Method and device for starting operation of equipment, self-moving equipment and readable storage medium

Abstract

The application discloses a working starting method and device of equipment, a self-moving equipment and a readable storage medium, relates to the technical field of equipment control, and actively intelligently identifies whether equipment tools for performing cleaning work in a maintenance station are in place, does not need manual participation in identification, guarantees that the cleaning work is performed under the condition that the tools are in place, avoids fan failure or dust discharge pollution of air, and is better in intelligence. The method comprises the following steps: collecting an equipment image to be identified of the maintenance station in a pile returning process; performing image identification analysis on the equipment image to be identified, and obtaining an identification result; and when the identification result indicates that the equipment tools are in place, controlling the maintenance station to start cleaning work.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of device control, and in particular to a device work starting method and device, a self-moving device, and a readable storage medium. BACKGROUND

[0002] A self-moving device is a device capable of moving automatically under the control of a program, such as a sweeping self-moving device, a mopping self-moving device, a window wiping self-moving device, etc. The self-moving device can perform work by itself, and return to a maintenance station for charging when the work is completed, a return instruction is received, or the power is insufficient. The maintenance station can be a fixed station with functions of cleaning, charging, dust collection, etc. After the self-moving device returns to the maintenance station, the maintenance station will start the work of dust collection, charging, or cleaning the mop, etc. to charge or maintain the self-moving device according to the state of the self-moving device.

[0003] In the related art, after the maintenance station starts the dust collection work, it will separate dust based on a dust collection bag or a cyclone separator. However, when the dust collection bag or the cyclone separator in the maintenance station is not placed or installed, starting the dust collection work will cause unfiltered garbage to be sucked into the fan of the maintenance station, resulting in fan failure, or causing dust to be discharged through the air outlet to pollute the ambient air. Therefore, how to start the dust collection work after determining that the dust collection bag or the separator is in place has become a problem to be solved. SUMMARY

[0004] Therefore, the present application provides a device work starting method and device, a self-moving device, and a readable storage medium, which mainly aims to accurately identify whether a device tool for performing cleaning work is in place.

[0005] According to a first aspect of the present application, a device work starting method is provided, which comprises:

[0006] The self-moving device collects a to-be-identified device image of the maintenance station during a return process;

[0007] The to-be-identified device image is subjected to image recognition analysis to obtain an identification result;

[0008] Based on the identification result, the maintenance station is controlled to start or not to start a cleaning work after the self-moving device successfully returns to the station.

[0009] Optionally, the self-moving device collects a to-be-identified device image of the maintenance station during a return process, which comprises:

[0010] The self-moving device collects at least one device signal, determines a signal source of the at least one device signal according to a signal strength of the at least one device signal, and takes a position indicated by the signal source as a current device position of the maintenance station.

[0011] selecting a target position at a preset distance from the current position of the equipment according to the current position and orientation of the equipment of the maintenance station, and moving to the target position;

[0012] starting an image acquisition component when the mobile equipment reaches the target position, and acquiring an image of the maintenance station based on the image acquisition component to obtain the image of the equipment to be identified.

[0013] Optionally, the image recognition analysis on the image of the equipment to be identified includes:

[0014] obtaining an original equipment image, which is obtained by image acquisition on the maintenance station with or without equipment tools in place;

[0015] comparing image feature values of the image of the equipment to be identified and the original equipment image based on an image recognition algorithm to obtain a feature value matching degree;

[0016] The feature value matching degree is taken as the recognition result.

[0017] Optionally, the image recognition analysis on the image of the equipment to be identified includes:

[0018] obtaining an image recognition model, which is obtained by model training on an original equipment image obtained by image acquisition on the maintenance station with or without equipment tools in place;

[0019] inputting the image of the equipment to be identified into the image recognition model to obtain a feature value matching degree of the image of the equipment to be identified and the original equipment image output by the image recognition model;

[0020] The feature value matching degree is taken as the recognition result.

[0021] Optionally, after the image recognition analysis on the image of the equipment to be identified to obtain the recognition result, the method further includes:

[0022] determining a feature value matching degree indicated by the recognition result, obtaining a matching degree threshold, and comparing the feature value matching degree with the matching degree threshold;

[0023] When the feature value matching degree is less than the matching degree threshold, it is determined that the installation state of the equipment tool of the maintenance station is opposite to the installation state of the equipment tool in the original equipment image.

[0024] When the feature value matching degree is greater than or equal to the matching degree threshold, it is determined that the installation state of the equipment tool in the maintenance station at present matches the installation state of the equipment tool in the original equipment image.

[0025] Optionally, the method further comprises:

[0026] If the original equipment image is obtained by image acquisition on the maintenance station where the equipment tool is not in place, when it is determined that the installation state of the equipment tool in the maintenance station at present is opposite to the installation state of the equipment tool in the original equipment image, a comparison result indicating that the equipment tool at present is in place is generated, and when it is determined that the installation state of the equipment tool in the maintenance station at present matches the installation state of the equipment tool in the original equipment image, a comparison result indicating that the equipment tool at present is not in place is generated;

[0027] If the original equipment image is obtained by image acquisition on the maintenance station where the equipment tool is in place, when it is determined that the installation state of the equipment tool in the maintenance station at present is opposite to the installation state of the equipment tool in the original equipment image, a comparison result indicating that the equipment tool at present is not in place is generated, and when it is determined that the installation state of the equipment tool in the maintenance station at present matches the installation state of the equipment tool in the original equipment image, a comparison result indicating that the equipment tool at present is in place is generated.

[0028] Optionally, the control of the maintenance station to start or not to start cleaning work after the self-moving device successfully returns the pile based on the identification result comprises:

[0029] When the comparison result obtained based on the identification result indicates that the equipment tool is not in place, the maintenance station is controlled not to start cleaning work;

[0030] When the comparison result obtained based on the identification result indicates that the equipment tool is in place, the maintenance station is instructed to start cleaning work.

[0031] Optionally, the control of the maintenance station to start cleaning work comprises:

[0032] An in-place identifier indicating that the equipment tool is in place is obtained, the identification result is marked with the in-place identifier, and the marked identification result is transmitted to the maintenance station, so that the maintenance station identifies the in-place identifier and starts cleaning work; or,

[0033] A start instruction for starting cleaning work is generated, and the start instruction is transmitted to the maintenance station, so that the maintenance station starts cleaning work after receiving the start instruction.

[0034] Optionally, the instructing the maintenance station not to start the cleaning operation includes:

[0035] Obtaining a reminder mark indicating that the equipment tool is not in place, marking the recognition result with the reminder mark, and transmitting the marked recognition result to the maintenance station so that the maintenance station recognizes the reminder mark and does not start the cleaning work after the self-moving equipment successfully returns to the pile; or

[0036] A reminder instruction for reminding that the cleaning work cannot be started is generated, and the reminder instruction is transmitted to the maintenance station, so that the maintenance station does not start the cleaning work after receiving the reminder instruction and the self-moving device successfully returns to the pile.

[0037] Optionally, before the self-mobile device collects the image of the device to be identified at the maintenance station during the return process, the method further includes:

[0038] Start the image acquisition component at the target location, and acquire an image of the maintenance station where the equipment tool is not in place or in place based on the image acquisition component to obtain an original equipment image;

[0039] Performing image feature value extraction on the original device image to obtain a plurality of original image feature values;

[0040] The plurality of original image feature values ​​are used to perform model training to generate an image recognition model.

