Control method of automatic pool cleaning device, automatic pool cleaning device and computer storage medium
By acquiring multiple frames of images from the automatic pool cleaning device and using an image recognition model to determine the cleaning target, the problem of misidentifying damaged areas on the pool bottom or walls was solved, thus improving cleaning efficiency.
Patent Information
- Application Number
- CN202511253984.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-02
- Publication Date
- 2025-10-17
AI Technical Summary
Automatic pool cleaning devices can easily misidentify damaged areas on the pool bottom or walls as dirt, leading to repeated cleaning and reduced cleaning efficiency.
Multiple frames of images are captured by image acquisition equipment, and combined with image recognition models to determine whether the cleaning target is a real cleaning target, thus avoiding repeated cleaning of non-real cleaning targets.
This improved the cleaning efficiency of the automatic pool cleaning device and reduced misidentification of non-real cleaning targets and repeated cleaning.
Smart Images
Figure CN120802961A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] Embodiments of the present application relate to the technical field of cleaning devices, in particular to a control method of a pool automatic cleaning device, a pool automatic cleaning device and a computer storage medium. BACKGROUND
[0002] With the development of computer technology, robot technology has also developed rapidly. At present, underwater robots are increasingly widely used in various fields, and can assist people in working in water, including underwater cleaning, underwater exploration, underwater sightseeing, etc.
[0003] The robot for cleaning the pool, such as the pool automatic cleaning device, can identify and clean the target to be cleaned in the pool when cleaning the pool. However, there may be features in the pool that are easily mistaken for the target to be cleaned, such as damaged areas of the pool bottom or pool wall that are easily mistaken for dirt (such as leaves). In the case of misjudgment, the pool automatic cleaning device is prone to repeatedly attempting to clean the misidentified target to be cleaned, resulting in a decrease in the cleaning efficiency of the pool automatic cleaning device. SUMMARY
[0004] According to a first aspect of the present application, a control method of a pool automatic cleaning device is provided, the pool automatic cleaning device comprising an image acquisition device, the control method comprising: acquiring an image of an object around the pool automatic cleaning device by the image acquisition device during movement of the pool automatic cleaning device in a pool; after identifying a cleaning target based on the image, controlling the pool automatic cleaning device to move to the cleaning target; during movement to the cleaning target, acquiring multiple frames of images with the cleaning target by the image acquisition device; determining whether the cleaning target is a real cleaning target based on the multiple frames of images, and if not, controlling the pool automatic cleaning device to no longer clean the cleaning target when the pool automatic cleaning device identifies the cleaning target again.
[0005] According to a second aspect of the present application, a pool automatic cleaning device is provided, wherein the pool automatic cleaning device can perform the control method according to any one of the above.
[0006] According to a third aspect of the present application, a computer storage medium is provided, the storage medium storing a computer program, the computer program being executed by a processor to implement the control method according to any one of the above.
[0007] The embodiments described in the present application have the following beneficial effects:
[0008] The control method of the pool automatic cleaning device provided in the application can collect images of objects around the pool automatic cleaning device through an image collection device during movement of the pool automatic cleaning device, can control the pool automatic cleaning device to move to a cleaning target after identifying the cleaning target according to the images, and can collect multiple images during the movement, and then rejudge the cleaning target according to the collected multiple images to determine whether the identified cleaning target is a real cleaning target. By rejudging the identified cleaning target, the occurrence of misidentification can be effectively reduced, and in the case that the cleaning target is not a real cleaning target, the cleaning target is no longer cleaned when the cleaning target is identified again, so that repeated attempts to clean the non-real cleaning target by the pool automatic cleaning device can be avoided, and the cleaning efficiency of the pool automatic cleaning device is effectively improved. BRIEF DESCRIPTION OF DRAWINGS
[0009] In order to more clearly illustrate the technical solutions of the embodiments of the application, the drawings needed to be used in the description of the embodiments will be briefly introduced. The drawings in the following description are only exemplary embodiments of the application.
[0010] Figure 1 is a flowchart of the control method of the pool automatic cleaning device provided by the application; and
[0011] Figure 2 is a schematic view of the pool automatic cleaning device provided by the application when cleaning the pool bottom;
[0012] Figure 3 is a schematic view of the pool automatic cleaning device provided by the application when cleaning the water surface. DETAILED DESCRIPTION
[0013] In order to make the purpose, technical solutions and advantages of the application more clear, the technical solutions in the application will be clearly and completely described below in combination with the drawings in the application. Obviously, the described embodiments are part of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the application.
