A kind of recognition methods of cradle, device, robot and storage mediumTechnical field
This application involves intelligent robot technology field more particularly to a kind of recognition methods of cradle, device, robotAnd storage medium.
Background technique
With the development of science and technology, intelligent robot has gradually been deep into the every field of people's daily life, gives peopleLife and work bring many conveniences, improve the working efficiency and quality of life of people.
The recharging of intelligent robot refer to when the electricity of intelligent robot is lower than default power threshold, contexture by self roadLine finds cradle and charges.Currently, the recognition methods of cradle is based primarily upon list in intelligent robot during rechargingPure laser radar or simple infrared signal.Its process be when intelligent robot electricity lower than some threshold value need to charge when,By laser radar or infrared sensor, the general pose of cradle is searched for.Then further according to the laser radar signal detectedOr infrared signal, it further determines that the position of cradle, eventually moves to charge on cradle.
But for using the scheme of laser radar merely, because the identification distance of laser radar is very short, usually effectivelyIdentification distance only has 1m or so, and it is too many to will lead to the region for needing to search for, so that recharging efficiency reduction.And for using infrared biographyThe scheme of sensor, because infrared signal, there are very strong reflecting effect, intelligent robot can not be determined according to infrared signal to be filledThe exact position of electric seat may be directed to robot under complex environment the region of search of mistake, increase searchComplexity reduces and recharges efficiency.
Summary of the invention
In order to solve above-mentioned technical problem of the existing technology, this application provides a kind of recognition methods of cradle,Device, robot and storage medium are able to ascend accuracy and reliability when identification cradle, and then improve robotRecharge efficiency.
This application provides a kind of recognition methods of cradle, be applied to robot, the robot include camera andLaser radar, which comprises
The identification model of cradle is obtained in advance；The identification model from the shooting image of the camera for identifyingThe cradle；
When being determined according to the identification model in the shooting image of the camera there are when doubtful cradle target, pass throughThe laser radar obtains the position coordinates of the doubtful cradle target, obtains the doubtful charging according to the identification modelSeat target is the probability of cradle, and the position coordinates and the probability are recorded in information table；
Into when recharging mode, according to the corresponding coordinate position of probability described in the sequential search of the probability from high to low；
When the laser radar obtain flag information matched with the flag information on preset charged seat and camera identificationThere are when cradle, there are cradles for confirmation current location.
Optionally, before entrance recharges mode, the method also includes: what is recorded in information table described in real-time update is describedPosition coordinates and probability.
Optionally, the position coordinates and probability recorded in information table described in the real-time update include:
According to the position coordinates captured in real-time image and the identification model, determine existing for the position coordinatesThe doubtful cradle target is the real-time probability of cradle, and the real-time probability is updated in the information table；
When the real-time probability be lower than predetermined probabilities when, deleted from the information table the real-time probability and it is described in real timeThe corresponding position coordinates of probability.
Optionally, described to include: according to the corresponding coordinate position of probability described in probability sequential search from high to low
The position coordinates that distance is less than pre-determined distance threshold value are divided into same class position coordinates, in every class position coordinatesThe sum of corresponding probability of all position coordinates is the probability of such position coordinates；
According to the coordinate position of the probability every class of sequential search from high to low of every class position coordinates.
Optionally, the preparatory identification model for obtaining cradle includes:
It advances with the camera and obtains shooting image of the cradle under different angle and different distance, utilizeDeep learning algorithm Faster R-CNN is trained to obtain the identification model of the cradle shooting image.
Optionally, the cradle is not found yet after having searched for all position coordinates, the method also includes:
The path for planning a traversal working environment utilizes the camera and institute during the path is mobileState laser radar search cradle.