[0041] Optionally, the method further includes:

[0042] querying the height of the object corresponding to the recognition result, transmitting the height of the object to the maintenance station, so that when the maintenance station determines that the height of the object reaches a height threshold, a tool processing reminder is generated, and the tool processing reminder is transmitted to the user's terminal; or,

[0043] Obtain the height threshold, compare the height of the object contained with the height threshold, generate a tool processing instruction when the height of the object contained reaches the height threshold, and transmit the tool processing instruction to the maintenance station so that the maintenance station transmits the tool processing instruction to the user's terminal.

[0044] According to a second aspect of the present application, a device for starting a device is provided, the device comprising:

[0045] The acquisition module is used to collect images of the equipment to be identified at the maintenance station from the mobile device during the return process;

[0046] An analysis module, configured to perform image recognition analysis on the image of the device to be identified to obtain an identification result;

[0047] The control module is configured to control the maintenance station to start or not to start cleaning work after the self-moving device successfully returns to the pile based on the identification result.

[0048] Optionally, the collection module is configured to collect at least one device signal by the self-moving device, determine a signal source of the at least one device signal according to a signal strength of the at least one device signal, and take a position indicated by the signal source as a current device position of the maintenance station; select a target position at a preset distance from the device position according to the current device position and a device orientation of the maintenance station, and move to the target position; and start an image collection component by the self-moving device when the target position is reached, collect an image of the maintenance station based on the image collection component, and obtain the to-be-identified device image.

[0049] Optionally, the analysis module is configured to obtain an original device image, the original device image being obtained by collecting an image of a maintenance station with or without a device tool in place; compare image feature values of the to-be-identified device image and the original device image based on an image recognition algorithm, and obtain a feature value matching degree; and take the feature value matching degree as the identification result.

[0050] Optionally, the analysis module is configured to obtain an image recognition model, the image recognition model being obtained by model training on an original device image obtained by collecting an image of a maintenance station with or without a device tool in place; input the to-be-identified device image into the image recognition model, obtain a feature value matching degree of the to-be-identified device image and the original device image output by the image recognition model; and take the feature value matching degree as the identification result.

[0051] Optionally, the apparatus further comprises:

[0052] The comparison module is configured to determine a feature value matching degree indicated by the identification result, obtain a matching degree threshold, and compare the feature value matching degree with the matching degree threshold.

[0053] The determination module is configured to determine that an installation state of the device tool of the maintenance station is opposite to an installation state of the device tool in the original device image when the feature value matching degree is less than the matching degree threshold.

[0054] The determination module is further configured to determine that the installation state of the device tool of the maintenance station matches the installation state of the device tool in the original device image when the feature value matching degree is greater than or equal to the matching degree threshold.

[0055] Optionally, the apparatus further comprises:

[0056] The generating module is configured to, if the original equipment image is obtained by image collection on the maintenance station where the equipment tool is not in place, generate a comparison result indicating that the equipment tool is currently in place when it is determined that the installation state of the equipment tool at the maintenance station is opposite to the installation state of the equipment tool in the original equipment image, and generate a comparison result indicating that the equipment tool is currently not in place when it is determined that the installation state of the equipment tool at the maintenance station matches the installation state of the equipment tool in the original equipment image.

[0057] The generating module is further configured to, if the original equipment image is obtained by image collection on the maintenance station where the equipment tool is in place, generate a comparison result indicating that the equipment tool is currently not in place when it is determined that the installation state of the equipment tool at the maintenance station is opposite to the installation state of the equipment tool in the original equipment image, and generate a comparison result indicating that the equipment tool is currently in place when it is determined that the installation state of the equipment tool at the maintenance station matches the installation state of the equipment tool in the original equipment image.

[0058] Optionally, the control module is configured to control the maintenance station to start cleaning work when the comparison result based on the identification result indicates that the equipment tool is not in place, and instruct the maintenance station not to start cleaning work when the comparison result based on the identification result indicates that the equipment tool is in place.

[0059] Optionally, the control module is configured to acquire an in-place identifier indicating that the equipment tool is in place, mark the identification result with the in-place identifier, and transmit the marked identification result to the maintenance station, so that the maintenance station identifies the in-place identifier and starts cleaning work; or, generate a start instruction for starting cleaning work, and transmit the start instruction to the maintenance station, so that the maintenance station starts cleaning work after receiving the start instruction.

[0060] Optionally, the control module is configured to acquire a reminder identifier indicating that the equipment tool is not in place, mark the identification result with the reminder identifier, and transmit the marked identification result to the maintenance station, so that the maintenance station identifies the reminder identifier and does not start cleaning work after the self-moving device successfully returns to the pile; or, generate a reminder instruction for reminding that cleaning work cannot be started, and transmit the reminder instruction to the maintenance station, so that the maintenance station does not start cleaning work after the self-moving device successfully returns to the pile after receiving the reminder instruction.

[0061] Optionally, the collection module is further configured to start an image collection component at a target position, and collect images of the maintenance station where the equipment tool is not in place or in place based on the image collection component to obtain an original equipment image.

[0062] The device also includes:

[0063] An extraction module configured to extract image feature values from the original device image to obtain a plurality of original image feature values.

[0064] A training module configured to train a model using the plurality of original image feature values to generate an image recognition model.

[0065] Optionally, the control module is further configured to query a holding object height corresponding to the recognition result, transmit the holding object height to the maintenance station, so that the maintenance station generates a tool processing reminder when the holding object height reaches a height threshold, and transmits the tool processing reminder to a terminal of a user; or, obtain the height threshold, compare the holding object height with the height threshold, and generate a tool processing instruction when the holding object height reaches the height threshold, and transmit the tool processing instruction to the maintenance station, so that the maintenance station transmits the tool processing instruction to the terminal of the user.

[0066] According to a third aspect of the present application, a self-moving device is provided, which includes a memory and a processor, the memory stores a computer program, and the processor includes an image acquisition component, a communication unit, and an analysis and transportation unit. The image acquisition component is configured to acquire images. The communication unit is configured to transmit data with a maintenance station and a terminal. The analysis and transportation unit is configured to perform image recognition analysis. The processor executes the computer program to implement the steps of the method of any one of the first aspect.

[0067] According to a fourth aspect of the present application, a readable storage medium is provided, which stores a computer program. The computer program is executed by a processor to implement the steps of the method of any one of the first aspect.

[0068] By means of the above technical solutions, the device working start method and device, self-moving device, and readable storage medium provided by the present application are provided. In the present application, the self-moving device collects the to-be-recognized device image of the maintenance station during the pile returning process, performs image recognition analysis on the to-be-recognized device image, obtains a recognition result, and controls the maintenance station to start the cleaning work when the recognition result indicates that the device tool is in place. The device tool used for cleaning work in the maintenance station is actively identified, without the need for manual participation in identification, so that the cleaning work is ensured to be performed when the tool is in place, the possibility of fan failure or dust discharge pollution of air is avoided, and the intelligence is better.

