Automatic pool cleaning equipment and control method thereof

Abstract

Disclosed are an automatic pool cleaning apparatus and a method of controlling the same. The method comprises the following steps: acquiring detection data of a detection object in the advancing direction of the automatic pool cleaning equipment on the water surface, the detection data comprises intensity information of a reflection signal which is captured by a distance sensor of the automatic pool cleaning equipment and comes from the detection object and distance information between the detection object and the automatic pool cleaning equipment; determining a category of the detection object based on the intensity information and the distance information; and adjusting a moving path of the automatic pool cleaning equipment based on the type of the detection object.

Description

Technical Field

[0001] The present disclosure relates to an automatic pool cleaning device and a method for controlling the automatic pool cleaning device in the field of automatic cleaning. Background Art

[0002] For pool facilities such as swimming pools, automatic pool cleaning equipment can be utilized to perform automatic cleaning or auxiliary cleaning. For example, the automatic pool cleaning equipment can be designed to filter the pool water and absorb dirt while moving on the bottom, wall and / or surface of the pool. Summary of the Invention

[0003] Disclosed is a method for controlling an automatic pool cleaning device, comprising: obtaining detection data about a detection object in a direction of travel of the automatic pool cleaning device on a water surface, the detection data comprising: intensity information of a reflection signal of the detection object detected by a distance sensor and distance information between the detection object and the automatic pool cleaning device; determining a category of the detection object based on the intensity information and the distance information; and adjusting a moving path of the automatic pool cleaning device based on the category of the detection object.

[0004] In one or more embodiments, determining the category of the detected object based on the intensity information and the distance information includes: determining at least one data classification of the detection data based on the intensity information and the distance information; and determining the category of the detected object based on the data classification to which the data point corresponding to the detected object in the detection data belongs.

[0005] In one or more embodiments, determining at least one data classification of the detection data based on the intensity information and the distance information includes determining a classification hyperplane for the detection data. Determining the category of the detected object based on the data classification to which data points corresponding to the detected object in the detection data belong includes predicting the category of the detected object based on the classification hyperplane and the data points corresponding to the detected object in the detection data.

[0006] In one or more embodiments, determining a classification hyperplane regarding the detection data includes: determining the classification hyperplane based on historical detection data regarding a plurality of historical detection objects, the historical detection data including historical intensity information of reflected signals from the plurality of historical detection objects captured by the distance sensor and historical distance information between the plurality of historical detection objects and the automatic pool cleaning device.

[0007] In one or more embodiments, determining a classification hyperplane for the detection data includes: determining at least one data classification of the detection data by a nonlinear support vector machine; and determining the classification hyperplane by fitting based on the at least one data classification.

[0008] In one or more embodiments, the category of the detected object includes at least one of a fixed object and a floating object.

[0009] In one or more embodiments, adjusting the moving path of the automatic pool cleaning device based on the category of the detected object includes: when the category of the detected object is determined to be a fixed object, controlling the automatic pool cleaning device to turn and move to another path when the distance between the detected object and the detected object reaches a specified distance.

[0010] In one or more embodiments, adjusting the movement path of the automatic pool cleaning device based on the category of the detected object includes: controlling the automatic pool cleaning device to move toward or avoid the detected object when the category of the detected object is determined to be a floating object.

[0011] Also disclosed is an automatic pool cleaning device, comprising: a distance sensor configured to detect the distance between the automatic pool cleaning device and surrounding objects; and a controller configured to execute the method described above.

[0012] In one or more embodiments, the distance sensor includes at least one of an ultrasonic sensor, a lidar, an infrared sensor, a time-of-flight sensor, and a visual sensor. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 An example of an automatic pool cleaning device according to an embodiment of the present disclosure is schematically shown.

[0014] Figure 2 An example of a method for controlling an automatic pool cleaning device according to an embodiment of the present disclosure is schematically shown.

[0015] Figure 3 An example of detection data according to an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0016] The embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. In the drawings, the same or corresponding parts are given the same reference numerals, and their description will not be repeated.

