CLEANING METHOD, CLEANING APPARATUS, CLEANING DEVICE, AND STORAGE MEDI

Abstract

A cleaning method comprising: (01) determining a target cleaning area; (02) generating a cleaning trajectory within the target cleaning area in response to a path planning command; and (03) performing a cleaning operation along the cleaning trajectory. The present application further discloses a cleaning apparatus, a cleaning device, and a storage medium.

Description

[Technical field]

[0001] This application claims priority to and the benefit of patent applications under Patent Application Nos. 202210504138.X, 202210597028.2, 202210613623.0, and 202210608104.5, filed with the State Intellectual Property Office of China on May 10, 2022, May 30, 2022, and May 31, 2022, the entire contents of which are incorporated herein by reference.

[0002] The present application relates to the field of cleaning technology, and in particular to a cleaning method, a cleaning apparatus, a cleaning device and a storage medium. [Background technology]

[0003] With the advent of the smart era, various smart devices bring great convenience to people. Smart cleaning devices can reduce people's burden of housework. However, at present, the cleaning strategy of cleaning devices is difficult to achieve a high full-room cleaning coverage rate, and the cleaning effect is not ideal in many aspects such as map construction, route planning, cleaning parameter setting, dynamic scenes, etc. For example, in dynamic scenes, when people walk, pets move, doors open and close, furniture is moved, etc., the cleaning device may not be able to recognize dynamic obstacles, resulting in cleaning omissions. Summary of the Invention [Problem to be solved by the invention]

[0004] SUMMARY OF THE DISCLOSURE Embodiments of the present invention provide a cleaning method, a cleaning apparatus, a cleaning device and a storage medium. [Means for solving the problem]

[0005] A cleaning method according to an embodiment of the present invention is used in a cleaning device, the cleaning method comprising: determining a target cleaning area; generating a cleaning trajectory within the target cleaning area in response to a path planning command; performing a cleaning action along the cleaning trajectory.

[0006] In some embodiments, determining the target cleaning area comprises: After activating the cleaning device, if the cleaning device recognizes that it is activated in a non-charging position, determining a target cleaning area for the cleaning device.

[0007] In some embodiments, determining a target cleaning area for the cleaning device comprises: The method includes a step of recognizing whether a first cleaning target exists within an area at a predetermined distance from the cleaning device, and if the first cleaning target exists, determining the area in which the first cleaning target is located as the target cleaning area.

[0008] In some embodiments, determining a target cleaning area for the cleaning device comprises: obtaining a semantic map of all scenes constructed by the cleaning device; determining the target cleaning area using the semantic map.

[0009] In some embodiments, determining the target cleaning area by the semantic map comprises: If the semantic map determines that the cleaning device is in a room of a predetermined type, determining the room of the predetermined type as the target cleaning area.

[0010] In some embodiments, determining the target cleaning area by the semantic map comprises: When the semantic map determines that the cleaning device is not present in a room of a specified type and that a second cleaning target is present within an area at a specified distance from the cleaning device, the step of determining the area in which the second cleaning target is located as the target cleaning area is included.

[0011] In some embodiments, determining the target cleaning area by the semantic map comprises: If the semantic map determines that the cleaning device is not present in a room of a specified type and that a second cleaning target does not exist within an area of ​​a specified distance from the cleaning device, determining an area of ​​a specified distance from the cleaning device as the target cleaning area.

[0012] In some embodiments, prior to determining a target cleaning area by the semantic map, the cleaning method comprises: The method further includes a step of recognizing whether a first cleaning target exists within an area at a predetermined distance from the cleaning device, and if the first cleaning target does not exist, confirming that the step of determining a target cleaning area is performed by the semantic map.

[0013] In some embodiments, performing a cleaning operation along the cleaning locus comprises: The cleaning method includes a step of cleaning the target cleaning area by cleaning along the edge and then cleaning in a bow shape, and after the cleaning is completed, returning to a charging position and charging.