[0069] The above description is only a summary of the technical solutions of the present application. In order to enable a more thorough understanding of the technical means of the present application, the present application can be implemented according to the content of the specification, and in order to enable the above and other purposes, characteristics and advantages of the present application to be more obvious and easy to understand, the following specific embodiments of the present application are described. BRIEF DESCRIPTION OF DRAWINGS

[0070] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The accompanying drawings are included to provide a description of the preferred embodiments and are not meant to limit the present application. Moreover, the same reference numerals in the attached drawings indicate the same or similar components. In the drawings:

[0071] FIG. 1A A flowchart of a method for starting the operation of a device is shown;

[0072] FIG. 1B A flowchart of a method for starting the operation of a device is shown;

[0073] FIG. 2A A flowchart of a method for starting the operation of a device is shown;

[0074] FIG. 2B A flowchart of a method for starting the operation of a device is shown;

[0075] FIG. 3 A flowchart of a method for starting the operation of a device is shown. DETAILED DESCRIPTION

[0076] Exemplary embodiments of the present application will be described herein below with reference to the accompanying drawings. Although exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. Rather, these embodiments are provided so that the present application can be more thoroughly understood and the scope of the present application can be accurately conveyed to those skilled in the art.

[0077] Many current self-moving devices such as floor cleaning robots, floor mopping robots, etc. are equipped with fixed stakes, which have functions such as charging, dust collection, and cleaning the mop. There are two main dust collection methods for the maintenance station. One is that the maintenance station has a detachable and installable dust collection bag in the dust collection barrel, which filters dust based on the dust collection bag. The other is that the maintenance station has a separator (such as a cyclone separator) in the dust collection barrel, which separates dust and gas to achieve the filtering effect. Both of these methods can separate dust to prevent dust from entering the fan compartment of the maintenance station and damaging the fan, or to prevent dust from being discharged through the air outlet and polluting the environment.

[0078] In fact, the maintenance station can also start the dust collection work when the dust collection bag or the separator is not placed, and in this case, starting the dust collection work will cause unfiltered garbage to be sucked into the fan, causing the fan to fail, so many maintenance stations currently detect whether the dust collection bag or the separator is in place. The detection of the dust collection bag or the separator can be achieved by designing the structure of the maintenance station, such as when the dust collection bag is not installed, the upper cover of the dust collection barrel of the maintenance station cannot be completely closed, thereby prompting the user; or adding a detection device to the maintenance station, when the detection device detects that the dust collection bag is not installed in place, the fan does not work, and the like. However, the above-mentioned ways of detecting whether the dust collection bag or the separator is in place have certain defects. Although the dust collection barrel cover cannot be closed when the dust collection bag is not in place, if the user does not notice that the dust collection barrel cover is not closed, the maintenance station fan can still work, which will still cause dust to leak into the dust collection barrel and the air, causing pollution. The scheme of adding a detection device to the maintenance station will cause the design of the maintenance station to be relatively complex, resulting in high cost of the maintenance station. Therefore, the present application provides a device work starting method, which collects the image of the to-be-identified device of the maintenance station during the back-pile process of the self-moving device, performs image recognition analysis on the to-be-identified device image, obtains an identification result, and controls the maintenance station to start the cleaning work when the identification result indicates that the device tool is in place. The device tool in the maintenance station is actively and intelligently identified whether it is in place, without manual participation, ensures that the dust collection work is performed when the dust collection tool is in place, avoids fan failure or dust discharge pollution, and has good intelligence.

[0079] The embodiment of the present application provides a device work starting method, as shown in the method, the method comprises the following steps of: FIG. 1A

[0080] 101、Self-moving device collects the image of the to-be-identified device of the maintenance station during the back-pile process.

[0081] The self-moving device comprises a memory and a processor, the memory stores a computer program, the processor comprises an image acquisition component, a communication unit and an analysis transportation unit, the image acquisition component is used for image acquisition and can be a camera, a camera and the like; the communication unit is used for data transmission with the maintenance station and a terminal; and the analysis transportation unit is used for identifying and analyzing the image collected by the image acquisition component.

[0082] ​There is a data link between the self-moving device and the terminal held by the user, and the user can issue mobile control instructions such as starting cleaning and returning to charging to the self-moving device by operating on the terminal, thereby freeing the user's hands. Therefore, the self-moving device will receive a mobile control instruction issued by the user based on the terminal to control the self-moving device to return to the maintenance station to perform charging, backwashing and other operations on the pile, so that the self-moving device will perform the return-to-pile operation in response to the mobile control instruction. It should be noted that the self-moving device will also automatically perform the return-to-pile operation after completing the corresponding work, and then return to the pile to perform charging, dust collection, cleaning and other operations, so the return-to-pile process of the self-moving device can be an active behavior or a passive behavior, which is not limited in the present application.

[0083] The maintenance station can be a maintenance station with dust collection, cleaning, charging and other functions, and is the destination of the return-to-pile process of the self-moving device. In the embodiments of the present application, the maintenance station has a communication module, and the self-moving device establishes a communication link with the maintenance station based on a wireless communication method such as Bluetooth, WIFI (Wireless-Fidelity, wireless network) or a wired method such as a serial port, I2C (Inter-Integrated Circuit, connection microcontroller bus), so that the two can communicate information through bidirectional communication. In the return-to-pile process, in order to determine whether the maintenance station is ready for the device tool, and further determine whether the maintenance station can start the cleaning work, the self-moving device will collect the image of the to-be-identified device of the maintenance station, so as to determine whether the dust collection bag, separator and other device tools are in place and whether the cleaning work can be started by identifying the image of the to-be-identified device in the subsequent process. It should be noted that the self-moving device can collect the image of the to-be-identified device through the image collection component such as the camera configured thereon.

[0084] 102, performing image recognition analysis on the to-be-identified device image to obtain a recognition result.

[0085] After the to-be-identified device image is collected, the self-moving device will perform image recognition analysis on the to-be-identified device image to obtain a recognition result indicating whether the device tool is in place, so as to determine whether to control the requested device to work according to the content indicated by the recognition result in the subsequent process.

[0086] The device tool can be a dust collection bag or a separator. In the embodiments of the present application, the dust collection bucket shell of the maintenance station is made of transparent or translucent material, and the image features of the dust collection bucket without the dust collection bag or the separator and the dust collection bucket with the dust collection bag or the separator are obviously different. Therefore, by performing image recognition analysis on the to-be-identified device image, the self-moving device can easily identify whether the device tool in the dust collection bucket is in place.

[0087] 103、based on the recognition result corresponding control maintenance station from mobile equipment back pile success after the start of cleaning work or do not start cleaning work.

[0088] Since the equipment tool meets the condition of starting the cleaning work when it is in place, the maintenance station can normally start the cleaning work, and the equipment tool is not in place, so the maintenance station cannot start the work, therefore, based on the recognition result corresponding control maintenance station from mobile equipment back pile success after the start of cleaning work or do not start cleaning work. Specifically, when the recognition result indicates that the equipment tool is in place, the self-moving equipment interacts with the maintenance station to control the maintenance station to start the cleaning work, otherwise, the maintenance station does not start the cleaning work. Wherein, when the equipment tool is a dust bag or a separator, the cleaning work started by the maintenance station is dust collection work; when the equipment tool is a sewage bucket, the cleaning work started by the maintenance station is mop cleaning work, etc., the cleaning work started by the maintenance station is determined according to the specific function of the equipment tool, and the specific content of the cleaning work is not limited in the present application.