[0014] The application provides a control method of an automatic pool cleaning device. It can be understood that the automatic pool cleaning device can clean a pool, for example, a pool-shaped building, and the automatic pool cleaning device comprises an image acquisition device, so that the automatic pool cleaning device can obtain an environmental image around the automatic pool cleaning device through the image acquisition device when cleaning the pool. The pool-shaped building can be a swimming pool, a water storage pool, a hydrotherapy pool, a water storage tank, a water storage tank, etc. The automatic pool cleaning device can be a device such as an automatic cleaning device, a pool cleaning robot, etc., which can clean the pool-shaped building. The specific presentation of the pool-shaped building in the application is not limited, as long as the principle of the application can be realized.
[0015] Hereinafter, if not specially stated, the robot will be described as an example of the automatic pool cleaning device, and the swimming pool will be described as an example of the pool or pool-shaped building. Hereinafter, if not specially stated, the terms "pool bottom", "swimming pool bottom surface" and "swimming pool bottom" all represent the pool bottom surface of the swimming pool.
[0016] During the movement cleaning of the robot in the swimming pool, the cleaning target can be identified, and path planning, cleaning task execution and other operations can be performed. In the case of a real cleaning target (such as dirt such as leaves and plastic bags), the cleaning target will usually move with the water flow in the pool. For example, during the movement cleaning of the robot in the pool, the water pump of the robot will disturb the water flow, and then the water flow will drive the cleaning target around the robot to move, for example, under the influence of the water flow pushed forward by the robot, the leaves will also float forward as the robot moves forward. However, the robot may also mistakenly identify a non-real cleaning target (for example, a damaged area of the pool bottom, a floor lamp, a wall lamp, a step, etc.) as a real cleaning target, for example, mistakenly identifying a damaged area of the swimming pool as a leaf, and cleaning the damaged area. However, because the target cannot be cleaned, the robot will clean the non-real cleaning target every time it encounters it, which reduces the cleaning efficiency of the robot. The control method provided by the application can distinguish between real cleaning targets and non-real cleaning targets, thereby improving the cleaning efficiency.
[0017] According to a first aspect of the application, a control method of an automatic pool cleaning device is provided, and the control method 100 of the automatic pool cleaning device is described in detail below with reference to the accompanying drawings. Figure 1 is a flowchart of the control method of the automatic pool cleaning device provided by the application. As Figure 1 shown, steps 101 to 103 are described in detail below.
[0018] Step 101, during the movement of the pool automatic cleaning device in the pool, the image acquisition device is used to acquire the image of the object around the pool automatic cleaning device.
[0019] Exemplarily, the robot can include multiple cleaning modes when cleaning the pool, such as pool bottom cleaning mode, pool wall cleaning mode and water surface cleaning mode, and the embodiments of the present application can be applied to multiple cleaning modes. The robot is provided with an image acquisition device, and during the movement of the robot in the pool, the image acquisition device can be used to acquire the image of the object around the robot, for example, during the movement of the robot in the pool for cleaning, the image acquisition device can be used to acquire the image of the object in front of the robot, wherein the front can include the front, the left front, the right front and the front bottom, etc.
[0020] The image acquisition device of the robot can include at least one camera. The camera can adopt one or more of the following types according to actual needs: visible light camera (RGB camera), which is used to acquire color images under sufficient light conditions; low light camera (such as high photosensitive sensor camera), which is suitable for image acquisition in weak light environment; infrared camera (IR camera), which can work in turbid water or low visibility conditions combined with infrared fill light; underwater special camera, which has the characteristics of waterproof, pressure resistance and corrosion resistance, and is suitable for long-term underwater operation; 3D camera or depth camera (such as binocular vision, structured light or ToF camera), which is used to acquire three-dimensional spatial information of obstacles; multispectral or polarized camera, which is used to enhance the image recognition ability in specific scenes (such as stain classification or reflection suppression); panoramic camera, which provides a wider field of view to improve environmental perception ability. The above-mentioned cameras are only a limited enumeration, and other types of cameras can be used in practice as long as they can realize the technical concept of the present application. It should be noted that the camera used should have a sealing and waterproof function to ensure that it can work stably in water for a long time. The camera can be arranged at the front, upper, side and / or rear of the robot according to needs to optimize the image acquisition requirements in different directions.