Present invention also provides a kind of identification device of cradle, described device includes: the first acquisition module, the second acquisitionModule, search module and confirmation module；
Described first obtains module, for obtaining the identification model of cradle in advance；The identification model is used for from camera shootingThe cradle is identified in the shooting image of head；
Described second obtains module, for existing when in the shooting image for determining the camera according to the identification modelWhen doubtful cradle target, the position coordinates of the doubtful cradle target are obtained by laser radar, according to the identification mouldType obtains the doubtful cradle target and is the probability of cradle, and records in information table position coordinates and described generalRate；
Described search module, when recharging mode for entering, according to general described in the sequential search of the probability from high to lowThe corresponding coordinate position of rate；
The confirmation module, for when the flag information on the flag information and preset charged seat that the laser radar obtainsMatching and camera identify that there are cradles for confirmation current location there are when cradle.
Optionally, described device further include: update module；
The update module, the position coordinates and probability for being recorded in information table described in real-time update.
Optionally, described search module includes: classification submodule and search submodule；
The classification submodule, the position coordinates for distance to be less than pre-determined distance threshold value are divided into same class position and sitIt marks, the sum of corresponding probability of all position coordinates in every class position coordinates is the probability of such position coordinates；
Described search submodule, the coordinate for the every class of sequential search of the probability according to every class position coordinates from high to lowPosition.
Present invention also provides a kind of robot, the robot includes camera, laser radar and any of the above-described instituteThe identification device for the cradle stated.
Present invention also provides a kind of storage medium, the storage medium include storage program, described program execute onState the recognition methods of cradle described in any one.
Herein described method has the advantage that
Method provided by the present application obtains the identification model of cradle in advance, which is used for the shooting from cameraThe cradle is identified in image；When there are doubtful cradle targets in the shooting image for determining camera according to identification modelWhen, the position coordinates of the doubtful cradle target are obtained by laser radar, and doubtful cradle mesh is obtained according to identification modelIt is designated as the probability of cradle, and records the position coordinates and the probability in information table, due to the working range of cameraCommonly greater than the working range of laser radar, therefore the position of doubtful cradle target can be obtained faster, to rechargeWhen reduce searching times, improve search efficiency；When robot, which enters, recharges mode, according to the sequential search of probability from high to lowThe corresponding coordinate position of the probability, because probability is higher, a possibility that there are cradles, is bigger, therefore can search for fasterTo cradle；When the laser radar obtain flag information matched with the flag information on preset charged seat and camera identificationThere are when cradle, there are cradles for confirmation current location, confirm cradle jointly by laser radar and camera, as a result moreWith reliability.
In conclusion accuracy when being able to ascend identification cradle using the recognition methods of cradle provided by the present applicationAnd reliability, and then improve robot recharges efficiency.
Detailed description of the invention
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, to embodiment or will show belowThere is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only thisThe some embodiments recorded in application, for those of ordinary skill in the art, without creative efforts,It can also be obtained according to these attached drawings other attached drawings.
Fig. 1 is a kind of flow chart of the recognition methods for cradle that the embodiment of the present application one provides；
Fig. 2 is the schematic diagram of a scenario that the embodiment of the present application one provides；
Fig. 3 is the corresponding coordinate position distribution map of table 1 that the embodiment of the present application one provides；
Fig. 4 is the corresponding coordinate position distribution map of table 1 after the cluster that the embodiment of the present application one provides；
Fig. 5 is the schematic diagram of the flag information for the cradle that the embodiment of the present application one provides；
Fig. 6 is a kind of schematic diagram of the identification device for cradle that the embodiment of the present application two provides；
Fig. 7 is a kind of schematic diagram for robot that the embodiment of the present application three provides.
Specific embodiment
Currently, in intelligent robot during recharging, the recognition methods of cradle be based primarily upon simple laser radar,Or simple infrared signal.Its process is to pass through laser radar when intelligent robot electricity needs to charge lower than some threshold valueOr infrared sensor, search for the general pose of cradle.Then further according to the laser radar signal or infrared signal detected,The position for further determining that cradle eventually moves to charge on cradle.