[0014] In some embodiments, prior to performing a cleaning operation along the cleaning locus, the cleaning method comprises: The method further includes a step of audibly instructing the cleaning device on a cleaning strategy for the target cleaning area.

[0015] In some embodiments, the first cleaning target includes any one of the following types of debris: dirt, small particles of debris, and hair-like debris.

[0016] In some embodiments, the second cleaning target is an appliance or furniture.

[0017] In some embodiments, generating a cleaning trajectory within the target cleaning area in response to a path planning command comprises: in response to a path planning command, determining a current cleanliness state of the mop and a start point and an end point corresponding to said path planning command; determining a cleaning trajectory from the starting point to the end point according to a geographical situation map, a cleaning situation map and the cleanliness status.

[0018] In some embodiments, the step of determining a current cleanliness state of the mop comprises: determining a current cleanliness state of the mop as clean if the current cleaning stage of the cleaning device is a starting stage after self-cleaning in the water station; The method includes determining that the current cleanliness state of the mop is dirty if the current cleaning stage of the cleaning device is moving from the area where the current cleaning is completed to an area waiting to be cleaned, or returning to the water station to wash the mop.

[0019] In some embodiments, the step of determining a cleaning trajectory from the starting point to the end point according to a geographical situation map, a cleaning situation map, and the cleanliness status includes: determining an obstacle-free area according to the geographical situation map; determining a cleaning trajectory from the starting point to the end point according to the obstacle-free area and the cleaning status map, according to a predetermined traffic strategy corresponding to the cleanliness state.

[0020] In some embodiments, the step of determining a cleaning trajectory from the starting point to the end point according to a predetermined traffic strategy corresponding to the cleanliness state according to the obstacle-free area and the cleaning status map includes: determining a currently cleaned area from the cleaning status map based on the cleanliness being clean; determining all areas between the start point and the end point that are free of obstacles and have been cleaned based on the obstacle-free areas and the currently cleaned areas; The method includes a step of determining that at least one continuous movement trajectory can be constructed between the starting point, all areas that are free of obstacles and have been cleaned, and the end point, and determining the shortest movement trajectory among the at least one movement trajectory as the cleaning trajectory from the starting point to the end point.

[0021] In some embodiments, the step of determining a cleaning trajectory from the starting point to the end point according to a predetermined traffic strategy corresponding to the cleanliness state according to the obstacle-free area and the cleaning status map includes: determining all areas waiting to be cleaned from the cleaning status map based on the cleanliness status being dirty; determining all areas between the start point and the end point that are free of obstacles and waiting to be cleaned based on the obstacle-free areas and all areas waiting to be cleaned; The method includes a step of determining that at least one continuous movement trajectory can be constructed between the starting point, all areas free of obstacles and waiting to be cleaned, and the end point, and determining the shortest movement trajectory among the at least one movement trajectory as the cleaning trajectory from the starting point to the end point.

[0022] In some embodiments, the cleaning method includes: The method further includes determining that the continuous movement trajectory cannot be constructed, and determining a cleaning trajectory from the starting point to the end point based on the obstacle-free area.

[0023] In some embodiments, before responding to a path planning command, The method further includes dividing the target cleaning area into a plurality of areas and determining a cleaning sequence for each area.

[0024] In some embodiments, determining a start point and an end point corresponding to the path planning command comprises: determining a start point and an end point corresponding to the path planning command according to a current cleaning stage of the cleaning device and a cleaning sequence of each area;

[0025] In some embodiments, the cleaning method includes: marking obstacles within the target cleaning area and performing a cleaning operation to avoid the obstacles within the target cleaning area; Detecting the state of obstacles in the target cleaning area and determining which obstacles have changed state; and performing a cleaning operation at the location of the obstacle whose state has changed.

[0026] In some embodiments, the step of marking obstacles within the target cleaning area comprises: marking the location, type and shape of obstacles within the target cleaning area, the types including dynamic obstacles.

[0027] In some embodiments, the step of detecting the state of an obstacle within the target cleaning area and determining an obstacle whose state has changed includes: detecting a position and shape of the moving obstacle within the target cleaning area; determining a dynamic obstacle whose state has changed based on the dynamic obstacle whose position or shape has changed.