[0089] In the actual application process, the technical scheme of the present application can be applied in the process of starting the dust collection work, taking the dust collection work of the maintenance station completed by the separator as an example, referring to FIG. 1B , the maintenance station can be a complete machine equipment with dust collection, charging, mop cleaning and other functions, including a transparent dust collection barrel, which is equipped with a detachable separator, which is also the equipment tool that needs to be identified whether it is in place in the present application.

[0090] In this way, through the technical scheme of the present application, based on the self-moving equipment with rich sensors and relatively strong computing power, the relevant data or operation instructions collected and analyzed are sent to the maintenance station which is not intelligent enough, so that the self-moving equipment and the maintenance station are linked together to realize the intelligent work of the maintenance station.

[0091] The method provided by the embodiment of the present application collects the image of the equipment to be identified of the maintenance station in the back pile process, analyzes the image to be identified, obtains the recognition result, and controls the maintenance station to start the cleaning work when the recognition result indicates that the equipment tool is in place. The equipment tool used for cleaning work in the maintenance station is actively identified whether it is in place, without the need for manual identification, which ensures that the cleaning work is carried out under the condition that the tool is in place, avoids the possibility of fan failure or dust pollution of air, and has good intelligence.

[0092] Further, as a refinement and expansion of the specific implementation manner of the above embodiment, in order to completely describe the specific implementation process of the embodiment, the embodiment of the present application provides another equipment work starting method, as shown in FIG. 2A , the method comprises:

[0093] 201、in the back pile process, the self-moving equipment moves to the target position according to the current equipment position and equipment orientation of the maintenance station.

[0094] In the embodiment of the present application, after the pile returning operation is started, the self-moving device moves towards the maintenance station, and after the self-moving device returns to the maintenance station, whether the maintenance station can normally start the cleaning work depends on whether the self-moving device can identify the device tool for performing the cleaning work in the collected image of the to-be-identified device. Considering that some device tools installed in the maintenance station can only be photographed at some fixed positions, for example, the dust collecting barrel of some dust collecting piles is at the left part of the dust collecting pile, and only when the photographing is performed at the position opposite to the dust collecting pile or at the left position of the dust collecting pile, the dust collecting barrel can be completely photographed, so as to identify whether the dust collecting bag or the separator has been installed in the dust collecting barrel. Therefore, during the pile returning process, the self-moving device selects a target position at a preset distance from the current device position of the maintenance station and moves towards the target position. That is, at least one better photographing position is selected as the target position, and the maintenance station is photographed after the self-moving device moves to the target position. In an optional embodiment, the process of determining the target position is as follows:

[0095] The self-moving device continuously moves, and when the self-moving device moves into the signal coverage range of the maintenance station, the device signal is continuously received, and the closer the self-moving device is to the maintenance station, the stronger the strength of the received device signal is, and the farther the self-moving device is from the maintenance station, the weaker the strength of the received device signal is. Therefore, after at least one device signal is collected, the self-moving device can determine the signal source of the at least one device signal according to the signal strength of the at least one device signal, and take the position indicated by the signal source as the current device position of the maintenance station.

[0096] Since the position of the signal unit for transmitting the device signal on the maintenance station is fixed, and the structure of the self-moving device matched with the maintenance station is known, the self-moving device can actually determine the device position of the maintenance station according to the signal source, and then the self-moving device can select a target position at a preset distance from the maintenance station according to the device position and move towards the target position. The preset distance can be set according to the photographing angle of the self-moving device and the height of the maintenance station, so that the self-moving device can photograph a clear image when being at a distance of the preset distance from the maintenance station, and ensure the accurate identification of the device tool later.

[0097] It should be noted that the above process of selecting the target position can be implemented based on a pile searching technology. In actual application, if the image of the to-be-identified device collected by the self-moving device at the target position is not complete or has missing parts, the self-moving device can also adjust the position until a clear and usable image of the to-be-identified device is collected. In addition, in order to improve the accuracy of identification, multiple target positions can be selected, and the image of the to-be-identified device is collected at each target position, and the multiple collection results are integrated to determine whether the to-be-identified device has been installed in the maintenance station, further improving the accuracy of identification. The way of determining the target position is not limited in the present application.

[0098] 202. When reaching the target position, start the image collection component, collect the image of the maintenance station based on the image collection component, and obtain the image of the to-be-identified device.

[0099] In the embodiment of the present application, when the self-moving device reaches the target position, the image collection component is started, the image of the maintenance station is collected based on the image collection component, and the image of the to-be-identified device is obtained. The image collection component can be a camera, a camera, etc. equipped on the self-moving device. The self-moving device shoots the camera towards the direction of the maintenance station, and the image collection can be realized, and the image of the to-be-identified device is obtained.

[0100] In addition, since the maintenance station is matched with the self-moving device when it leaves the factory, the shape, color and structure of the maintenance station are known to the self-moving device. In fact, when the self-moving device directs the image collection component towards the maintenance station, the area where the device tool should be installed can be easily identified. Therefore, when collecting the image of the to-be-identified device, the self-moving device can directly collect part of the image of the area where the device tool is located as the image of the to-be-identified device, without the need to collect the complete image of the maintenance station. Thus, subsequent identification is only performed on this part of the image, reducing the workload of image identification and improving the speed of image identification. The specific content of the image of the to-be-identified device is not limited in the present application.

[0101] 203. Perform image recognition analysis on the image of the to-be-identified device to obtain an identification result.

[0102] In the embodiment of the present application, after the image of the to-be-identified device is collected, the self-moving device starts to perform image recognition analysis on the image of the to-be-identified device, and then obtains an identification result. The image analysis unit built in the self-moving device can be used to perform recognition analysis on the image of the to-be-identified device. In an optional embodiment, the image of the to-be-identified device can be recognized in the following two ways:

[0103] The first way is to use a pre-stored image recognition algorithm for recognition. The mobile device pre-stores an original device image, which can be obtained by image acquisition of a maintenance station where the device tool is not in place, or by image acquisition of a maintenance station where the device tool is in place, i.e., when the maintenance station does not have a dust bag or separator installed, or when the maintenance station has a dust bag or separator installed at the target position, the image acquisition component is started. In actual work, the mobile device calls the built-in image analysis unit, compares the image feature values of the to-be-identified device image and the original device image based on the image recognition algorithm, obtains the feature value matching degree, and takes the feature value matching degree as the recognition result. The image recognition algorithm is used to extract image features such as LBP (Local Binary Pattern), HOG (Histogram of Oriented Gridients), etc. from the to-be-identified device image and the original device image, and a machine learning method is used to classify the extracted image features or a SVM (Support Vector Machine), BP (Back Propagation), logistic regression, etc. algorithm is used to fit, so as to obtain the feature value matching degree.