But for using the scheme of laser radar merely, because the identification distance of laser radar is very short, usually effectivelyIdentification distance only has 1m or so, and the information content that synchronization obtains is insufficient, therefore it is too many to will lead to the region for needing to search for, machinePeople may need to be compared intensive search, need a large amount of search time, can to recharge efficiency reduction.
And for the scheme using infrared sensor, because there are very strong reflecting effect, intelligent robots for infrared signalThe exact position of cradle can not be determined according to infrared signal, may collide cradle so as to cause robot, it is any to damagePseudo-machine people and cradle.Furthermore.When working environment complexity, it is also possible to robot can be directed to the region of search of mistake,The complexity for increasing search, reduces and recharges efficiency.
In order to solve the above-mentioned technical problem, it this application provides a kind of recognition methods of cradle, device, robot and depositsStorage media combines the position for determining cradle by camera and laser radar, not only solves and is based purely on laser radar calculationThe problem of method information content is insufficient, causes search range too big, error rate is too high, inefficiency；And the shooting for passing through cameraImage, which is identified, to be also avoided that and is identified mistake due to caused by infrared reflection, furthermore carries out precise search knowledge in laser radarThe shooting image for also passing through camera when other carries out recognition and verification, further improves the accuracy of cradle pose and reliableProperty.
In order to make those skilled in the art more fully understand application scheme, below in conjunction in the embodiment of the present applicationAttached drawing, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described embodiment is only thisApply for a part of the embodiment, instead of all the embodiments.Based on the embodiment in the application, those of ordinary skill in the art existEvery other embodiment obtained under the premise of creative work is not made, shall fall in the protection scope of this application.
Embodiment one:
The embodiment of the present application one provides a kind of recognition methods of cradle, illustrates with reference to the accompanying drawing.
Referring to Fig. 1, which is a kind of flow chart of the recognition methods for cradle that the embodiment of the present application one provides.
The embodiment of the present application the described method comprises the following steps:
Step 101: obtaining the identification model of cradle in advance；Identification model from the shooting image of camera for identifyingCradle.
Since the identification effective distance of laser radar is shorter, in order to obtain bigger information content and reduce search space, thisApplication identifies cradle using camera.Therefore the identification for identifying cradle from the shooting image of camera is trained firstModel.
It advances with camera and obtains shooting image of the cradle under different angle and different distance, utilize deep learningAlgorithm Faster R-CNN is trained to obtain the identification model of cradle shooting image.
The cradle is cradle present in robot manipulating task map, utilizes the shooting of identification model identification video cameraWhen whether there is cradle in image, the probability in image there are cradle can be obtained.
Step 102: when being determined according to identification model in the shooting image of camera there are when doubtful cradle target, passing throughLaser radar obtains the position coordinates of doubtful cradle target, and obtaining doubtful cradle target according to identification model is cradleProbability, and record position coordinate and probability in information table.
Referring to fig. 2, which is the schematic diagram of a scenario that the embodiment of the present application one provides.
Robot 200 includes camera 201 and laser radar 202.Generally include multiple laser radars 202.
Shooting when robot build figure or execute task to working environment, according to identification model to cameraImage identified in real time, can according to the shooting image of camera when recognizing there are when doubtful cradle target 300It determines angle of the doubtful cradle target 300 relative to robot, and the doubtful charging can be obtained according to preset modelSeat target 300 is the probability of cradle.
Then laser radar 202 is scanned detection according to the angle, and multiple laser radars 202 can obtain multiple groups detectionData carry out calculating of averaging to the multiple groups detection data of acquisition, obtain the position coordinates of doubtful cradle target 300, the positionSetting coordinate includes abscissa and ordinate, identifies the position of the doubtful cradle target 300 in current map, coordinate origin,Change in coordinate axis direction and coordinate length unit etc. can be set according to actual working environment, and the application is not especially limited this.