[0028] In some embodiments, the cleaning action includes a first cleaning action and a second cleaning action, wherein a first cleaning trajectory determined by performing the first cleaning action is a closed loop, and a second cleaning trajectory determined by performing the second cleaning action can fill the target cleaning area.

[0029] In some embodiments, determining the target cleaning area comprises: performing the first cleaning operation; and determining the target cleaning area based on a cleaning trajectory of the first cleaning operation.

[0030] In some embodiments, the step of avoiding the obstacles and performing a cleaning operation within the target cleaning area includes: performing a second cleaning operation within the target cleaning area; performing a first cleaning action when it is determined that the obstacle has been encountered, and continuing to perform a second cleaning action after avoiding the obstacle.

[0031] In some embodiments, performing a cleaning operation along the cleaning locus comprises: monitoring the amount of dust collected in a dust box of the cleaning device; determining a target operation mode of the cleaning device based on a magnitude relationship between the dust collection amount and a predetermined dust amount threshold, where the target operation mode is used to change a cleaning strength and a cleaning area of ​​the cleaning device; and performing the cleaning operation by driving the cleaning device to operate in the target operating mode.

[0032] In some embodiments, the step of determining a target operation mode of the cleaning device based on a magnitude relationship between the dust collection amount and a predetermined dust amount threshold value includes: Detecting that the dust amount is greater than the predetermined dust amount threshold and determining to control the cleaning device to operate a target operating mode of a cleaning function at a first cleaning intensity and a first cleaning area; or The method includes detecting that the dust collection amount is not greater than the predetermined dust amount threshold and determining to control the cleaning device to operate a target operating mode of a cleaning function at a second cleaning intensity and a second cleaning area, where the second cleaning intensity is lower than the first cleaning intensity and the second cleaning area is smaller than the first cleaning area.

[0033] In some embodiments, the step of detecting that the dust volume is greater than the predetermined dust volume threshold and determining to control the cleaning device to operate a target operating mode of a cleaning function at a first cleaning intensity and a first cleaning area comprises: determining that the dust amount is greater than the predetermined dust amount threshold and recording the location of the cleaning device; The method further includes the step of marking an area within a predetermined range from the location as a high cleaning area and controlling the cleaning device to operate a cleaning function at a first cleaning intensity and a first cleaning area in the high cleaning area.

[0034] In some embodiments, the step of driving the cleaning device to operate in the target operating mode includes: determining at least one of the following items represented by the target operating mode: a rotation speed of a suction motor, a travel interval, a cleaning area, and a number of cleaning repetitions; The method includes driving the cleaning device to perform a cleaning function with at least one of a rotation speed of a suction motor, a running interval, a cleaning area, and a number of cleaning repetitions, and the running interval is used to reflect the distance between two adjacent running routes of the cleaning device.

[0035] In some embodiments, the step of driving the cleaning device to perform a cleaning function at the travel interval comprises: determining an initial interval between two adjacent travel routes of the current cleaning device; adjusting the initial interval to a target interval whose distance is less than the initial interval; The method includes a step of driving the cleaning device and operating a cleaning function such that the distance between two adjacent travel routes becomes the target interval.

[0036] In some embodiments, the step of determining a target operation mode of the cleaning device based on a magnitude relationship between the dust collection amount and a predetermined dust amount threshold value includes: detecting that the dust amount is greater than the predetermined dust amount threshold and activating a photographing device of the cleaning device; Using the photographing device to capture an area image in a travel direction of the cleaning device; and determining a target operation mode of the cleaning device based on the recognition result of the area image and a magnitude relationship between the dust collection amount and a predetermined dust amount threshold.

[0037] In some embodiments, after the step of capturing an image of an area in a direction of travel of the cleaning device using the image capture device, Recognizing area object features in the area image using a predetermined image detection model, the area object features including at least one of a size feature, a color feature, and a contour feature; The method further includes a step of determining a recognition result of the area image based on the area object characteristics, where the recognition result is used to reflect objects waiting to be cleaned that exist in the area where the cleaning device is currently located.

