Control method of swimming pool robot
By equipping the pool robot with image acquisition devices and attitude sensors, and combining attitude and image detection, the abnormal state of the pool robot can be detected and processed efficiently and accurately, ensuring efficient cleaning operations and solving the problem of pool water pollution.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- XINGMAI INNOVATION TECH (SUZHOU) CO LTD
- Filing Date
- 2026-01-05
- Publication Date
- 2026-04-21
AI Technical Summary
After use, pools may accumulate dirt and debris, leading to water pollution. Existing technologies struggle to efficiently and accurately detect and handle abnormal conditions in pool robots.
The swimming pool robot is equipped with image acquisition devices and posture sensors. It can initially detect suspected abnormal conditions by using posture information, confirm abnormal conditions by using image acquisition devices, and combine depth map analysis with the pool bottom and pool wall planes to execute abnormal handling and cleaning strategies.
This improves the accuracy and efficiency of anomaly detection in pool robots, ensuring efficient cleaning operations and reducing misjudgments and omissions.
Smart Images
Figure CN121900496A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of pool robot technology, and more particularly to a control method for a pool robot. Background Technology
[0002] After use, swimming pools may accumulate various dirt and debris, polluting the water and affecting their appearance, thus requiring cleaning. With the development of artificial intelligence technology, pool robots are increasingly used in pool cleaning. Therefore, anomaly detection of pool robots during their application is particularly important. Summary of the Invention
[0003] In view of this, embodiments of this application provide a control method for a swimming pool robot to ensure the accuracy and efficiency of anomaly detection in the swimming pool robot.
[0004] A first aspect of this application provides a control method for a swimming pool robot. The swimming pool robot is equipped with at least one image acquisition device and one attitude sensor. The image acquisition device is used to acquire images of the swimming pool, and the attitude sensor is used to acquire attitude information of the swimming pool robot. The method includes: If the swimming pool robot is detected to be in a suspected abnormal state through posture information, the image acquisition device is activated to acquire images, or images are obtained from an already activated image acquisition device. The image acquired by the image acquisition device is used to confirm whether the swimming pool robot is in an abnormal state.
[0005] In some embodiments, determining whether a pool robot is in an abnormal status based on images acquired by an image acquisition device includes: If the difference between images collected within a preset duration is less than a preset difference condition, the pool robot is determined to be in an abnormal confirmation state; or... Depth maps are obtained from images acquired by image acquisition devices. If the difference between depth maps is less than a preset difference condition within a continuously preset time period, the pool robot is determined to be in an abnormal confirmation state.
[0006] In some embodiments, the method further includes: During the operation of the pool robot, depth maps are obtained from images acquired by image acquisition devices; based on the depth maps, the current plane where the pool robot is located and the adjacent planes of the current plane are obtained; if the current plane is the bottom plane of the pool, the adjacent planes are determined to be the pool wall planes.
[0007] In some embodiments, defining adjacent planes as pool wall planes includes: Obtain the angle between the normal vector direction of the adjacent plane and the normal vector direction of the current plane; If the included angle is greater than the first preset angle threshold, then the adjacent plane is determined as the pool wall plane.
[0008] In some embodiments, after determining the adjacent plane as the pool wall plane, the method further includes: Control the pool robot to move along the boundary of the pool wall plane; or, Control the pool robot to move to the pool wall plane.
[0009] In some embodiments, the method further includes: During the operation of the pool robot, depth maps are obtained from images acquired by image acquisition devices, and the target normal vector of the transition area between the pool bottom and the pool wall is obtained based on the depth maps. If there are two or more target normal vectors with different directions, and at least one of the following conditions is met, then the shape of the transition region is determined to be non-right-angled: The maximum angle difference between the target normal vectors is greater than the second preset angle threshold; The number of target normal vectors is greater than a preset threshold.
[0010] In some embodiments, the method further includes: During the operation of the pool robot, a depth map is obtained based on the images acquired by the image acquisition device, and an obstacle map is constructed based on the continuously acquired depth map; if the pool robot is in an abnormal confirmation state, an abnormal handling operation is performed based on the obstacle map.