[0104] The second way is to train a model based on a neural network and use the trained model for recognition. When training the image recognition model, the mobile device also obtains the original device image, extracts the image feature values from the original device image, obtains a plurality of original image feature values, and uses the plurality of original image feature values for model training to generate an image recognition model. The image recognition model can be trained based on a CNN (Convolutional Neural Networks), RBF (Radial-Basis Function), etc. neural network algorithm, and the mobile device can collect original device images at the time of factory shipment or directly obtain original device images input by workers for training of the image recognition model, and then use the image recognition model for image recognition in subsequent image recognition, thereby realizing fast image recognition. In this way, since the image recognition model has been trained, when the to-be-identified device image is analyzed for image recognition, the mobile device obtains the image recognition model, inputs the to-be-identified device image into the image recognition model, obtains the feature value matching degree of the to-be-identified device image and the original device image output by the image recognition model, and takes the feature value matching degree as the recognition result.

[0105] 204、determine the feature value matching degree indicated by the recognition result, obtain a matching degree threshold, compare the feature value matching degree with the matching degree threshold, when the comparison result indicates that the current device tool is in place, control the maintenance station to start the cleaning work; when the comparison result indicates that the current device tool is not in place, instruct the maintenance station not to start the cleaning work.

[0106] In the embodiment of the present application, after obtaining the recognition result, it is determined whether the maintenance station has installed the device tool according to the indication of the recognition result. The feature value matching degree indicated by the recognition result actually indicates the similarity between the to-be-recognized device image and the original device image. Therefore, the higher the similarity between the to-be-recognized device image and the original device image, the more similar the installation state of the current device tool of the maintenance station to the installation state of the device tool in the original device image, and vice versa. In order to determine whether the device tool is in place according to the high or low of the feature value matching degree, a matching degree threshold is set in the embodiment of the present application. The feature value matching degree is evaluated by the matching degree threshold. When the feature value matching degree is less than the matching degree threshold, it indicates that the similarity between the to-be-recognized device image and the original device image is low, and it can be determined that the installation state of the current device tool is opposite to the installation state of the device tool in the original device image. When the feature value matching degree is greater than or equal to the matching degree threshold, it indicates that the similarity between the to-be-recognized device image and the original device image is high, and the installation state of the device tool on the maintenance station is matched with the installation state of the device tool in the original device image.

[0107] Correspondingly, if the original device image is obtained by image collection on the maintenance station where the device tool is not in place, when it is determined that the installation state of the current device tool of the maintenance station is opposite to the installation state of the device tool in the original device image, the comparison result indicating that the current device tool is in place is generated, and when it is determined that the installation state of the current device tool of the maintenance station is matched with the installation state of the device tool in the original device image, the comparison result indicating that the current device tool is not in place is generated. If the original device image is obtained by image collection on the maintenance station where the device tool is in place, when it is determined that the installation state of the current device tool of the maintenance station is opposite to the installation state of the device tool in the original device image, the comparison result indicating that the current device tool is not in place is generated, and when it is determined that the installation state of the current device tool of the maintenance station is matched with the installation state of the device tool in the original device image, the comparison result indicating that the current device tool is in place is generated.

[0108] It should be noted that the matching degree threshold can be obtained through experiments. Image acquisition is performed on the maintenance station of the installed device tool and the maintenance station of the uninstalled device tool to obtain two sample pictures. The sample similarity of the two sample pictures is calculated, and the two sample pictures are respectively shot at multiple angles and the sample similarity is calculated to obtain multiple sample similarities. The average of the multiple sample similarities is taken as the matching degree threshold, or the sample similarity with the highest occurrence frequency can also be directly taken as the matching degree threshold. The present application does not limit this. In addition, in the actual application, the matching degree threshold can be updated every 3 days, 5 days, 10 days, etc., so as to ensure accurate identification of whether the device tool is in place.

[0109] In addition, there are two ways for the mobile device to control the maintenance station to start the cleaning work. One way is to transmit the identification result to the maintenance station, so that the maintenance station reads the identification result by itself and judges whether the cleaning work can be started by itself. That is, the mobile device obtains the in-place identifier indicating that the device tool is in place, marks the identification result with the in-place identifier, and transmits the marked identification result to the maintenance station, so that the maintenance station identifies the in-place identifier and starts the cleaning work.

[0110] The other way is that the mobile device generates an instruction for controlling the maintenance station to start the work, and sends the instruction to the maintenance station, so that the maintenance station directly works according to the instruction without reading the identification result again. That is, the mobile device generates a start instruction for starting the cleaning work, and transmits the start instruction to the maintenance station, so that the maintenance station starts the cleaning work after receiving the start instruction.

[0111] In the actual application, since the cleaning work needs to start after the mobile device successfully returns to the pile and contacts the maintenance station, it is meaningless to perform the cleaning work when the mobile device does not return to the pile. Therefore, if the identification result or the start instruction for indicating that the maintenance station can start the cleaning work is received by the maintenance station before the mobile device successfully returns to the pile, the cleaning operation can be temporarily suspended, and the cleaning work can be started after it is detected that the mobile device successfully returns to the pile, so as to avoid the waste of electric energy caused by starting the cleaning work when the mobile device does not return to the pile. Alternatively, the mobile device can send the above-mentioned identification result or start instruction to the maintenance station only after successfully returning to the pile each time. The maintenance station starts the cleaning work as soon as it receives the identification result or the start instruction, thereby ensuring the continuity of the business. In the embodiment of the present application, when the current state of the maintenance station is more matched with the state of the uninstalled device tool, it can be considered that the device tool is not in place. Therefore, the mobile device determines that the identification result indicates that the device tool is not in place, and reminds the maintenance station that the cleaning work cannot be started temporarily and the user needs to be alarmed.

[0112] Similarly, the self-moving device can also adopt two ways to remind the maintenance station. One way is to transmit the recognition result to the maintenance station, so that the maintenance station reads the recognition result by itself and judges whether the cleaning work can be started by itself. That is, the self-moving device obtains the reminder identifier indicating that the device tool is not in place, marks the recognition result with the reminder identifier, and transmits the marked recognition result to the maintenance station, so that the maintenance station identifies the reminder identifier and does not start the cleaning work after the self-moving device is successfully returned to the pile. Another way is that the self-moving device generates a reminder instruction for reminding the maintenance station, and transmits the reminder instruction to the maintenance station, so that the maintenance station does not start the cleaning work after the self-moving device is successfully returned to the pile. That is, the self-moving device generates a reminder instruction for reminding that the cleaning work cannot be started, transmits the reminder instruction to the maintenance station, so that the maintenance station does not start the cleaning work for the self-moving device successfully returned to the pile after receiving the reminder instruction.

[0113] In addition, in addition to not starting the cleaning work for the self-moving device successfully returned to the pile, in the actual application process, the self-moving device or the maintenance station can also generate warning or reminder texts such as “dust collection bag not placed” and “separator not in place”, and intuitively remind the user by playing the texts as reminder information by voice, indicating light, and the like. The texts can also be pushed to the terminal bound to the self-moving device or the maintenance station in the form of a pop-up window or a banner for display, so as to achieve the purpose of reminding the user.