[0021] Exemplarily, the image acquisition device is a monocular image acquisition device or a binocular image acquisition device. Of course, the image acquisition device can also be a surround view image acquisition device. Different image acquisition devices have different advantages, for example, the cost of monocular image acquisition device is low; binocular image acquisition device can obtain sufficient depth information; surround view image acquisition device can acquire first images in multiple directions without the need to control the robot or the image acquisition device to rotate, which can save the energy consumption of the robot.
[0022] It should be noted that the above description of the type of image acquisition device is only exemplary, and those skilled in the art can adjust the type of image acquisition device according to the actual situation, as long as the technical principles of the present application can be realized.
[0023] Step 102, after identifying the cleaning target based on the image, controlling the water tank automatic cleaning device to move to the cleaning target.
[0024] Specifically, after collecting the image of the object around the robot, target recognition can be performed through the collected image to identify whether there is a cleaning target around the robot. The cleaning target is generally garbage that needs to be cleaned, such as leaves, etc., but the cleaning target identified by the robot at this time may not be the real cleaning target, such as the pool wall damage may be misidentified as leaves. After identifying the cleaning target based on the image, the robot needs to be controlled to move to the cleaning target. For example, the relative coordinates of the cleaning target relative to the robot can be calculated according to the pixel position of the cleaning target in the image, and then a control command is generated based on the relative coordinates to control the movement of the robot (such as controlling the robot to turn, move forward, move backward, etc.), so that the robot moves to the cleaning target. The above description of step 102 is only exemplary, and those skilled in the art can select the identification method of the cleaning target, the movement control method of the robot, etc., as long as the technical principles of the present application can be realized.
[0025] Step 103, in the process of moving to the cleaning target, collecting multiple images with the cleaning target through the image acquisition device.
[0026] Specifically, in the process of controlling the robot to move to the cleaning target, further, multiple images with the cleaning target are collected through the image acquisition device, that is, the cleaning target is tracked and multiple images are collected in the process of moving to the cleaning target. Among them, the multiple images can be continuous multiple images or non-continuous multiple images. That is, the multiple images can be multiple images continuous in time sequence, or multiple images collected every interval of a predetermined time length, or multiple images randomly collected. For example, the robot identifies that there may be leaves in front of it while moving in the pool, and then needs to control the robot to move to the identified leaves, and collect multiple images in the process of moving, so as to further determine whether the identified leaves are real leaves based on the multiple images.
[0027] The above description of multiple images is only exemplary, and those skilled in the art can collect multiple images according to the actual situation, as long as the technical principles of the present application can be realized.
[0028] In step 104, it is determined whether the cleaning target is a real cleaning target based on the multiple images. If not, the water pool automatic cleaning device is controlled to stop cleaning the cleaning target when the water pool automatic cleaning device identifies the cleaning target again.
[0029] Specifically, the cleaning target is further determined based on the collected multiple images. If the cleaning target is a real cleaning target, the robot is controlled to clean the cleaning target. If the cleaning target is not a real cleaning target, the robot is controlled to stop cleaning the cleaning target subsequently. For example, the controller of the robot can be configured with a pre-trained image recognition model. The image recognition model is constructed by a neural network and pre-trained by a large number of image data sets labeled with different types of pool bottom dirt (including real cleaning targets such as branches, leaves, etc. and non-real cleaning targets such as steps, wall lamps, etc.). If the multiple images are input into the image recognition model, the recognition result output by the model can be obtained, so as to determine whether the cleaning target is a real cleaning target.
[0030] For example, the real cleaning target includes leaves, algae or plastic products. Of course, the above-mentioned real cleaning target is only an example, and the real cleaning target can also be other objects that need to be cleaned in the pool. The embodiments of the present application do not make specific limitations here.