The coordinate position for all doubtful cradle targets 300 that will confirm that and to because probability be stored in information table, sitThe corresponding ID of cursor position is determined by natural increase.Table 1 gives an example of information table, and the coordinate data of the table is correspondingUnit is rice (m).
1 information table of table
| ID | Abscissa | Ordinate | Probability |
| 1 | 1 | 2 | 0.6 |
| 2 | 3 | 7 | 0.8 |
| 3 | 2 | 7 | 0.3 |
| 4 | 2 | 9 | 0.9 |
| 5 | 9 | 10 | 0.5 |
| 6 | 8 | 1 | 0.3 |
| 7 | 10 | 2 | 0.7 |
In order to facilitate understanding, it is illustrated below with coordinate position distribution map.
Referring to Fig. 3, which is the corresponding coordinate position distribution map of table 1 that the embodiment of the present application one provides.
The position coordinates recorded in information table correspond to the different coordinate points in distribution map, which is robot building ringThe top view in border.
Further, when robot build figure or execute task to working environment, can be believed with real-time updateThe position coordinates and probability recorded in breath table, specifically includes the following steps:
Step 102a: it according to the captured in real-time image and identification model to position coordinates, determines and is doubted existing for position coordinatesIt is the real-time probability of cradle like cradle target, real-time probability is updated in information table.
Robot identifies the shooting image real-time perfoming of different location coordinate during mobile advance, and then obtainsThe doubtful cradle target under different shooting angles is obtained, the doubtful charging under different shooting angles can be obtained according to identification modelSeat target is the real-time probability of cradle.Then real-time probability is updated in information table, which may increase, it is also possible toReduce.
Step 102b: when real-time probability is lower than predetermined probabilities, real-time probability and the probability pair are deleted from information tableThe position coordinates answered.
When real-time probability is lower than predetermined probabilities, showing the doubtful cradle target on the position coordinates at this time not is chargingSeat, therefore this group of data are deleted from information table, and then reduce the quantity of position coordinates, to reduce search range, raising is searchedRope efficiency.The embodiment of the present application is not especially limited predetermined probabilities.
Step 103: when entrance recharges mode, according to the corresponding coordinate position of sequential search probability of probability from high to low.
When robot access recharges mode, the probability in information table is ranked up by numerical values recited, is first searched for largerThe corresponding position coordinates of probability show that doubtful cradle target 300 on the position coordinates is cradle because probability is biggerA possibility that highest, therefore can more quickly search cradle, reduce the space of search, avoid unnecessary search workMake.
Due to existing when obtaining angle of the doubtful cradle target 300 relative to robot according to the shooting image of cameraThere are measurement error when measurement error and laser radar 202 position doubtful cradle target 300, stored in information table certainA little coordinate positions may indicate the same doubtful cradle target 300, may result in work when robot search cradleWork amount increase further increases search efficiency to solve the above-mentioned problems, and this method is further comprising the steps of:
Step 103a: the position coordinates that distance is less than pre-determined distance threshold value are divided into same class position coordinates, every class positionSet the probability that the sum of corresponding probability of all position coordinates in coordinate is such position coordinates.
Referring to fig. 4, which is the corresponding coordinate position distribution map of table 1 after the cluster that the embodiment of the present application one provides.
When robot, which enters, recharges mode, robot just starts the search program to cradle.First in information tableThe position coordinates in face carry out cluster operation.As shown in figure 4, being clustered according to pre-determined distance threshold value R to position coordinates.For example,As R=3, it is classified as same category if the distance of two position coordinates is less than R, the doubtful cradle mesh of 7 in table 1Target position coordinates are clustered into four classes, are indicated respectively with A, B, C and D in figure.Then, it is determined that the center of each class andThe probability of such position coordinates.
It is often the probability of such position coordinates with the sum of the corresponding probability of all position coordinates in class position coordinates, then respectivelyThe probability of class position coordinates is respectively P (A)=0.6, P (B)=0.8+0.3+0.9=2.0, P (C)=0.5, P (D)=0.3+0.7=1.