[0017] Figure 1 An exemplary automatic pool cleaning device 100 (hereinafter also referred to as “device 100 ”) in an embodiment of the present disclosure is illustratively shown.

[0018] The device 100 may be configured with a housing, and a travel mechanism such as travel wheels, tracks, water nozzles, propellers, etc. The housing may be provided with a water inlet and outlet, etc., and a suction device, a filtering device, and a drive mechanism, etc., may be provided within the housing. The suction device may, for example, include at least one water pump. The filtering device may, for example, include a trash basket having at least one layer of filter mesh. The drive mechanism may, for example, include components such as a motor, a water pump, and gears that can provide and / or transmit driving force to drive the travel mechanism to operate, thereby driving the device 100 to move or swim in the water, on the water surface, and / or on the pool wall.

[0019] For example, when the device 100 moves on the bottom, wall or water surface of the pool, the suction device of the device 100 can work to suck the water in the pool together with the garbage or dirt in the water from the water inlet of the device 100 into the filtering device of the device 100, and then suck the water in the filtering device out of the filtering device and guide it to the water outlet of the device 100, and finally discharge it into the pool from the water outlet, while the garbage or dirt in the pool water is adsorbed by the filter element of the filtering device or intercepted in the filtering device, thereby cleaning the pool.

[0020] like Figure 1 As shown, the device 100 may also be configured with a controller 110. The controller 110 may include any circuit and / or module with data processing capability and / or instruction execution capability and suitable for the device 100, such as a central processing unit (CPU), a graphics processing unit (GPU), a field programmable gate array (FPGA), etc., and may be configured to perform data processing and / or control related to the cleaning operation and / or other functions of the device 100 according to a program stored in a memory (not shown) of the device 100 and / or a signal and / or instruction from a control panel or a control terminal (not shown) of the device 100 and / or sensing data from one or more sensors (e.g., a spatial attitude sensor, an odometer, etc.) of the device 100.

[0021] In addition, the device 100 may also include at least one distance sensor 120 such as an ultrasonic sensor, a lidar, a visual sensor, an infrared sensor, a time-of-flight (ToF) sensor, etc., so as to detect the distance between the device 100 and surrounding objects (for example, cleanable objects in the pool such as garbage, uncleanable objects or fixed obstacles such as pool walls, escalators, etc., etc.) when the device 100 moves on the water surface or in the pool.

[0022] In a pool or on the water surface, there may be various types of floating objects, such as leaves, lifebuoys, water toys, and fixed objects such as pool walls and ladders. For example, when the device 100 performs a cleaning task on the water surface, the detection data of the distance sensor 120 may come from different objects. If the detection data of the distance sensor 120 itself does not contain semantic information, it may not be possible to accurately determine whether the detected object is an uncleanable obstacle such as the pool wall or a cleanable floating object such as a leaf, thereby affecting the environmental perception and path planning capabilities of the device 100.

[0023] Figure 2 An exemplary method 200 for controlling the device 100 according to an embodiment of the present disclosure is schematically shown. The method 200 may be implemented by the controller 110 by executing corresponding program instructions, for example, and may improve the environmental perception capability and path planning capability of the device 100.

[0024] like Figure 2 As shown, method 200 may include steps 210 , 220 , and 230 .

[0025] For example, when the device 100 moves on the water surface, the controller 110 can execute step 210 to obtain detection data about the detection object in the direction of travel of the device 100 on the water surface, wherein the acquired detection data may include intensity information of the reflected signal from the detection object captured by the distance sensor 120 of the device 100 and distance information between the detection object and the device 100.

[0026] For example, in the case where the distance sensor 120 includes an ultrasonic sensor, when the device 100 moves on the water surface, the distance sensor 120 may transmit ultrasonic signals toward the surroundings or a designated direction and receive reflected signals.

[0027] The distance sensor 120 or the controller 110 can then determine the real-time distance between the device 100 and a detected object, for example, in the direction of travel of the device 100 on the water surface based on the time difference between the transmitted ultrasonic signal and the received reflected signal, the propagation speed of the ultrasonic wave in the air and / or water surface and / or water, etc.