[0038] In some embodiments, the step of monitoring the amount of dust collected in a dust box of the cleaning device comprises: The method includes a step of monitoring the amount of dust collected at the dust box inlet for a predetermined period of time, or a step of monitoring a change in weight of the dust box for a predetermined period of time.

[0039] In an embodiment of the present invention, a cleaning device includes: a location planning module used to determine a target cleaning area; a trajectory planning module adapted to generate a cleaning trajectory within the target cleaning area in response to a path planning command; a cleaning module adapted to perform a cleaning operation along the cleaning trajectory.

[0040] A cleaning device in an embodiment of the present invention includes a memory, a processor, and a computer program stored in the memory and executable by the processor, which, when executed by the processor, implements the cleaning method.

[0041] A computer-readable storage medium in an embodiment of the present invention stores a computer program, which, when executed by a processor, realizes the cleaning method.

[0042] Additional aspects and advantages of the invention will be set forth in part in the description that follows, and in part will be obvious from the description, or may be learned by practice of the invention. [Brief description of the drawings]

[0043] The above and / or additional aspects and advantages of the present invention will become apparent and be readily understood from the following description of the embodiments taken in conjunction with the drawings.

[0044] [Figure 1] FIG. 2 is a schematic diagram of a cleaning method process according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a schematic diagram of a cleaning method process according to an embodiment of the present invention. [Diagram 3] FIG. 2 is a schematic diagram of a cleaning method process in an embodiment of the present invention. [Figure 4] FIG. 2 is a schematic diagram of a cleaning method process according to an embodiment of the present invention. [Diagram 5] FIG. 2 is a schematic diagram of a cleaning method process in an embodiment of the present invention. [Figure 6] FIG. 2 is a schematic diagram of a cleaning method process according to an embodiment of the present invention. [Figure 7] FIG. 2 is a schematic diagram of a cleaning method process according to an embodiment of the present invention. [Figure 8] FIG. 2 is a schematic diagram of a cleaning method process according to an embodiment of the present invention. [Figure 9] FIG. 2 is a schematic diagram of a cleaning method process in an embodiment of the present invention. [Figure 10] FIG. 2 is a schematic diagram of a scene in an embodiment of the present invention. [Figure 11] FIG. 2 is a schematic diagram of a scene in an embodiment of the present invention. [Figure 12] FIG. 2 is a schematic diagram of a scene in an embodiment of the present invention. [Figure 13] FIG. 2 is a schematic diagram of a cleaning method process in an embodiment of the present invention. [Figure 14] FIG. 2 is a schematic diagram of a scene in an embodiment of the present invention. [Figure 15] FIG. 2 is a schematic diagram of a cleaning method process in an embodiment of the present invention. [Figure 16] FIG. 2 is a schematic diagram of a scene in an embodiment of the present invention. [Figure 17] FIG. 2 is a schematic diagram of a scene in an embodiment of the present invention. [Figure 18] FIG. 2 is a schematic diagram of a cleaning method process in an embodiment of the present invention. [Figure 19] FIG. 2 is a schematic diagram of a module of a cleaning device in an embodiment of the present invention. [Figure 20] FIG. 2 is a schematic diagram of a module of a cleaning device in an embodiment of the present invention. [Figure 21] FIG. 2 is a schematic diagram of a module of a cleaning device in an embodiment of the present invention. [Figure 22] 1 is a schematic diagram of a cleaning device according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0045] In order to make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention are described below clearly and completely with reference to the accompanying drawings, and obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without any creative efforts are included in the protection scope of the present invention.

[0046] In the description of this specification, reference to the terms "one embodiment", "some embodiments", "examples", "specific examples" or "several examples" means that the specific features, structures, materials or characteristics described in the corresponding embodiment or example are included in at least one embodiment or example of the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only and do not indicate or imply relative importance. Illustrative expressions of the above terms in this specification do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials or characteristics described can be combined in any suitable manner in any embodiment or example. In addition, if not mutually inconsistent, a person skilled in the art can combine different embodiments or examples described in this specification and features of different embodiments or examples.