Step 103b: according to the coordinate position of the probability every class of sequential search from high to low of every class position coordinates.
Then according to the probability height of every class position coordinates, from high to low, all classes are ranked up, obtain result successivelyFor B, D, A, C.The corresponding center of every class coordinate position is successively searched for according to the sequence.
In one possible implementation, the abscissa of the center is all position coordinates for including with suchThe average value of abscissa, the ordinate of the center are the average value of the ordinate for all position coordinates for including with such,Then the corresponding center of B class is (7/3,23/8), and the corresponding center of D class is (9,1.5), the corresponding center of A classIt is (9,10) for the corresponding center of (1,2) C class.
Storage corresponding with such center, forms new letter after the probability of every class position coordinates is arranged from high to lowTable is ceased, as shown in table 2.
2 new information table of table
| ID | Abscissa | Ordinate | Probability |
| 1 | 7/3 | 23/8 | 2.0 |
| 2 | 9 | 1.5 | 1.0 |
| 3 | 1 | 2 | 0.6 |
| 4 | 9 | 10 | 0.5 |
Step 104: when the flag information that laser radar obtains matches and camera with the flag information on preset charged seatIdentification is there are when cradle, and there are cradles for confirmation current location.
According to new information table obtained in the previous step, successively scan for traversing.One type position coordinates are scanned forWhen, it first passes through cartographic information and cooks up reasonable path, the center of guided robot to class.
When robot reaches center, starts with laser radar and ambient enviroment is scanned, to obtain markWill information, when the flag information that laser radar obtains is matched with the flag information on preset charged seat, laser radar confirmation is depositedIn cradle.Flag information on the preset charged seat is stored in the memory of robot.
When laser radar recognizes flag information, exact position and the direction of cradle are determined by flag information.TogetherWhen, factors such as interference of environmental factor, such as wallpaper, barrier in order to prevent, it is also necessary to by camera to determining accuratePosition carries out further recognition and verification, and then ensures the position really with the presence of cradle.
Referring to Fig. 5, which is the schematic diagram of the flag information for the cradle that the embodiment of the present application one provides.
Cradle passes through specially treated, and the flag information after laser radar scanning can be the small cube at black and white interval.
Determine exact position and the direction of cradle to robot, and after confirmed reliability by camera.MachineDevice people adjusts position and the angle of itself, is directed at cradle, then charges close to cradle.
When searching for a doubtful cradle target corresponding position coordinates, filled if laser radar does not recognizeFlag information or camera on electric seat can not finally confirm the presence for having cradle, then abandon the position, carry out nextThe search of position class.
Further, after all doubtful cradle targets corresponding position coordinates have been searched for by robot, if still withoutCradle is found, then robot will do it primary last full figure search, and process is exactly to plan the entire building ring of traversalThe path in border utilizes camera and laser radar to search for cradle when robot is in the path moving process.If robotIt covers entire path and does not all search cradle, then recharge mission failure, there is no cradle in report environmental, end, which recharges, appointsBusiness.
Method provided by the embodiments of the present application obtains the identification model of cradle in advance, which is used for from cameraShooting image in identify the cradle；When there are doubtful cradles in the shooting image for determining camera according to identification modelWhen target, the position coordinates of the doubtful cradle target are obtained by laser radar, and doubtful charging is obtained according to identification modelSeat target is the probability of cradle, and the position coordinates and the probability are recorded in information table, due to the work of cameraRange therefore can obtain the position of doubtful cradle target commonly greater than the working range of laser radar faster, so as toSearching times are reduced when recharging, and improve search efficiency；When robot, which enters, recharges mode, according to the sequence of probability from high to lowThe corresponding coordinate position of the probability is searched for, because probability is higher, a possibility that there are cradles is bigger, therefore can be fasterSearch cradle；When the flag information that the laser radar obtains matches and camera with the flag information on preset charged seatThere are when cradle, there are cradles for confirmation current location, confirm cradle jointly by laser radar and camera for identification, tieFruit has more reliability.