[0031] The control method of the water pool automatic cleaning device provided in the present application can collect the images of the objects around the water pool automatic cleaning device by the image collection device during the movement of the water pool automatic cleaning device. After the cleaning target is identified based on the images, the water pool automatic cleaning device is controlled to move towards the cleaning target and collect multiple images during the movement. Then, the cleaning target is determined again based on the collected multiple images to determine whether the identified cleaning target is a real cleaning target. By determining the identified cleaning target again, the misidentification can be effectively reduced. When it is determined that the cleaning target is not a real cleaning target, the cleaning target is not cleaned again when the cleaning target is identified again subsequently. In this way, the water pool automatic cleaning device can avoid repeatedly attempting to clean the non-real cleaning target, and the cleaning efficiency of the water pool automatic cleaning device is effectively improved.
[0032] In step 104, determining whether the cleaning target is a real cleaning target based on the multiple images includes: determining the depth change of the cleaning target and / or the depth change of an environmental object in the environment where the cleaning target is located during the movement of the pool automatic cleaning device to the cleaning target based on the multiple images; and determining whether the cleaning target is the real cleaning target according to the depth change of the cleaning target and / or the depth change of the environmental object in the environment where the cleaning target is located.
[0033] As shown in Figure 2 , Figure 3 When the machine moves on the pool bottom or water surface, the image acquisition device on the robot can acquire images of objects around the machine. Since the shape of the damaged pool wall or pool bottom is similar to that of a leaf, the machine is likely to identify the damage as a real cleaning target such as a leaf. However, the real cleaning target is generally a light object such as a leaf or a plastic product. Therefore, when the robot moves towards the cleaning target, the forward movement of the robot will push the water in the pool to generate a wave, thereby pushing the leaf to move forward. The non-real cleaning target is generally fixed in the pool and is not or less affected by the water flow generated by the robot. Therefore, the depth changes of the real cleaning target and the non-real cleaning target in the multiple images acquired during the movement of the robot to the cleaning target are different. Therefore, the depth change of the cleaning target and / or the depth change of the environmental object in the environment where the cleaning target is located during the movement of the robot to the cleaning target can be determined based on the multiple images, and then whether the cleaning target is the real cleaning target can be determined according to the depth change of the cleaning target and / or the depth change of the environmental object in the environment where the cleaning target is located. For example, if the depth of the cleaning target identified before based on the multiple images does not decrease with the movement of the robot, it can be considered that the cleaning target is a real cleaning target. If the depth of the cleaning target decreases with the movement of the robot, it can be considered that the cleaning target is a non-real cleaning target.
[0034] Further, determining whether the cleaning target is the real cleaning target according to the depth change of the cleaning target and / or the depth change of the environmental object in the environment where the cleaning target is located includes: determining that the cleaning target is not the real cleaning target when the depth change of the cleaning target is greater than a preset depth change threshold; or determining that the cleaning target is not the real cleaning target when the difference between the depth change of the cleaning target and the depth change of the environmental object is less than a predetermined difference threshold.
[0035] For example, as described above, a real cleaning target such as a leaf will move forward together with the robot, and the change in the distance between the leaf and the robot is not obvious. If it is determined based on multiple images that the depth of the previously identified cleaning target changes significantly (greater than a preset depth change threshold) as the robot moves, and the speed of the robot is positively correlated, it can be considered that the cleaning target is not a real cleaning target, but a fixed object such as a pool wall damage that looks like a leaf.
[0036] On the other hand, the difference between the depth changes of the cleaning target and the environmental object in the environment (for example, the environmental object in the pool, such as a fixed object such as a ground lamp, a step, etc.) can be determined. If the difference is large, it indicates that the cleaning target moves due to the fluctuation of the water flow during the movement of the robot, and it is further determined that the cleaning target is a movable object, i.e., a real cleaning target; if the difference is small, it indicates that the cleaning target does not move or moves less with the fluctuation of the water flow during the movement of the robot, and it is further determined that the cleaning target is not movable, i.e., the cleaning target is not a real cleaning target. Based on the above principle, a predetermined difference threshold can be set, and when the difference between the depth changes of the cleaning target and the environmental object is less than the predetermined difference threshold, it can be determined that the cleaning target is not a real cleaning target.
[0037] In step 104, when the pool automatic cleaning device identifies the cleaning target again, the pool automatic cleaning device is controlled to stop cleaning the cleaning target, including: when the cleaning target appears again in the sensing range of the image acquisition device, the pool automatic cleaning device is controlled to avoid the cleaning target or the pool automatic cleaning device is controlled to stop moving towards the cleaning target.