[0011] In some embodiments, after obtaining the current plane where the pool robot is located and the adjacent planes of the current plane based on the depth map, the method further includes: Based on the pre-defined correspondence between the pool bottom, pool walls and anomaly handling strategies, the target anomaly handling strategy corresponding to the current plane is determined; and anomaly handling operations are executed according to the target anomaly handling strategy corresponding to the current plane.
[0012] In some embodiments, after performing exception handling operations through a target exception handling strategy corresponding to the current plane, the method further includes: If the pool robot is still detected to be in an abnormal confirmation state, an alarm message will be issued and the trust value of the target abnormal handling strategy will be reduced. Among them, the initial trust value of each abnormal handling strategy is the highest, and the corresponding relationship will be terminated when the trust value is reduced to a preset value. If the pool robot is detected to have returned to normal operation from an abnormal confirmation state, the trust value of the target abnormal handling strategy is increased. If the increased trust value is higher than the initial trust value, the increased trust value is updated to the initial trust value.
[0013] In some embodiments, after obtaining the current plane where the pool robot is located and the adjacent planes of the current plane based on the depth map, the method further includes: Based on the pre-defined correspondence between the pool bottom, pool walls and cleaning strategies, the cleaning strategy corresponding to the current plane is determined; the cleaning operation is then performed using the cleaning strategy corresponding to the current plane.
[0014] The beneficial effects of this application embodiment compared with the prior art are as follows: The swimming pool robot is equipped with at least one image acquisition device and one attitude sensor. The image acquisition device is used to acquire images of the pool, and the attitude sensor is used to acquire attitude information of the swimming pool robot. When the swimming pool robot is detected to be in a suspected abnormal state through attitude information, the image acquisition device can be activated to acquire images or images can be acquired from the activated image acquisition device. The images can then be used to confirm whether the swimming pool robot is in an abnormal state. This achieves the goal of confirming whether the swimming pool robot is in an abnormal state through both attitude information and images, thereby improving detection accuracy while ensuring detection efficiency. Attached Figure Description
[0015] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0016] Figure 1 This is a flowchart illustrating a control method for a swimming pool robot provided in an embodiment of this application; Figure 2 This is a schematic diagram illustrating the relationship between the normal vectors of a plane provided in an embodiment of this application; Figure 3 This is a schematic diagram of the swimming pool robot moving along the plane of the pool wall provided in the embodiments of this application; Figure 4 This is a schematic diagram of the swimming pool robot moving to the pool wall plane according to an embodiment of this application; Figure 5 This is a schematic diagram of the control device for the swimming pool robot provided in an embodiment of this application. Detailed Implementation
[0017] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.
[0018] The following describes in detail, with reference to the accompanying drawings, a control method for a swimming pool robot according to an embodiment of this application.
[0019] This application provides a cleaning device for performing cleaning, disinfection, and rescue tasks in a target area. The target area can be a water-containing area where the cleaning device can move. For example, the target area may include, but is not limited to, swimming pools, water tanks, oil wells, sewers, etc. The shape of the target area can be a regular shape such as rectangle, circle, or ellipse, or an irregular shape, and is not limited thereto. The following description uses a swimming pool robot as the cleaning device and a swimming pool as the target area. For a swimming pool, the pool includes at least a pool bottom and pool walls.
[0020] Pool robots can clean only the pool bottom and walls. Alternatively, they can clean the pool bottom, walls, and surface. A pool robot can be entirely submerged, with its bottom in contact with the pool bottom, moving along the bottom (bottom-movement posture). Or, it can be entirely submerged, but its bottom doesn't contact the bottom or walls, suspending in the water (suspended posture). A pool robot can also be partially above and partially below the water surface, with its bottom not in contact with the walls, moving on the surface (surface-movement posture). Alternatively, its bottom may contact the walls, moving along them (wall-movement posture).
[0021] For pool robots, they can be robots that power their power-consuming units via built-in rechargeable batteries, or devices that power their power-consuming units via external cables.
[0022] The pool robot is equipped with at least one image acquisition device and one attitude sensor. The image acquisition device is used to acquire images of the pool, and the attitude sensor is used to acquire attitude information of the pool robot.