[0114] The above-described process is a process of learning the state of whether the device tool is not installed or installed on the maintenance station, so as to determine whether the device tool is in place by image recognition. In the actual application process, some dust collection bags are provided with transparent windows, or the dust collection barrels are transparent. The height of the dust collection bag or the current containing object can actually be recognized through the transparent window and the transparent dust collection barrel. Therefore, in the embodiment of the present application, the self-moving device can learn the device tool containing objects of different heights in advance, determine the picture features of the device tool containing objects of different heights, and then determine the height of the containing object in the device tool according to the recognition result obtained by image recognition analysis, and determine whether the user needs to be reminded to dump the containing object based on the height of the containing object. Specifically, whether the user needs to be reminded to dump the containing object can be executed by the self-moving device or the maintenance station, and the process is as follows:

[0115] If the maintenance station executes, the self-moving device can query the height of the containing object corresponding to the recognition result, and transmit the height of the containing object to the maintenance station, so that the maintenance station obtains the height threshold set by itself, and generates a tool processing reminder when it is determined that the height of the containing object reaches the height threshold. The tool processing reminder is transmitted to the terminal of the user to remind the user to dump or replace the dust collection bag.

[0116] If the self-moving device is executed, the self-moving device can directly acquire the height threshold, compare the container height with the height threshold, generate a tool processing instruction when the container height reaches the height threshold, transmit the tool processing instruction to the maintenance station, so that the maintenance station directly transmits the tool processing instruction to the bound terminal to remind the user to dump or replace the dust bag, without the need for height comparison operation.

[0117] It should be noted that since the self-moving device itself has communication function, in fact, when the container height reaches the height threshold, the self-moving device can generate a tool processing reminder or instruction to the user's terminal by itself, directly reminding the user to dump, without relying on the maintenance station for reminding, shortening the data transmission link and improving the reminding rate. In this way, the self-moving device is pre-learned by the algorithm to learn pictures of various different heights of containers, so that the self-moving device can detect whether the device tool has reached the maximum capacity height, and then intelligently prompt the user to dump the dust bucket dust or replace the dust bag.

[0118] In summary, the technical scheme of the present application is summarized as follows:

[0119] Referring to FIG. 2B , the self-moving device starts to return to the stake, and the image acquisition component is started to collect image data of the maintenance station as a to-be-identified device image in the process of returning to the stake. Then, the self-moving device identifies the to-be-identified device image to determine whether the device tool for performing cleaning work is in place. When it is determined that the device tool is in place, the maintenance station is controlled to start the cleaning work, and when it is determined that the device tool is not in place, the cleaning work is not started, and a reminder information is fed back.

[0120] The method provided by the embodiment of the present application collects the to-be-identified device image of the maintenance station in the process of returning to the stake, performs image recognition analysis on the to-be-identified device image to obtain an identification result, and controls the maintenance station to start the cleaning work when the identification result indicates that the device tool is in place. The device tool in the maintenance station for performing cleaning work is actively identified without manual participation, ensuring that the cleaning work is performed when the tool is in place, avoiding the possibility of fan failure or dust discharge polluting the air, and having good intelligence.

[0121] Further, as a specific implementation of the method, FIG. 1A the embodiment of the present application provides a device work starting device, as shown in FIG. 3 , the device includes an acquisition module 301, an analysis module 302 and a control module 303.

[0122] The acquisition module 301 is used to collect the to-be-identified device image of the maintenance station in the process of returning to the stake by the self-moving device;

[0123] The analysis module 302 is configured to perform image recognition analysis on the device image to be identified to obtain an identification result.

[0124] The control module 303 is configured to control the maintenance station to start or not to start cleaning work after the self-moving device successfully returns to the pile based on the identification result.

[0125] In a specific application scenario, the collection module 301 is configured to collect at least one device signal by the self-moving device, determine a signal source of the at least one device signal according to a signal strength of the at least one device signal, and take a position indicated by the signal source as a current device position of the maintenance station; select a target position at a preset distance from the device position according to the current device position and a device orientation of the maintenance station, and move to the target position; and when the target position is reached, the self-moving device starts an image collection component, performs image collection on the maintenance station based on the image collection component, and obtains the device image to be identified.

[0126] In a specific application scenario, the analysis module 302 is configured to obtain an original device image, the original device image being obtained by performing image collection on a maintenance station with or without a device tool in place; compare image feature values of the device image to be identified and the original device image based on an image recognition algorithm to obtain a feature value matching degree; and take the feature value matching degree as the identification result.

[0127] In a specific application scenario, the analysis module 302 is configured to obtain an image recognition model, the image recognition model being obtained by model training on an original device image obtained by performing image collection on a maintenance station with or without a device tool in place; input the device image to be identified into the image recognition model to obtain a feature value matching degree of the device image to be identified and the original device image output by the image recognition model; and take the feature value matching degree as the identification result.

[0128] In a specific application scenario, the device further comprises:

[0129] A comparison module is configured to determine a feature value matching degree indicated by the identification result, obtain a matching degree threshold, and compare the feature value matching degree with the matching degree threshold.

[0130] A determination module is configured to determine that an installation state of the device tool of the maintenance station is opposite to an installation state of the device tool in the original device image when the feature value matching degree is less than the matching degree threshold.

[0131] The determining module is further configured to determine that the installation state of the equipment tool in the maintenance station currently matches the installation state of the equipment tool in the original equipment image when the feature value matching degree is greater than or equal to the matching degree threshold.

[0132] In a specific application scenario, the apparatus further includes:

[0133] The generating module is configured to, if the original equipment image is obtained by image collection on the maintenance station where the equipment tool is not in place, generate a comparison result indicating that the equipment tool is in place currently when it is determined that the installation state of the equipment tool in the maintenance station currently is opposite to the installation state of the equipment tool in the original equipment image, and generate a comparison result indicating that the equipment tool is not in place currently when it is determined that the installation state of the equipment tool in the maintenance station currently matches the installation state of the equipment tool in the original equipment image.

[0134] The generating module is further configured to, if the original equipment image is obtained by image collection on the maintenance station where the equipment tool is in place, generate a comparison result indicating that the equipment tool is not in place currently when it is determined that the installation state of the equipment tool in the maintenance station currently is opposite to the installation state of the equipment tool in the original equipment image, and generate a comparison result indicating that the equipment tool is in place currently when it is determined that the installation state of the equipment tool in the maintenance station currently matches the installation state of the equipment tool in the original equipment image.

[0135] In a specific application scenario, the control module 303 is configured to control the maintenance station to start cleaning work when the comparison result based on the identification result indicates that the equipment tool is not in place, and instruct the maintenance station not to start cleaning work when the comparison result based on the identification result indicates that the equipment tool is in place.

[0136] In a specific application scenario, the control module 303 is configured to acquire an in-place identifier indicating that the equipment tool is in place, mark the identification result by using the in-place identifier, and transmit the marked identification result to the maintenance station, so that the maintenance station identifies the in-place identifier and starts cleaning work; or, generate a start instruction for starting cleaning work, and transmit the start instruction to the maintenance station, so that the maintenance station starts cleaning work after receiving the start instruction.

[0137] In a specific application scenario, the control module 303 is configured to acquire a prompt identifier indicating that the device tool is not in place, mark the identification result by using the prompt identifier, and transmit the marked identification result to the maintenance station, so that the maintenance station identifies the prompt identifier and does not start the cleaning work after the self-moving device successfully returns to the pile; or, generate a prompt instruction for prompting that the cleaning work cannot be started, and transmit the prompt instruction to the maintenance station, so that the maintenance station does not start the cleaning work after the self-moving device successfully returns to the pile upon receiving the prompt instruction.

