 Edge evaluation method and device based on gray scale or position information
CN115933706A (en) * 2023-02-07 2023-04-07 科大讯飞股份有限公司 Robot charging method and device, robot and robot system
WO2023236642A1 (en) * 2022-06-07 2023-12-14 速感科技(北京)有限公司 Autonomous mobile device, autonomous mobile device operation method and system, and storage medium

Citations (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1853552A (en) * 2005-04-20 2006-11-01 Lg电子株式会社 Cleaning robot having auto-return function to charching-stand and method using the same
CN102770864A (en) * 2010-01-29 2012-11-07 香港科技大学 Architectural pattern detection and modeling in images
CN103997082A (en) * 2014-04-21 2014-08-20 伍小军 Autonomous charging device and autonomous charging method of mobile robot
CN105259918A (en) * 2015-09-18 2016-01-20 莱克电气股份有限公司 Automatic charging returning method of robot dust collector
CN105354866A (en) * 2015-10-21 2016-02-24 郑州航空工业管理学院 Polygon contour similarity detection method
US20160144735A1 (en) * 2010-04-19 2016-05-26 Joseph C. Haddad Automated electric vehicle charging system and method
CN106200640A (en) * 2016-07-24 2016-12-07 广东大仓机器人科技有限公司 A kind of automatic charging system based on image recognition location technology and charging method thereof
CN106235954A (en) * 2016-09-22 2016-12-21 蔡展 A kind of automatic cleaning machine people and charging system thereof
CN106556428A (en) * 2016-11-30 2017-04-05 河海大学常州校区 A kind of robot power station cruising inspection system and method
CN106826821A (en) * 2017-01-16 2017-06-13 深圳前海勇艺达机器人有限公司 The method and system that robot auto-returned based on image vision guiding charges
CN107392962A (en) * 2017-08-14 2017-11-24 深圳市思维树科技有限公司 A kind of robot charging docking system and method based on pattern identification
CN107481238A (en) * 2017-09-20 2017-12-15 众安信息技术服务有限公司 Image quality measure method and device
CN107539160A (en) * 2017-09-29 2018-01-05 深圳悉罗机器人有限公司 Charging pile and its recognition methods, intelligent mobile robot
CN107590836A (en) * 2017-09-14 2018-01-16 斯坦德机器人(深圳)有限公司 A kind of charging pile Dynamic Recognition based on Kinect and localization method and system

Patent Citations (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1853552A (en) * 2005-04-20 2006-11-01 Lg电子株式会社 Cleaning robot having auto-return function to charching-stand and method using the same
CN102770864A (en) * 2010-01-29 2012-11-07 香港科技大学 Architectural pattern detection and modeling in images
US20160144735A1 (en) * 2010-04-19 2016-05-26 Joseph C. Haddad Automated electric vehicle charging system and method
CN103997082A (en) * 2014-04-21 2014-08-20 伍小军 Autonomous charging device and autonomous charging method of mobile robot
CN105259918A (en) * 2015-09-18 2016-01-20 莱克电气股份有限公司 Automatic charging returning method of robot dust collector
CN105354866A (en) * 2015-10-21 2016-02-24 郑州航空工业管理学院 Polygon contour similarity detection method
CN106200640A (en) * 2016-07-24 2016-12-07 广东大仓机器人科技有限公司 A kind of automatic charging system based on image recognition location technology and charging method thereof
CN106235954A (en) * 2016-09-22 2016-12-21 蔡展 A kind of automatic cleaning machine people and charging system thereof
CN106556428A (en) * 2016-11-30 2017-04-05 河海大学常州校区 A kind of robot power station cruising inspection system and method
CN106826821A (en) * 2017-01-16 2017-06-13 深圳前海勇艺达机器人有限公司 The method and system that robot auto-returned based on image vision guiding charges
CN107392962A (en) * 2017-08-14 2017-11-24 深圳市思维树科技有限公司 A kind of robot charging docking system and method based on pattern identification
CN107590836A (en) * 2017-09-14 2018-01-16 斯坦德机器人(深圳)有限公司 A kind of charging pile Dynamic Recognition based on Kinect and localization method and system
CN107481238A (en) * 2017-09-20 2017-12-15 众安信息技术服务有限公司 Image quality measure method and device
CN107539160A (en) * 2017-09-29 2018-01-05 深圳悉罗机器人有限公司 Charging pile and its recognition methods, intelligent mobile robot