[0023] In some embodiments, the pool robot is equipped with at least one image acquisition device, which is used to acquire images of the pool area and may also acquire images of objects on the shore.
[0024] There may be one or more image acquisition devices. In some embodiments, when the pool robot moves on the bottom of the pool, the image acquisition device is located below the water surface to acquire images of the pool below the water surface; and / or, when the pool robot moves on the water surface, the image acquisition device is located above the water surface to acquire images of the pool above the water surface.
[0025] In some embodiments, the attitude sensor includes, but is not limited to, an accelerometer, an angle sensor, a direction sensor, a magnetic sensor, etc. Accordingly, the attitude information may include, but is not limited to, acceleration information, velocity information, attitude angles (including pitch angle, roll angle, and yaw angle), and direction information.
[0026] like Figure 1 As shown, the control method for the pool robot includes the following steps: Step 101: If the swimming pool robot is detected to be in a suspected abnormal state through posture information, turn on the image acquisition device to acquire images, or obtain images from the already turned-on image acquisition device. Step 102: Confirm whether the pool robot is in an abnormal state based on the image acquired by the image acquisition device.
[0027] Suspected abnormal states include the pool robot appearing to be stuck or its movement seemingly stopped.
[0028] When detecting whether a swimming pool robot is in a suspected abnormal state using posture information, the difference between the same posture information of the robot within a preset duration can be checked. If the difference is less than a preset threshold, the robot is considered to be in a suspected abnormal state. For example, assuming the posture information includes velocity, if the difference between multiple velocities within a preset duration is less than a preset threshold (e.g., the difference between velocities is less than or equal to a preset value, which can be zero), the robot is considered to be in a suspected abnormal state. Similarly, assuming the posture information includes pitch angle, if the difference between multiple pitch angles within a preset duration is less than a preset threshold (e.g., the difference between pitch angles is less than or equal to a preset value, which can be zero), the robot is considered to be in a suspected abnormal state.
[0029] If the pool robot is detected to be in a suspected abnormal state, the image captured by the image acquisition device can be obtained directly from the image acquisition device that is turned on. Alternatively, if the image acquisition device is not turned on, it can be turned on to acquire images.
[0030] In some embodiments, the external indication that the image acquisition device is turned on can be that the indicator light on the image acquisition device is lit.
[0031] This embodiment first detects whether the pool robot is in a suspected abnormal state by using posture information. If so, it continues to detect whether the pool robot is in an confirmed abnormal state by using images acquired by an image acquisition device. This realizes a two-level detection mechanism for abnormal detection of the pool robot, which improves the accuracy of abnormal detection while ensuring the efficiency of abnormal detection.
[0032] In some embodiments, if the posture information indicates that the pool robot is not in a suspected abnormal state, it can be directly considered to be in a normal working state. Alternatively, if the posture information indicates that the pool robot is not in a suspected abnormal state, it can be further checked using images acquired by an image acquisition device to determine if the pool robot is not in a suspected abnormal state. If the images confirm that the pool robot is not in a suspected abnormal state, it can be considered to be in a normal working state.
[0033] In some embodiments, determining whether a pool robot is in an abnormal status based on images acquired by an image acquisition device includes: If the difference between images collected within a preset duration is less than a preset difference condition, the pool robot is determined to be in an abnormal confirmation state.
[0034] Specifically, the degree of difference between images can include structural differences or pixel-level differences. Structural differences can include variations in the distribution of object shapes, edges, or textures within the image. Pixel-level differences refer to comparing the RGB or grayscale values of an image pixel by pixel; common methods include mean squared error (MSE) and pixel difference images. Correspondingly, if the degree of difference between images is structural, the preset difference condition could be that the image similarity index is less than a preset value; if the degree of difference between images is pixel-level, the preset difference condition could be that the number of pixels with a pixel difference exceeding a preset threshold is less than a preset number.
[0035] The preset duration can be 30 seconds, 40 seconds, etc., and can be set according to needs. If the difference between the images collected within the preset duration is less than the preset difference condition, it means that the pool robot has been continuously collecting images of the same scene within the preset duration, and it can be determined that the pool robot is in an abnormal confirmation state.