In conclusion being able to ascend when identifying cradle using the recognition methods of cradle provided by the embodiments of the present applicationAccuracy and reliability, and then improve robot recharges efficiency.
Embodiment two:
The recognition methods of the cradle provided based on the above embodiment, the embodiment of the present application two additionally provide a kind of cradleIdentification device, illustrate with reference to the accompanying drawing.
Referring to Fig. 6, which is a kind of schematic diagram of the identification device for cradle that the embodiment of the present application two provides.
Device provided by the embodiments of the present application includes: that the first acquisition module 601, second obtains module 602, search module603 and confirmation module 604.
First obtains module 601, and for obtaining the identification model of cradle in advance, which is used for from cameraCradle is identified in shooting image.
Specifically, the first acquisition module 601 advance with the camera obtain the cradle in different angle andShooting image under different distance is trained to obtain the shooting image using deep learning algorithm Faster R-CNNThe identification model of the cradle.
Second obtains module 602, determines that there are doubtful chargings in the shooting image of camera according to identification model for working asWhen seat target, the position coordinates of doubtful cradle target are obtained by laser radar, and described doubtful fill is obtained according to identification modelElectric seat target is the probability of cradle, and record position coordinate and probability in information table；.
Search module 603, when recharging mode for entering, according to the corresponding seat of sequential search probability of probability from high to lowCursor position.
Further, search module 603 can also include: classification submodule 603a and search submodule 603b.
Classification submodule 603a is used to the position coordinates that distance is less than pre-determined distance threshold value being divided into same class position and sitIt marks, the sum of corresponding probability of all position coordinates in every class position coordinates is the probability of such position coordinates.
Search for the coordinate of the every class of sequential search of submodule 603b from high to low for the probability according to every class position coordinatesPosition.
Confirmation module 604, the flag information for obtaining when laser radar are matched with the flag information on preset charged seatAnd camera identification, there are when cradle, there are cradles for confirmation current location.
Further, described device can also include: update module 605.
Position coordinates and probability of the update module 605 for being recorded in real-time update information table.
When not finding the cradle yet after having searched for all position coordinates, described search module 603 can be withThe path for planning a traversal working environment utilizes the camera and the laser thunder during path is mobileUp to search cradle.
The first acquisition module that device provided by the embodiments of the present application includes obtains the identification model of cradle, the knowledge in advanceOther model is for identifying the cradle from the shooting image of camera.Second obtains module can be when true according to identification modelDetermine to obtain the doubtful cradle target by laser radar there are when doubtful cradle target in the shooting image of cameraPosition coordinates obtain the probability that doubtful cradle target is cradle according to identification model, and record institute's rheme in information tableCoordinate and the probability are set, it, can be very fast since the working range of camera is commonly greater than the working range of laser radarThe position of the doubtful cradle target of acquisition improve search efficiency to reduce searching times when recharging.When robot entersWhen recharging mode, search module is according to the corresponding coordinate position of probability described in probability sequential search from high to low, because of probabilityHigher, a possibility that there are cradles, is bigger, therefore can search cradle faster.Confirmation module works as the laser radarThe flag information of acquisition is matched with the flag information on preset charged seat and camera identification is there are when cradle, confirms present bitIt sets there are cradle, cradle is confirmed by laser radar and camera jointly, as a result have more reliability.
Further, the identification device of the cradle includes processor and memory, and above-mentioned first obtains module, secondIt obtains module, search module and confirmation module etc. to store in memory as program module, be stored in by processor executionAbove procedure module in memory realizes corresponding function.
Include kernel in processor, is gone in memory to transfer corresponding program module by kernel.Kernel can be set oneOr more, the identification of cradle is carried out by adjusting kernel parameter.