[0028] In some embodiments, the device 100 may also include a temperature sensor to compensate for the propagation speed of the ultrasonic wave according to the monitored temperature value, and then determine the real-time distance between, for example, the device 100 and the detected object in the direction of travel of the device 100 on the water surface based on the time difference between the transmitted ultrasonic signal and the received reflected signal, the compensated ultrasonic propagation speed, etc.

[0029] For example, when the distance sensor 120 includes a lidar, the controller 110 can determine, for example, the real-time distance between the device 100 and a detected object in the direction of water travel of the device 100 based on point cloud data about the surrounding environment of the device 100 obtained by the lidar.

[0030] For example, in the case where the distance sensor 120 includes a visual sensor such as a monocular camera or a binocular camera, the controller 110 can identify the detection object in the water surface travel direction of the device 100 based on the real-time image of the surrounding environment of the device 100 obtained through the visual sensor, through any suitable image analysis and processing method or model such as an image convolutional neural network or a target recognition method or model, and determine the real-time distance between the device 100 and the identified detection object.

[0031] Affected by factors such as water surface fluctuations, the position of the device 100 and / or the detection object of the floating object type on the water surface may not be fixed on the water surface. For example, the distance sensor 120 may sometimes be able to sense the detection object, but sometimes not sense the detection object, making the distance detection value between the device 100 and the detection object or the change in the distance detection value unstable. For example, the change amplitude of the distance data of adjacent frames may be different. On the other hand, the detection object of the fixed object type such as the pool wall will not float or drift on the water surface. As the device 100 moves, the change in the distance between the device 100 and the detection object of the fixed object type such as the pool wall is relatively stable, for example, it may show a trend of gradually decreasing.

[0032] Furthermore, different detection objects may have different materials or shapes. Therefore, for ultrasonic, infrared, or laser signals, the reflection coefficients of the transmission signals from the distance sensor 120 may be different. This means that even if the detection objects are at the same distance from the device 100, the reflected signals from objects of different materials or shapes may have different signal strengths, penetration depths, and / or detection sensitivities. In swimming pools, fixed objects (e.g., pool walls) and floating objects (e.g., leaves, toys) are typically made of different materials.

[0033] It can be seen that the distance between the detection object and the device 100 and the signal strength of the reflected signal of the detection object can reflect the type of the detection object to a certain extent, especially can identify whether the detection object is a fixed object or a floating object.

[0034] like Figure 3As shown, in the detection data about the detection object in the water surface travel direction of the device 100 obtained in step 210, in addition to the distance information about the real-time distance between the device 100 and the detection object, for example, when the distance between the device 100 and the detection object is determined based on the transmission signal of the distance sensor 120 such as an ultrasonic sensor, an infrared sensor or a lidar and the captured reflection signal, it can also include, for example, the intensity information of the reflection signal from the detection object received by the distance sensor 120.

[0035] Then, the controller 110 may execute step 220 to determine the category of the detected object based on the intensity information and distance information acquired in step 210 .

[0036] For example, in step 220, at least one data classification of the detection data acquired in step 210 may be determined based on the intensity information and the distance information. For example, a suitable method or model, such as a nonlinear support vector machine or a deep learning network, may be used to determine the at least one data classification of the detection data based on the detection data including the intensity information and the distance information.

[0037] Each data classification may correspond to a category of detection objects, for example, including but not limited to: uncleanable fixed objects or fixed obstacles such as pool walls and ladders; uncleanable floating objects such as water toys and inflatable lifebuoys; cleanable floating objects such as fallen leaves floating on the water surface; and so on.

[0038] Then, for the current detection object, it can be determined to which data category of the at least one determined data category the data point corresponding to the detection object in the detection data belongs, thereby determining or predicting the category of the detection object.

[0039] For example, a suitable method or model, such as a nonlinear support vector machine or a deep learning network, can be used to determine at least one data classification of the detection data based on the detection data including intensity information and distance information. Then, based on the determined at least one data classification, a classification hyperplane for the detection data is determined through data fitting. The category of the current detection object can then be predicted based on the classification hyperplane and data points in the detection data corresponding to the current detection object.