[0036] In some embodiments, when confirming whether the pool robot is in an abnormal confirmation state based on the image acquired by the image acquisition device, a depth map can also be obtained based on the image acquired by the image acquisition device. If it is detected that the difference between the depth maps is less than a preset difference condition within a continuous preset time period, the pool robot is determined to be in an abnormal confirmation state.
[0037] Specifically, a depth map can be obtained based on the acquired images. The depth map may contain distance information between the pool robot and various objects in the image. If the difference between the depth maps obtained from the acquired images is less than a preset difference condition within a preset time period, it indicates that the pool robot has been continuously acquiring images of the same scene within the preset time period. In this case, it can be determined that the pool robot is in an abnormal confirmation state.
[0038] When detecting whether the difference between depth maps is less than a preset difference condition within a preset detection period, the difference can be compared from aspects such as pixel-level difference, feature matching, and structural similarity to detect whether the difference is less than the corresponding preset difference condition.
[0039] Optionally, if the degree of difference is analyzed from the pixel level, a difference map can be generated by comparing the depth value differences pixel by pixel. Specifically, the depth difference of the corresponding pixels in the depth map within a preset time period can be calculated. If the number of pixels in the difference map whose depth difference is less than the preset difference is greater than the preset number, then the pool robot can be determined to be in an abnormal confirmation state.
[0040] If the degree of difference is analyzed by feature matching, key points or edge features of the depth map within a preset time period can be extracted. The corresponding region can be found by descriptor matching. If the similarity of the corresponding region is greater than the preset similarity, the pool robot can be determined to be in an abnormal confirmation state.
[0041] If we analyze the degree of difference from the perspective of structural similarity, we can calculate the structural similarity index between depth maps within a preset time period to measure the overall structural consistency. The closer the value of the structural similarity index is to 1, the more similar the two maps are. If the structural similarity index is greater than the preset threshold, then the pool robot can be determined to be in an abnormal confirmation state.
[0042] In some implementations, the distances between the pool robot and various objects in the pool can be obtained based on the depth map. These real-time distances can be fed back to a terminal device connected to the pool robot and displayed on the terminal device, improving the user's interactive experience. The terminal device includes, but is not limited to, remote controls, mobile phones, tablets, laptops, desktop computers, smartwatches, and smart speakers. An application related to the pool robot can be installed on the terminal device.
[0043] In some embodiments, during the operation of the pool robot, a depth map is obtained from images acquired by an image acquisition device; based on the depth map, the current plane where the pool robot is located and its adjacent planes are obtained; if the current plane is the pool bottom plane, the adjacent planes are determined to be the pool wall planes. The current plane can be the clean surface where the pool robot is currently located, such as the pool bottom plane, the pool wall plane, the water surface, etc.
[0044] Based on the depth map, it is possible to distinguish the current plane where the pool robot is located and the adjacent planes of the current plane.
[0045] Specifically, to determine whether the current plane is the pool bottom, the pitch angle obtained from the attitude sensor can be used. If the pitch angle is less than a preset angle threshold and the pool robot is in the water, then the current plane can be determined to be the pool bottom. Alternatively, if the pitch angle is less than the preset angle threshold, the pool robot is in the water, and the bottom of the pool robot is in contact with the cleaning surface, then the current plane can be determined to be the pool bottom. Whether the pool robot is in the water can be determined by the detection data of one or more sensors, including a water pressure sensor, a light sensor, a sound wave sensor, and a capacitance-resistance sensor. Whether the bottom of the pool robot is in contact with the cleaning surface can be determined by the detection data of one or more sensors, including a contact sensor installed on the bottom of the pool robot and a distance sensor whose detection direction is towards the bottom of the pool robot.
[0046] In some embodiments, defining adjacent planes as pool wall planes includes: Obtain the angle between the normal vector direction of the adjacent plane and the normal vector direction of the current plane; If the included angle is greater than the first preset angle threshold, then the adjacent plane is determined as the pool wall plane.