[0038] Specifically, in the case where it is determined that the cleaning target is not a real cleaning target (for example, a damaged area of a pool wall), if the cleaning target is repeatedly cleaned, the cleaning efficiency of the robot will be seriously affected. Therefore, in the case where it is determined that the cleaning target is not a real cleaning target, and when it is found that the cleaning target appears again in the sensing range of the image acquisition device, the robot can be controlled to directly avoid the cleaning target or the robot can be controlled to stop moving towards the cleaning target, so as to improve the cleaning efficiency of the robot.
[0039] Further, the control method further includes: when it is determined based on the multiple images that the cleaning target is not the real cleaning target, recording object information of the cleaning target, so that the pool automatic cleaning device determines whether the cleaning target is identified again during subsequent movement based on the object information; wherein the object information includes at least one of image information, position information and contour information of the cleaning target.
[0040] Specifically, in the case of judging that the cleaning target is a non-real cleaning target, it is necessary to record the object information of the cleaning target, which may include at least one of image information, position information and contour information of the cleaning target. When the machine moves in the pool, if the cleaning target appears around the machine, the recorded object information can be used to determine whether the cleaning target has been identified before and does not belong to the real cleaning target, such as comparing the current collected image with the previously recorded image, or comparing the current position of the machine with the recorded position, or comparing the contour of the current collected object with the recorded contour information, and the machine can be controlled not to clean it, such as not needing the machine to clean it in a targeted manner, or even can control the machine to bypass the position. Of course, the above information is only exemplary, and the object information can also include other information that can represent the cleaning target, which is not limited in the present application. Those skilled in the art can set the object information according to the actual situation, as long as the technical principles of the present application can be realized.
[0041] Further, the method further comprises: before the pool automatic cleaning device moves to the cleaning target, if it is judged based on the plurality of images that the cleaning target is not the real cleaning target, controlling the pool automatic cleaning device to stop moving to the cleaning target. That is, in the case of finding that the cleaning target is a non-real cleaning target before the robot moves to the cleaning target, it is not necessary to move the cleaning to the cleaning target, so the robot can be controlled to stop moving to the cleaning target.
[0042] According to the second aspect of the present application, a pool automatic cleaning device is also provided. The pool automatic cleaning device can perform the control method described above with reference to various embodiments. The description of the pool automatic cleaning device performing the control method of various embodiments is omitted here, and the principles and schemes of the control method are described above in combination with various embodiments and the accompanying drawings. Here, no longer be described.
[0043] According to a third aspect of the present application, a non-transitory computer readable storage medium is also provided, which stores a computer program. When the computer program is executed by a processor, the computer program implements the control method of the pool automatic cleaning device provided by the above embodiments. The pool automatic cleaning device comprises an image acquisition device. The control method comprises: acquiring, by the image acquisition device, images of objects around the pool automatic cleaning device during movement of the pool automatic cleaning device in the pool; after identifying a cleaning target based on the images, controlling the pool automatic cleaning device to move to the cleaning target; during movement to the cleaning target, acquiring, by the image acquisition device, multiple images with the cleaning target; determining whether the cleaning target is a real cleaning target based on the multiple images, and if not, controlling the pool automatic cleaning device not to clean the cleaning target again when the pool automatic cleaning device identifies the cleaning target again. The principle and scheme of the control method are described above in combination with the embodiments and the drawings, and will not be repeated here.
[0044] According to a fourth aspect of the present application, a computer program product is also provided. The computer program product comprises a computer program, which can be stored on a non-transitory computer readable storage medium. When the computer program is executed by a processor, the computer can execute the control method of the pool automatic cleaning device provided by the above methods. The pool automatic cleaning device comprises an image acquisition device. The control method comprises: acquiring, by the image acquisition device, images of objects around the pool automatic cleaning device during movement of the pool automatic cleaning device in the pool; after identifying a cleaning target based on the images, controlling the pool automatic cleaning device to move to the cleaning target; during movement to the cleaning target, acquiring, by the image acquisition device, multiple images with the cleaning target; determining whether the cleaning target is a real cleaning target based on the multiple images, and if not, controlling the pool automatic cleaning device not to clean the cleaning target again when the pool automatic cleaning device identifies the cleaning target again. The principle and scheme of the control method are described above in combination with the embodiments and the drawings, and will not be repeated here.