[0040] For example, when the detection data is data in a two-dimensional space, the determined classification hyperplane can be a straight line in the two-dimensional space (plane), and the two sides of the straight line correspond to a data classification. When the detection data is data in a three-dimensional space, the determined classification hyperplane can be a plane in the three-dimensional space, and the two sides of the plane correspond to a data classification. Similarly, when the detection data is data in an n-dimensional space, the determined classification hyperplane can be an n-1 dimensional geometric object in the n-dimensional space, and the two sides of the n-1 dimensional geometric object correspond to a data classification. There can be one or more classification hyperplanes corresponding to the detection data, and each classification hyperplane can correspond to a type of detection object. Based on the classification hyperplane, it can be identified whether the detection object is a floating object or a fixed object, and it can even be identified which specific floating object or fixed object it is.

[0041] For example, see Figure 3 A classification hyperplane can be determined based on the intensity information and distance information of data points or data items corresponding to historical sampling time points or historical frames t0, t1, t2, t3, etc. in the historical detection data of multiple historical detection objects. Then, for the current sampling time point or current frame tc, based on the data points at tc corresponding to the current detection object in the detection data and / or one or more data points at one or more previous sampling time points or previous frames before tc in the detection data corresponding to the current detection object, a suitable method or model such as a nonlinear support vector machine or a deep learning network can be used to determine which side of the classification hyperplane the data points corresponding to the current detection object are located, thereby determining or predicting the category of the current detection object.

[0042] In another embodiment, methods or models such as nonlinear support vector machines and deep learning networks can be trained using historical ultrasonic data collected by the device 100 during previous actual water surface cleaning tasks, or real-value data pre-collected or designed during the factory commissioning phase of the device 100. For example, hyperparameters of the nonlinear support vector machine can be determined or screened using methods such as grid search. Then, for the current sampling time point or current frame tc, the trained method or model can be used to determine which side of the classification hyperplane the data point corresponding to the current detected object is located on based on the data point at tc corresponding to the current detected object in the detection data, and / or one or more data points at one or more previous sampling time points or previous frames before tc corresponding to the current detected object in the detection data, thereby determining or predicting the category of the current detected object.

[0043] After determining the category of the detected object, the controller 110 may perform step 230 to adjust a movement path of the device 100 , for example, on a water surface, based on the category of the detected object determined in step 220 .

[0044] For example, in step 230, when the detection object is determined to be a fixed object such as a pool wall, a ladder, etc., the controller 110 can control the travel mechanism of the control device 100 through an instruction signal, and drive the device 100 to turn and move to another moving path when the distance between the device 100 and the detection object reaches a specified distance.

[0045] For example, in step 230, when the detection object is determined to be a floating object such as a fallen leaf, the controller 110 can control the travel mechanism of the control device 100 through an instruction signal to drive the device 100 to move toward the detection object or avoid the object.

[0046] For example, when the size of the detection object exceeds a predetermined threshold, the controller 110 can control the device 100 to avoid the detection object through an instruction signal; otherwise, the controller 110 can control the device 100 to move toward the detection object in order to clear the detection object.

[0047] For example, when the detection object is determined to be a cleanable floating object such as fallen leaves, the controller 110 can control the device 100 through a command signal to determine whether to move toward the detection object in order to clean the detection object; when the detection object is determined to be an uncleanable floating object such as a water toy, the controller 110 can control the device 100 through a command signal to avoid the detection object.

[0048] As described above, in method 200, the category of the detected object is determined based on the distance information and intensity information in the detection data, and the movement path of device 100 is adjusted based on the determined category of the detected object, thereby improving the environmental perception capability and navigation planning accuracy of device 100. In addition, method 200 is applicable to swimming pools of various material types and shapes, and is not limited to specific environments, thereby improving the adaptability of device 100.

[0049] The basic principles of the present disclosure have been described above in conjunction with the embodiments. However, it should be noted that the advantages, strengths, and effects mentioned in this disclosure are merely illustrative and non-restrictive, and should not be construed as necessarily possessed by each embodiment of the present disclosure. Furthermore, the aforementioned details are provided for illustrative purposes and to facilitate understanding, not for limitation, and do not limit the present disclosure to necessarily being implemented using the aforementioned details.