[0047] like Figure 2 As shown, the angle α between the normal vector direction m1 of the adjacent plane P1 and the normal vector direction m2 of the current plane P2 can be obtained. If the angle α is greater than a first preset angle threshold (e.g., 60 degrees or 70 degrees), the adjacent plane P1 can be determined as the pool wall plane. This avoids mistaking a slope with a small angle for the pool wall.
[0048] In some embodiments, after determining the adjacent plane as the pool wall plane, such as Figure 3 As shown, the pool robot 10 can be controlled to move along the boundary of the pool wall plane P1, thereby cleaning the vicinity of the pool wall. When the pool robot moves along the boundary of the pool wall plane, it can either contact the pool wall or maintain a preset distance from it.
[0049] In some embodiments, after determining the adjacent plane as the pool wall plane, such as Figure 4 As shown, the pool robot can be controlled to move to the pool wall plane, thereby enabling it to return through the pool wall or perform cleaning operations.
[0050] In some embodiments, during the operation of the pool robot, a depth map can be obtained from images acquired by an image acquisition device, and a target normal vector for the transition region between the pool bottom and the pool wall can be obtained based on the depth map. If there are two or more target normal vectors with different directions, and at least one of the following conditions is met, then the shape of the transition region is determined to be non-right-angled: The maximum angle difference between the target normal vectors is greater than the second preset angle threshold; The number of target normal vectors is greater than a preset threshold.
[0051] Non-right-angle shapes can be arcs or polygons, etc.
[0052] If the number of target normal vectors in the transition area between the pool bottom and the pool wall is greater than the preset threshold, for example, if the number of target normal vectors is greater than 5 or 7, since the normal vector is a vector perpendicular to the plane, it means that the transition area between the pool bottom and the pool wall is not facing the same plane. Therefore, it can be determined that the shape of the transition area is non-right angle.
[0053] If the maximum angle difference between the target normal vectors in the transition area between the pool bottom and the pool wall is greater than the second preset angle threshold, for example, if the maximum angle difference between the target normal vectors is greater than 30 degrees or 35 degrees, it can be said that there is an angle difference between the planes corresponding to the target normal vectors, and therefore the shape of the transition area can be determined to be non-right angle.
[0054] In some embodiments, during the operation of the pool robot, a depth map is obtained based on images acquired by an image acquisition device, and an obstacle map is constructed based on the continuously acquired depth map; if the pool robot is in an abnormal confirmation state, an abnormal handling operation is performed based on the obstacle map.
[0055] Depth maps obtained from continuously acquired images by image acquisition devices can be used to construct an obstacle map for the swimming pool. The obstacle map can include information such as the location of obstacles in the pool, their size and shape, their classification, and the distance between the pool robot and the obstacles. The obstacle classification can include movable and immovable obstacles, or it can include pool covers, roller blinds, wall lights, pillars, pool platforms, ladders, sun decks, hoses, balloons, etc.
[0056] When the pool robot is in an abnormal confirmed state, an abnormal handling operation can be performed with reference to the obstacle map. For example, an escape operation can be performed based on the obstacle map. Optionally, when performing the escape operation, the robot can escape by performing a rotation, or it can move back a certain distance and then perform a rotation to escape, or it can return to the original path to escape.
[0057] The obstacle map is used for escape, and because the obstacle map contains relatively accurate information such as obstacle information and distance information between the pool robot and the obstacle, the escape method and escape path are also more accurate.
[0058] In some embodiments, after obtaining the current plane where the pool robot is located and the adjacent planes of the current plane based on the depth map, the method further includes: Based on the pre-defined correspondence between the pool bottom, pool walls and anomaly handling strategies, determine the target anomaly handling strategy corresponding to the current plane; Execute exception handling operations using the target exception handling strategy corresponding to the current plane.
[0059] There are pre-set correspondences between the pool bottom, pool walls and anomaly handling strategies. For example, the anomaly handling strategy corresponding to the pool bottom may include: returning along the original path after encountering an obstacle, changing the walking path after encountering an obstacle, and sending a return-to-home message to the base station or user when the amount of trash in the pool robot's trash can exceeds a preset value.