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
刘梅 等: ""基于三维模型的定位算法研究"", 《计算机时代》 *
窦育民: ""基于轮廓的目标检测研究"", 《中国博士学位论文全文数据库 信息科技辑》 *

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109394220A (en) * 2018-11-28 2019-03-01 广东永衡良品科技有限公司 A kind of intelligence body height measuring device and its measurement method
CN109782770A (en) * 2019-02-02 2019-05-21 南京航空航天大学 A kind of method of grass trimmer recharging
CN109782770B (en) * 2019-02-02 2022-06-21 南京航空航天大学 Method for automatically charging mower
CN110109450A (en) * 2019-04-01 2019-08-09 深圳市女娲机器人科技有限公司 A kind of automatic recharging method of robot, device and terminal device
CN110378285A (en) * 2019-07-18 2019-10-25 北京小狗智能机器人技术有限公司 A kind of recognition methods of cradle, device, robot and storage medium
CN111027540A (en) * 2019-11-08 2020-04-17 深兰科技(上海)有限公司 Method and equipment for searching target object
CN111986221A (en) * 2020-09-07 2020-11-24 北京凌云光技术集团有限责任公司 Edge evaluation method and device based on gray scale or position information
WO2023236642A1 (en) * 2022-06-07 2023-12-14 速感科技(北京)有限公司 Autonomous mobile device, autonomous mobile device operation method and system, and storage medium
CN115933706A (en) * 2023-02-07 2023-04-07 科大讯飞股份有限公司 Robot charging method and device, robot and robot system

Similar Documents

Publication Publication Date Title
CN108596084A (en) A kind of charging pile automatic identifying method and device
CN109460709B (en) RTG visual barrier detection method based on RGB and D information fusion
CN110175576B (en) Driving vehicle visual detection method combining laser point cloud data
US11360571B2 (en) Information processing device and method, program and recording medium for identifying a gesture of a person from captured image data
CN111486855B (en) Indoor two-dimensional semantic grid map construction method with object navigation points
Baillard et al. Automatic reconstruction of piecewise planar models from multiple views
JP3660492B2 (en) Object detection device
CN110344621A (en) A kind of wheel points cloud detection method of optic towards intelligent garage
CN110674705B (en) Small-sized obstacle detection method and device based on multi-line laser radar
US7321669B2 (en) Method and apparatus for refining target position and size estimates using image and depth data
CN111123262B (en) Automatic driving 3D modeling method, device and system
CN110110678A (en) Determination method and apparatus, storage medium and the electronic device of road boundary
CN115083199A (en) Parking space information determination method and related equipment thereof
CN116895059A (en) BEV space target detection method and device for multi-view perspective image
JP2861014B2 (en) Object recognition device
CN114425774B (en) Robot walking road recognition method, robot walking road recognition device, and storage medium
Cootes et al. Medical image interpretation using active shape models: Recent advances
JPH08255245A (en) Landmark recognizing device and landmark recognizing method
CN113326715B (en) Target association method and device
CN201181487Y (en) Infrared opto-electric passenger traffic collection apparatus based on radial base neural network
Bartel et al. Real-time outdoor trail detection on a mobile robot
EP0307120A2 (en) Object recognition
CN112308904A (en) Vision-based drawing construction method and device and vehicle-mounted terminal
Asensio et al. Navigation Among Obstacles by the Cooperation of Trinocular Stereo Vision System and Laser Rangefinder
CN115661323B (en) Method for establishing three-dimensional virtual image in real time by using 3D underwater sonar system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
WW01 Invention patent application withdrawn after publication

Application publication date: 20180928

WW01 Invention patent application withdrawn after publication