[0050] The block diagrams of the devices, apparatuses, equipment, and systems involved in this disclosure are intended to be illustrative examples only and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. In different embodiments, these devices, apparatuses, equipment, and systems may be connected, arranged, or configured in any appropriate manner.

[0051] In addition, words such as "including," "comprising," and "having" are open-ended words that mean "including but not limited to," and are used interchangeably therewith. The words "or" and "and" used herein mean the words "and / or" and are used interchangeably therewith unless the context clearly indicates otherwise. The word "such as" used herein means the phrase "such as, but not limited to," and is used interchangeably therewith.

[0052] It should also be noted that in the apparatus, device, and method of the present disclosure, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present disclosure.

[0053] In this document, modifiers such as "first" and "second" without quantifiers are intended to distinguish different elements / components / circuits / modules / devices / steps, and are not used to emphasize the order, positional relationship, importance, priority, etc. In contrast, modifiers such as "first" and "second" with quantifiers can be used to emphasize the order, positional relationship, importance, priority, etc. of different elements / components / circuits / modules / devices / steps.

[0054] The above description is provided for the purpose of illustration and description. This description is not intended to limit the embodiments of the present disclosure to the forms disclosed herein. Although a number of example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions, and sub-combinations thereof.

Claims

1. A method for controlling an automatic pool cleaning device, comprising: Acquiring detection data about a detection object in the direction of travel of the automatic pool cleaning device on the water surface, the detection data including: intensity information of a reflection signal of the detection object detected by a distance sensor and distance information between the detection object and the automatic pool cleaning device; determining a category of the detected object based on the intensity information and the distance information; and The moving path of the automatic pool cleaning device is adjusted based on the category of the detected object.

2. The method according to claim 1, wherein Determining the category of the detected object based on the intensity information and the distance information includes: determining at least one data classification of the detection data based on the intensity information and the distance information; and The category of the detected object is determined based on the data category to which the data point corresponding to the detected object in the detection data belongs.

3. The method according to claim 2, wherein: Determining at least one data classification of the detection data based on the intensity information and the distance information includes: determining a classification hyperplane with respect to the detection data; and Determining the category of the detected object based on the data category to which the data points corresponding to the detected object in the detection data belong includes: predicting the category of the detected object based on the classification hyperplane and the data points corresponding to the detected object in the detection data.

4. The method according to claim 3, wherein: Determining a classification hyperplane on the detection data includes: The classification hyperplane is determined based on historical detection data about a plurality of historical detection objects, wherein the historical detection data includes historical intensity information of reflection signals from the plurality of historical detection objects captured by the distance sensor and historical distance information between the plurality of historical detection objects and the automatic pool cleaning device.

5. The method according to claim 3, wherein: Determining a classification hyperplane on the detection data includes: determining at least one data classification of the detection data by a nonlinear support vector machine; and The classification hyperplane is determined by fitting based on the at least one data classification.

6. The method according to any one of claims 1 to 5, wherein The category of the detected object includes at least one of a fixed object and a floating object.

7. The method according to claim 6, wherein: Adjusting the moving path of the automatic pool cleaning device based on the category of the detected object includes: In a case where the category of the detected object is determined to be a fixed object, the automatic pool cleaning device is controlled to turn and move to another path when the distance between the automatic pool cleaning device and the detected object reaches a specified distance.

8. The method of claim 6, wherein: Adjusting the moving path of the automatic pool cleaning device based on the category of the detected object includes: In a case where the category of the detected object is determined to be a floating object, the automatic pool cleaning device is controlled to move toward the detected object or avoid the detected object.

9. An automatic pool cleaning device comprising: a distance sensor configured to detect the distance between the automatic pool cleaning device and surrounding objects; as well as A controller configured to execute the method according to any one of claims 1 to 8.

10. The automatic pool cleaning device according to claim 9, wherein: The distance sensor includes at least one of an ultrasonic sensor, a laser radar, an infrared sensor, a time-of-flight sensor, and a visual sensor.

Images