[0060] The abnormal handling strategies for the pool wall can include: after detecting a backward roll and getting stuck, i.e., when the value of the z-axis accelerometer is less than the preset value, it can be restored to normal by swinging left and right; after detecting a lateral movement, it can return to the original path.
[0061] By executing an anomaly handling operation according to the target anomaly handling strategy corresponding to the current plane, the pool robot can efficiently resolve the abnormal state and restore normal operation through the matching anomaly handling operation.
[0062] In some embodiments, after performing exception handling operations through a target exception handling strategy corresponding to the current plane, the method further includes: If the pool robot is still detected to be in an abnormal confirmation state, an alarm message is issued and the trust value of the target abnormal handling strategy is reduced; among them, the initial trust value of each abnormal handling strategy is the highest, and the corresponding relationship is terminated when the trust value is reduced to a preset value. If the pool robot is detected to have returned to normal operation from an abnormal confirmation state, the trust value of the target abnormal handling strategy is increased. If the increased trust value is higher than the initial trust value, the increased trust value is updated to the initial trust value.
[0063] If the pool robot remains in an anomaly confirmation state after the anomaly handling operation, it indicates that the anomaly handling operation has not resolved the anomaly. In this case, an alarm message can be issued. The alarm message can be sent to the user terminal or base station, and can include messages such as "pool robot continues to malfunction" or "pool robot requires manual intervention."
[0064] Each anomaly handling strategy has an initial trust value preset to characterize its success rate. If the pool robot remains in anomaly confirmation state after executing an anomaly handling operation using the target anomaly handling strategy corresponding to the current plane, it indicates that the anomaly handling strategy was ineffective in this instance, and the trust level of the anomaly handling strategy needs to be reduced, i.e., the trust value of the target anomaly handling strategy can be lowered.
[0065] This allows you to set the amount by which the trust value decreases each time. For example, the trust value can decrease in steps. As an example, assuming the initial trust value is 100, the step decrease could be 5 or 10. If it's 5, the trust value decreases by 5 if the exception handling strategy fails to clear the pool robot from the exception confirmation state even once. As another example, the decrease in trust value can increase with the number of unsuccessful exception handling strategies. As an example, assuming the initial trust value is 100, if the number of unsuccessful exception handling strategies is 3 or less, the decrease in trust value could be 5; if the number of unsuccessful exception handling strategies is 3 or more, the decrease in trust value could be 20.
[0066] When the trust value drops to a preset value, the mapping between the exception handling strategy and its corresponding plane can be severed, thus avoiding the continuous use of a success rate-based exception handling strategy. The preset value can be limited according to requirements, for example, it can be 70 or 80.
[0067] Of course, if the pool robot is detected to have returned to normal working state from the abnormality confirmation state after passing through the target abnormality handling strategy, it means that the target abnormality handling strategy has a good abnormality handling effect on its corresponding plane. The trust value of the target abnormality handling strategy can be increased to indicate that the success rate of the abnormality handling strategy is high. Of course, the increased trust value is less than or equal to the initial trust value.
[0068] In some embodiments, after obtaining the current plane where the pool robot is located and the adjacent planes of the current plane based on the depth map, the method further includes: Based on the pre-defined correspondence between the pool bottom, pool walls and cleaning strategies, determine the cleaning strategy corresponding to the current plane; The cleaning operation is performed using the cleaning strategy corresponding to the current plane.
[0069] The bottom and walls of the pool can be pre-set with corresponding cleaning strategies. For example, the cleaning strategy corresponding to the bottom of the pool can include cleaning according to the first bow-shaped path, or targeted cleaning of the preset garbage area, or bow-shaped cleaning in a cross-shaped manner, etc.
[0070] Cleaning strategies for pool walls may include: repeated cleaning of pool wall corners and edges, and increasing cleaning power during the wall-climbing process.
[0071] By executing cleaning operations according to the cleaning strategy corresponding to the current plane, the pool robot can efficiently clean itself, thus improving the cleaning efficiency of the pool.