Memory may include the non-volatile memory in computer-readable medium, random access memory (RAM) and/Or the forms such as Nonvolatile memory, if read-only memory (ROM) or flash memory (flash RAM), memory include that at least one is depositedStore up chip.
The embodiment of the present application also provides a kind of storage mediums, are stored thereon with program, when which is executed by processorIt can be realized the recognition methods of cradle described in embodiment one.
Embodiment three:
The identification device of the cradle provided based on the above embodiment, the embodiment of the present application three additionally provide a kind of machinePeople illustrates with reference to the accompanying drawing.
Referring to Fig. 7, which is a kind of schematic diagram for robot that the embodiment of the present application three provides.
The robot 700 includes the identification device 703 of camera 701, laser radar 702 and cradle.
Wherein, which includes at least two laser radars, and the embodiment of the present application does not make the quantity of laser radarIt is specific to limit.
The identification device 703 of the cradle includes: the first acquisition module, the second acquisition module, search module and confirmation mouldBlock.The explanation and working principle of identification device 703 about cradle may refer to above-described embodiment one and embodiment two, this ShenPlease embodiment details are not described herein.
Since the robot includes the identification device of cradle, the first acquisition module which includes obtains charging in advanceThe identification model of seat, the identification model is for identifying the cradle from the shooting image of camera.Second obtains module energyEnough when being determined in the shooting image of camera according to identification model there are when doubtful cradle target, institute is obtained by laser radarThe position coordinates for stating doubtful cradle target obtain doubtful cradle target according to identification model as the probability of cradle, andThe position coordinates and the probability are recorded in information table, due to camera working range commonly greater than laser radar workRange, therefore the position of doubtful cradle target can be obtained faster, to reduce searching times when recharging, improve searchEfficiency.When robot, which enters, recharges mode, search module is corresponding according to probability described in probability sequential search from high to lowCoordinate position, because probability is higher, a possibility that there are cradles, is bigger, therefore can search cradle faster.ConfirmationThe flag information that module works as laser radar acquisition is matched with the flag information on preset charged seat and camera identification existsWhen cradle, there are cradles for confirmation current location, confirm cradle jointly by laser radar and camera, as a result have moreReliability.
Therefore, which is executing the accuracy and success rate that search efficiency when recharging task is high, and cradle identifiesIt is high.
It should be appreciated that in this application, " at least one (item) " refers to one or more, and " multiple " refer to two or twoMore than a."and/or" indicates may exist three kinds of relationships, for example, " A and/or B " for describing the incidence relation of affiliated partnerIt can indicate: only exist A, only exist B and exist simultaneously tri- kinds of situations of A and B, wherein A, B can be odd number or plural number.WordSymbol "/" typicallys represent the relationship that forward-backward correlation object is a kind of "or"." at least one of following (a) " or its similar expression, refers toAny combination in these, any combination including individual event (a) or complex item (a).At least one of for example, in a, b or c(a) can indicate: a, b, c, " a and b ", " a and c ", " b and c ", or " a and b and c ", and wherein a, b, c can be individually, can alsoTo be multiple.
All the embodiments in this specification are described in a progressive manner, same and similar portion between each embodimentDividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for device realityFor applying example, since it is substantially similar to the method embodiment, so describing fairly simple, related place is referring to embodiment of the methodPart explanation.The apparatus embodiments described above are merely exemplary, wherein described be used as separate part descriptionUnit and module may or may not be physically separated.Furthermore it is also possible to select it according to the actual needsIn some or all of unit and module achieve the purpose of the solution of this embodiment.Those of ordinary skill in the art are not payingIn the case where creative work, it can understand and implement.
The above is only the specific embodiment of the application, it is noted that for the ordinary skill people of the artFor member, under the premise of not departing from the application principle, several improvements and modifications can also be made, these improvements and modifications are also answeredIt is considered as the protection scope of the application.