[0045] The device embodiments described above are only schematic, and the units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e. they can be located in one place, or distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the present embodiment according to actual needs. Those skilled in the art can understand and implement it without creative labor.
[0046] Those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary universal hardware platform, and of course can be implemented by hardware. Based on such an understanding, the technical solutions described above can be embodied in the form of a software product, and the computer software product can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, or an optical disk, and includes a plurality of instructions used to cause a computer device (such as a personal computer, a server, or a network device) to execute the methods described in each embodiment or some parts of the embodiments.
[0047] In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, different embodiments or examples described in the present specification and the features of different embodiments or examples can be combined and combined by those skilled in the art without contradiction.
[0048] In addition, the terms "first", "second" are only for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include at least one of the features. In the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise specifically limited.
[0049] The above is only an exemplary embodiment of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of various changes or replacements within the technical scope disclosed in the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.
Claims
1. A method for controlling an automatic pool cleaning device, wherein the automatic pool cleaning device includes an image acquisition device, and the control method comprises: During the movement of the automatic pool cleaning device in the pool, the image capturing device captures images of objects around the automatic pool cleaning device; After identifying a cleaning target based on the image, controlling the automatic pool cleaning device to move to the cleaning target; In the process of moving toward the cleaning target, the image acquisition device acquires multiple frames of images with the cleaning target; Based on the multiple frames of images, it is determined whether the cleaning target is a real cleaning target. If not, when the automatic pool cleaning device recognizes the cleaning target again, the automatic pool cleaning device is controlled to no longer clean the cleaning target.
2. The control method according to claim 1, wherein: The method further comprises: When it is determined based on the multiple frames of images that the cleaning target is not the real cleaning target, the object information of the cleaning target is recorded so that the automatic pool cleaning device can determine whether to recognize the cleaning target again based on the object information during subsequent movement; wherein, the object information includes: at least one of the image information, position information and contour information of the cleaning target.
3. The control method according to claim 1, wherein: When the automatic pool cleaning device recognizes the cleaning target again, controlling the automatic pool cleaning device to no longer clean the cleaning target includes: When the cleaning target appears again within the sensing range of the image acquisition device, the automatic pool cleaning device is controlled to avoid the cleaning target or the automatic pool cleaning device is controlled not to move toward the cleaning target.
4. The control method according to claim 1, wherein: The actual cleaning targets include: leaves, algae or plastic products.
5. The control method according to claim 1, wherein: Determining whether the cleaning target is a real cleaning target based on the multiple frames of images includes: Determining, based on the multiple frames of images, a change in the depth of the cleaning target and / or a change in the depth of an environmental object in the environment where the cleaning target is located during the process of the automatic pool cleaning device moving to the cleaning target; Whether the cleaning target is the real cleaning target is determined based on the depth change of the cleaning target and / or the depth change of the environmental objects in the environment where the cleaning target is located.
6. The control method according to claim 5, wherein: Determining whether the cleaning target is the real cleaning target according to a change in the depth of the cleaning target and / or a change in the depth of an environmental object in the environment where the cleaning target is located includes: If the depth change of the cleaning target is greater than a preset depth change threshold, determining that the cleaning target is not a real cleaning target; or When the difference between the depth change of the cleaning target and the depth change of the environmental object is less than a predetermined difference threshold, it is determined that the cleaning target is not the real cleaning target.
7. The control method according to claim 1, wherein: The multiple frames of images are continuous multiple frames of images or discontinuous multiple frames of images.
8. The control method according to claim 1, wherein: The method further comprises: Before the automatic pool cleaning device moves to the cleaning target, if it is determined based on the multiple frames of images that the cleaning target is not the real cleaning target, the automatic pool cleaning device is controlled to stop moving toward the cleaning target.
9. The control method according to any one of claims 1 to 8, wherein: The image acquisition device is a monocular image acquisition device or a binocular image acquisition device.
10. An automatic pool cleaning device, wherein: The automatic pool cleaning device can execute the control method according to any one of claims 1 to 9.
11. A computer storage medium storing a computer program, wherein the computer program is executed by a processor to implement the control method according to any one of claims 1 to 9.