[0072] In some embodiments, the pool robot, in a specific cleaning mode or in any cleaning mode, can identify the location of trash (e.g., the relative position of the trash and the pool robot, or the absolute coordinates of the trash), and move to that location to clean it. During the cleaning process, if the location of the trash changes due to factors such as water flow disturbance, wind force, or collision with the pool robot, the pool robot can track the location of the trash and clean it. That is, the pool robot can change its cleaning path based on the real-time location of the trash to avoid missing any trash and improve the cleaning effect.
[0073] In some embodiments, when the pool robot is cleaning trash, if the trash to be cleaned consists of multiple clusters of trash or a single piece of trash covering an area larger than a threshold, the cleaning path of the pool robot can at least cover a first area. The first area refers to the area occupied by the trash on the cleaning surface. The pool robot can move within the first area according to a preset path (e.g., a bow-shaped path) or move randomly; there are no restrictions here. This avoids the problem of cleaning omissions caused by a single cleaning operation, thereby improving the effectiveness of trash cleaning.
[0074] All of the above-mentioned optional technical solutions can be combined in any way to form the optional embodiments of this application, and will not be described in detail here.
[0075] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.
[0076] Figure 5 This is a schematic diagram of a control device for a swimming pool robot provided in an embodiment of this application. The swimming pool robot is equipped with at least one image acquisition device and one attitude sensor. The image acquisition device is used to acquire images of the pool surface, and the attitude sensor is used to acquire attitude information of the swimming pool robot. Figure 5 As shown, the device includes: The acquisition module 501 is used to activate the image acquisition device to acquire images or acquire images from the activated image acquisition device when the swimming pool robot is detected to be in a suspected abnormal state through posture information. The judgment module 502 is used to confirm whether the pool robot is in an abnormal confirmation state based on the image acquired by the image acquisition device.
[0077] In some embodiments, the determination module is specifically used to determine that the pool robot is in an abnormal confirmation state if the difference between images collected within a preset duration is less than a preset difference condition; or... Depth maps are obtained from images acquired by image acquisition devices. If the difference between depth maps is less than a preset difference condition within a continuously preset time period, the pool robot is determined to be in an abnormal confirmation state.
[0078] In some embodiments, the apparatus further includes a determining module, configured to obtain a depth map from an image acquired by an image acquisition device during the operation of the pool robot; based on the depth map, determine the current plane where the pool robot is located and the adjacent planes of the current plane; and if the current plane is the bottom plane of the pool, determine the adjacent planes as the pool wall planes.
[0079] In some embodiments, the determining module is specifically used to obtain the angle between the normal vector direction of the adjacent plane and the normal vector direction of the current plane; if the angle is greater than a first preset angle threshold, the adjacent plane is determined as the pool wall plane.
[0080] In some embodiments, the device further includes a control module for controlling the pool robot to move along the boundary of the pool wall plane; or, controlling the pool robot to move to the pool wall plane.
[0081] In some embodiments, the determining module is further configured to obtain a depth map from an image acquired by an image acquisition device during the operation of the pool robot, and obtain a target normal vector of the transition area between the pool bottom and the pool wall based on the depth map; if there are two or more target normal vectors with different directions, and at least one of the following conditions is met, then the shape of the transition area is determined to be non-right angle: the maximum angle difference between the target normal vectors is greater than a second preset angle threshold; the number of target normal vectors is greater than a preset number threshold.
[0082] In some embodiments, the control module is further configured to, during the operation of the pool robot, obtain a depth map based on images acquired by an image acquisition device, and construct an obstacle map based on the continuously acquired depth map; if the pool robot is in an abnormal confirmation state, perform an abnormal handling operation based on the obstacle map.
[0083] In some embodiments, the control module is further configured to determine the target anomaly handling strategy corresponding to the current plane based on a pre-set correspondence between the pool bottom, pool wall and anomaly handling strategy; and execute anomaly handling operations according to the target anomaly handling strategy corresponding to the current plane.
[0084] In some embodiments, the control module is further configured to determine the cleaning strategy corresponding to the current plane based on a pre-set correspondence between the pool bottom, pool wall and cleaning strategy; and to perform cleaning operations using the cleaning strategy corresponding to the current plane.