[0138] In a specific application scenario, the acquisition module 301 is further configured to start an image acquisition component at the target position, perform image acquisition on the maintenance station with or without the device tool based on the image acquisition component, and obtain an original device image.

[0139] The device further includes:

[0140] The extraction module is configured to extract image feature values from the original device image to obtain a plurality of original image feature values.

[0141] The training module is configured to perform model training by using the plurality of original image feature values to generate an image recognition model.

[0142] In a specific application scenario, the control module 303 is further configured to query a holding object height corresponding to the identification result, transmit the holding object height to the maintenance station, so that the maintenance station generates a tool processing prompt when the holding object height reaches a height threshold, and transmits the tool processing prompt to a terminal of a user; or, acquire the height threshold, compare the holding object height with the height threshold, and generate a tool processing instruction when the holding object height reaches the height threshold, and transmit the tool processing instruction to the maintenance station, so that the maintenance station transmits the tool processing instruction to the terminal of the user.

[0143] The device provided by the embodiment of the present application collects the to-be-identified device image of the maintenance station in the returning process of the self-moving device, performs image recognition analysis on the to-be-identified device image to obtain an identification result, and controls the maintenance station to start the cleaning work when the identification result indicates that the device tool is in place. The device actively identifies whether the device tool for performing the cleaning work in the maintenance station is in place, without the need for manual participation in identification, ensures that the cleaning work is performed under the condition that the tool is in place, avoids the possibility of fan failure or dust discharge polluting the air, and has good intelligence.

[0144] It should be noted that other corresponding descriptions of the functions of the device work starting device provided by the embodiment of the present application can be referred to FIG. 1A and FIG. 2A to FIG. 2B FIG. 3The corresponding description in the foregoing embodiments is not repeated here.

[0145] In an exemplary embodiment, a self-moving device is also provided, comprising a memory and a processor, the memory storing a computer program, the processor comprising an image acquisition component, a communication unit, and an analysis transportation unit, characterized in that the image acquisition component is configured to acquire images, the communication unit is configured to transmit data with a maintenance station and a terminal, and the analysis transportation unit is configured to perform image recognition analysis, and the processor executes the computer program to execute the working start method of the device in the foregoing embodiments.

[0146] A computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the steps of the working start method of the device.

[0147] From the above description of the embodiments, those skilled in the art can clearly understand that the present application can be implemented by hardware, or by means of software and a necessary general hardware platform. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product. The software product can be stored in a non-volatile storage medium (which can be a CD-ROM, a U disk, a mobile hard disk, etc.), and includes several instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in various embodiments of the present application.

[0148] Those skilled in the art can understand that the accompanying drawings are only schematic of a preferred embodiment, and the modules or flows in the drawings are not necessarily required for implementing the present application.

[0149] Those skilled in the art can understand that the modules in the device in the embodiments can be distributed in the device in the embodiments as described, or can be changed and located in one or more devices different from the embodiments. The modules in the above embodiments can be combined into one module, or can be further split into multiple sub-modules.

[0150] The above application numbers are only for description, and do not represent the advantages or disadvantages of the embodiments.

[0151] The above disclosure is only a few specific embodiments of the present application, but the present application is not limited thereto. Any changes that can be thought of by those skilled in the art should fall within the protection scope of the present application.

Claims

1. A method of starting operation of an apparatus, characterized by, The method comprises the following steps: The self-moving device collects the device image to be identified of the maintenance station during the process of returning to the stake; Image recognition analysis is performed on the device image to be identified, and an identification result is obtained; Based on the identification result, the maintenance station is controlled to start or not to start the cleaning work after the self-moving device successfully returns to the stake; A tool processing reminder is transmitted to the terminal of the user; or, the maintenance station transmits a tool processing instruction to the terminal of the user, specifically: The height of the containing object corresponding to the identification result is queried, and the height of the containing object is transmitted to the maintenance station, so that the maintenance station generates a tool processing reminder when the height of the containing object reaches a height threshold; or, the height threshold is obtained, the height of the containing object is compared with the height threshold, and a tool processing instruction is generated when the height of the containing object reaches the height threshold, and the tool processing instruction is transmitted to the maintenance station; The height of the containing object is determined by the self-moving device learning the device tools containing different heights of containing objects and determining the picture features of the device tools containing different heights of containing objects, and the identification result obtained by image recognition analysis.

2. The method of claim 1, wherein, The self-moving device collects the device image to be identified of the maintenance station during the process of returning to the stake, comprising: The self-moving device collects at least one device signal, determines the signal source of the at least one device signal according to the signal strength of the at least one device signal, and takes the position indicated by the signal source as the current device position of the maintenance station; According to the current device position and the device orientation of the maintenance station, a target position within a preset distance from the device position is selected, and the self-moving device moves to the target position; When the target position is reached, the self-moving device starts an image collection component, performs image collection on the maintenance station based on the image collection component, and obtains the device image to be identified.

3. The method of claim 1, wherein, The method comprises the following steps: An original device image is obtained, which is obtained by image collection on the maintenance station with or without device tools in place; Based on an image recognition algorithm, the image feature values of the device image to be identified and the original device image are compared to obtain a feature value matching degree; The feature value matching degree is taken as the identification result.

4. The method of claim 1, wherein, The method comprises the following steps: An image recognition model is obtained, which is obtained by model training on the original device image obtained by image collection on the maintenance station with or without device tools in place; The device image to be identified is input into the image recognition model to obtain the feature value matching degree of the device image to be identified and the original device image output by the image recognition model; The feature value matching degree is taken as the identification result.

5. The method of claim 1, wherein, After the image recognition analysis on the device image to be identified is performed to obtain the identification result, the method further comprises the following steps: The feature value matching degree indicated by the identification result is determined, a matching degree threshold is obtained, and the feature value matching degree is compared with the matching degree threshold; determining that the installation state of the equipment tool of the maintenance station currently is opposite to the installation state of the equipment tool in the original equipment image when the feature value matching degree is less than the matching degree threshold; determining that the installation state of the equipment tool of the maintenance station currently matches the installation state of the equipment tool in the original equipment image when the feature value matching degree is greater than or equal to the matching degree threshold.

6. The method of claim 5, wherein, The method further comprises: if the original equipment image is obtained by image collection on the maintenance station where the equipment tool is not in place, generating a comparison result indicating that the equipment tool is currently in place when it is determined that the installation state of the equipment tool of the maintenance station currently is opposite to the installation state of the equipment tool in the original equipment image, and generating a comparison result indicating that the equipment tool is currently not in place when it is determined that the installation state of the equipment tool of the maintenance station currently matches the installation state of the equipment tool in the original equipment image; if the original equipment image is obtained by image collection on the maintenance station where the equipment tool is in place, generating a comparison result indicating that the equipment tool is currently not in place when it is determined that the installation state of the equipment tool of the maintenance station currently is opposite to the installation state of the equipment tool in the original equipment image, and generating a comparison result indicating that the equipment tool is currently in place when it is determined that the installation state of the equipment tool of the maintenance station currently matches the installation state of the equipment tool in the original equipment image.

7. The method of claim 1, wherein, The control of the maintenance station to start or not to start cleaning work after the self-moving device successfully returns to the pile based on the identification result comprises: controlling the maintenance station to not start cleaning work when the comparison result obtained based on the identification result indicates that the equipment tool is not in place; indicating the maintenance station to start cleaning work when the comparison result obtained based on the identification result indicates that the equipment tool is in place.

8. The method of claim 7, wherein, The control of the maintenance station to start cleaning work comprises: obtaining an in-place identifier for indicating that the equipment tool is in place, marking the identification result with the in-place identifier, and transmitting the marked identification result to the maintenance station, so that the maintenance station identifies the in-place identifier and starts cleaning work; or generating a start instruction for starting cleaning work, and transmitting the start instruction to the maintenance station, so that the maintenance station starts cleaning work after receiving the start instruction.

9. The method of claim 7, wherein, The indication of the maintenance station to not start cleaning work comprises: obtaining a reminder identifier for indicating that the equipment tool is not in place, marking the identification result with the reminder identifier, and transmitting the marked identification result to the maintenance station, so that the maintenance station identifies the reminder identifier and does not start cleaning work after the self-moving device successfully returns to the pile; or generating a reminder instruction for reminding that cleaning work cannot be started, and transmitting the reminder instruction to the maintenance station, so that the maintenance station does not start cleaning work after the self-moving device successfully returns to the pile after receiving the reminder instruction.

10. The method of claim 1, wherein, Before the self-moving device collects the to-be-identified device image of the maintenance station in the pile returning process, the method further comprises: starting an image collection component at the target position, and performing image collection on the maintenance station based on whether the device tool is in place or not in place, to obtain an original device image; extracting image feature values from the original device image to obtain a plurality of original image feature values; training a model using the plurality of original image feature values to generate an image recognition model.

11. An apparatus for starting operation of a device, characterized by comprising: Comprise: a collection module configured to collect a to-be-identified device image of a maintenance station by a self-moving device in a pile returning process; an analysis module configured to perform image recognition analysis on the to-be-identified device image to obtain an identification result; a control module configured to control the maintenance station to start or not to start cleaning work after the self-moving device successfully returns the pile based on the identification result; the control module is also configured to transmit a tool processing reminder to a terminal of a user; or, the maintenance station transmits the tool processing instruction to the terminal of the user, specifically: querying a containing object height corresponding to the identification result, and transmitting the containing object height to the maintenance station, so that the maintenance station generates a tool processing reminder when the containing object height reaches a height threshold; or, obtaining the height threshold, comparing the containing object height with the height threshold, and generating a tool processing instruction when the containing object height reaches the height threshold, and transmitting the tool processing instruction to the maintenance station; wherein the containing object height is obtained by the self-moving device learning device tools containing different heights of containing objects, determining picture features of device tools containing different heights of containing objects, and obtaining an identification result through image recognition analysis.

12. The apparatus of claim 11, wherein, The collection module is configured to collect at least one device signal, determine the signal source of the at least one device signal according to the signal strength of the at least one device signal, and take the position indicated by the signal source as the current device position of the maintenance station; According to the current device position and the device orientation of the maintenance station, a target position within a preset distance from the device position is selected, and the self-moving device moves to the target position; when reaching the target position, the self-moving device starts an image collection component, and performs image collection on the maintenance station based on the image collection component to obtain the to-be-identified device image.

13. The apparatus of claim 11, wherein, The analysis module is configured to obtain an original device image, which is obtained by performing image collection on a maintenance station with a device tool not in place or in place; based on an image recognition algorithm, comparing the image feature values of the to-be-identified device image and the original device image to obtain a feature value matching degree; The feature value matching degree is taken as the identification result.

14. The apparatus of claim 11, wherein, The analysis module is configured to obtain an image recognition model, which is obtained by training a model using an original device image obtained by performing image collection on a maintenance station with a device tool not in place or in place; inputting the to-be-identified device image into the image recognition model to obtain a feature value matching degree of the to-be-identified device image and the original device image output by the image recognition model; The feature value matching degree is used as the recognition result.

15. The apparatus of claim 11, wherein, The device further comprises: a comparison module, configured to determine a matching degree of a feature value indicated by the recognition result, obtain a matching degree threshold, and compare the matching degree of the feature value with the matching degree threshold; a determination module, configured to determine that the current installation state of the equipment tool at the maintenance station is opposite to the installation state of the equipment tool in the original equipment image when the feature value matching degree is less than the matching degree threshold; The determining module is further configured to determine that the current installation status of the equipment tool at the maintenance station matches the installation status of the equipment tool in the original equipment image when the feature value matching degree is greater than or equal to the matching degree threshold.

16. The apparatus of claim 15, wherein, The device further comprises: a generating module configured to, if the original device image is obtained by capturing an image of a maintenance station in which the device tool is not in place, generate a comparison result indicating that the device tool is currently in place when it is determined that the current installation state of the device tool at the maintenance station is opposite to the installation state of the device tool in the original device image, and to generate a comparison result indicating that the device tool is currently not in place when it is determined that the current installation state of the device tool at the maintenance station matches the installation state of the device tool in the original device image; The generation module is further configured to generate a comparison result indicating that the equipment tool is not currently in place when it is determined that the current installation status of the equipment tool at the maintenance station is opposite to the installation status of the equipment tool in the original equipment image, if the original equipment image is obtained by image capture of the maintenance station with the equipment tool in place; and to generate a comparison result indicating that the equipment tool is currently in place when it is determined that the current installation status of the equipment tool at the maintenance station matches the installation status of the equipment tool in the original equipment image.

17. The apparatus of claim 11, wherein, The control module is used to control the maintenance station to start cleaning work when the comparison result obtained based on the recognition result indicates that the equipment tool is not in place; when the comparison result obtained based on the recognition result indicates that the equipment tool is in place, instruct the maintenance station not to start cleaning work.

18. The apparatus of claim 17, wherein, The control module is used to obtain an on-site identifier indicating that the equipment tool is in place, mark the identification result with the on-site identifier, and transmit the marked identification result to the maintenance station so that the maintenance station recognizes the on-site identifier and starts the cleaning work; or, generate a start instruction for starting the cleaning work, transmit the start instruction to the maintenance station, so that the maintenance station starts the cleaning work after receiving the start instruction.

19. The apparatus of claim 17, wherein, The control module is configured to acquire a prompt identifier indicating that the equipment tool is not in place, mark the identification result with the prompt identifier, and transmit the marked identification result to the maintenance station to enable the maintenance station to identify the prompt identifier and not start the cleaning work after the self-moving equipment is successfully returned to the stake.

20. The apparatus of claim 11, wherein, The collection module is further configured to start an image collection component at the target position, perform image collection on the maintenance station with the equipment tool not in place or in place based on the image collection component, and obtain an original equipment image. The device further comprises: An extraction module configured to extract image feature values from the original equipment image to obtain a plurality of original image feature values; A training module configured to perform model training using the plurality of original image feature values to generate an image recognition model.

21. A self-moving device comprising a memory and a processor, the memory storing a computer program, the processor comprising an image acquisition component, a communication unit, an analysis and transport unit, characterized in that, The image collection component is configured to perform image collection, the communication unit is configured to perform data transmission with the maintenance station and the terminal, the analysis and transportation unit is configured to perform image recognition analysis, and the processor implements the steps of the method in any one of claims 1 to 10 when executing the computer program.

22. A readable storage medium, having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method in any one of claims 1 to 10.

Images