[0085] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.
[0086] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.
Claims
1. A control method for a swimming pool robot, wherein the swimming pool robot is equipped with at least one image acquisition device and one attitude sensor, the image acquisition device being used at least to acquire images of the pool surface, and the attitude sensor being used at least to acquire attitude information of the swimming pool robot; characterized in that, The methods include: If the swimming pool robot is detected to be in a suspected abnormal state through the posture information, the image acquisition device is activated to acquire an image, or an image is acquired from the already activated image acquisition device. The image acquired by the image acquisition device is used to confirm whether the swimming pool robot is in an abnormal state.
2. The method according to claim 1, characterized in that, Confirming whether the pool robot is in an abnormal state based on the images acquired by the image acquisition device includes: If the difference between images acquired within a preset duration is less than a preset difference condition, the pool robot is determined to be in an abnormal confirmation state; or... A depth map is obtained based on the image acquired by the image acquisition device. If the difference between the depth maps is less than a preset difference condition within a continuously preset time period, the pool robot is determined to be in an abnormal confirmation state.
3. The method according to claim 1, characterized in that, The method further includes: During the operation of the pool robot, depth maps are obtained from images acquired by the image acquisition device. Based on the depth map, the current plane where the pool robot is located and the adjacent planes of the current plane are obtained; If the current plane is the bottom plane of the pool, the adjacent plane is determined to be the pool wall plane.
4. The method according to claim 3, characterized in that, Defining the adjacent planes as pool wall planes includes: Obtain the angle between the normal vector direction of the adjacent plane and the normal vector direction of the current plane; If the included angle is greater than a first preset angle threshold, then the adjacent plane is determined as the pool wall plane.
5. The method according to claim 3, characterized in that, After determining the adjacent planes as pool wall planes, the method further includes: Control the pool robot to move along the boundary of the pool wall plane; or, Control the swimming pool robot to move to the surface of the pool wall.
6. The method according to claim 1, characterized in that, The method further includes: During the operation of the swimming pool robot, a depth map is obtained from the image acquired by the image acquisition device, and the target normal vector of the transition area between the pool bottom and the pool wall is obtained based on the depth map. If there are two or more target normal vectors with different directions, and at least one of the following conditions is met, then the shape of the transition region is determined to be non-right-angled: The maximum angle difference between the target normal vectors is greater than the second preset angle threshold. The number of target normal vectors is greater than a preset threshold.
7. The method according to claim 1, characterized in that, The method further includes: During the operation of the pool robot, a depth map is obtained based on the images acquired by the image acquisition device, and an obstacle map is constructed based on the continuously acquired depth map; If the pool robot is in the abnormal confirmation state, an abnormal handling operation is performed based on the obstacle map.
8. The method according to claim 3, characterized in that, After obtaining the current plane where the pool robot is located and the adjacent planes of the current plane based on the depth map, the method further includes: Based on the pre-set correspondence between the pool bottom, pool walls and anomaly handling strategies, the target anomaly handling strategy corresponding to the current plane is determined. Anomaly handling operations are performed using the target anomaly handling strategy corresponding to the current plane.
9. The method according to claim 8, characterized in that, After performing the exception handling operation through the target exception handling strategy corresponding to the current plane, the method further includes: If the pool robot is detected to still be in the abnormal confirmation state, an alarm message is issued and the trust value of the target abnormal handling strategy is reduced; wherein, the initial trust value of each abnormal handling strategy is the highest, and the corresponding relationship is terminated when the trust value is reduced to a preset value; If the pool robot is detected to have returned to normal operation from an anomaly confirmation state, the trust value of the target anomaly handling strategy is increased. If the increased trust value is higher than the initial trust value, the increased trust value is updated to the initial trust value.
10. The method according to claim 3, characterized in that, After obtaining the current plane where the pool robot is located and the adjacent planes of the current plane based on the depth map, the method further includes: Based on the pre-set correspondence between the pool bottom, pool walls and cleaning strategies, the cleaning strategy corresponding to the current plane is determined; The cleaning operation is performed using the cleaning strategy corresponding to the current plane.