[0047] Hereinafter, the present invention will be described in more detail with reference to the accompanying drawings and examples. The following examples are intended to illustrate the present invention but are not intended to limit the scope of the present invention.

[0048] Referring to FIG. 1, an embodiment of the present invention discloses a cleaning method for use in a cleaning device, the cleaning method comprising: 01. Step of determining a target cleaning area; 02. generating a cleaning trajectory within a target cleaning area in response to a path planning command; 03. performing a cleaning operation along the cleaning trajectory.

[0049] In addition, in actual dynamic application scenes, such as people walking, pets moving, doors opening and closing, and furniture moving, the execution terminal of the embodiment method of the present invention may be a cleaning device such as a cleaning robot, an automatic floor washer, a robot combining cleaning and mopping, an automatic cleaning cleaner, a smart vacuum cleaner, or a robot vacuum cleaner.

[0050] Furthermore, when performing a cleaning task, the cleaning device first determines a target cleaning area of ​​the cleaning task in the cleaning scene, and if an obstacle is encountered when determining the target cleaning area, the cleaning device avoids the detected obstacle and marks the detected obstacle. In the subsequent cleaning process, if the obstacle detected when determining the target cleaning area is a dynamic obstacle and its position moves, the cleaning device will clean the area again.

[0051] 2 is an exemplary flowchart for determining a target cleaning area according to some embodiments of the present disclosure. In some embodiments, step 01 of FIG. 1 can be performed based on step 011 of FIG.

[0052] 011. After activating the cleaning device, if it is determined that the cleaning device is activated in a non-charging position, a target cleaning area of ​​the cleaning device is determined.

[0053] In addition, in this embodiment, the entity that executes the proposed cleaning method may be a cleaning apparatus that is a cleaning device. In order to more clearly explain the cleaning method proposed in the present invention, in the following embodiment, an illustrative explanation is provided using a cleaning apparatus that is a cleaning device as an example that executes the cleaning method.

[0054] In this embodiment, the cleaning device is activated in a non-charging position when the user carries the cleaning device to a certain position and manually activates it, at which time the cleaning device enters an operating mode for local cleaning.

[0055] Specifically, after the cleaning device is started, the cleaning device first recognizes whether the starting position of the cleaning device is a non-charging position. Specifically, when the cleaning device is in an awake state, the position recognition component in the cleaning device is also started accordingly, and the position recognition component can detect whether the cleaning device is in a charging position, and the cleaning device can determine whether the starting position of the cleaning device is a non-charging position according to the recognition result of the position recognition component.

[0056] Furthermore, when the cleaning device is determined to be activated in a non-charging position, i.e. when the cleaning device is brought to a location and manually activated, the cleaning device needs to determine a target cleaning area in which the cleaning device needs to clean.

[0057] Specifically, when the cleaning device determines that the cleaning device is activated in a non-charging position, it indicates that the cleaning device is in a local cleaning mode, and at this time, it can be determined that the user's cleaning intention is to clean a certain area, and therefore the cleaning device needs to determine the target cleaning area, so as to accurately clean according to the user's cleaning intention in subsequent steps.

[0058] Specifically, a sensing component is provided in the cleaning device, which is used to sense situations near the location of the cleaning device, a semantic map of all scenes is stored in the cleaning device, and the cleaning apparatus can analyze the user's cleaning intention based on the sensing result of the sensing component and / or the semantic map, and further determine the target cleaning area of ​​the cleaning device based on the user's cleaning intention.

[0059] In this embodiment, the cleaning device first determines whether the cleaning device is in a state to be activated in the non-charging position, and when the cleaning device is in a state to be activated in the non-charging position, determines the area that the user wants to clean (i.e., the target cleaning area), and then controls the cleaning device to clean the area that the user wants to clean. In this embodiment of the present invention, when the cleaning device determines that the cleaning device is in a state to be activated in the non-charging position, it can automatically analyze the user's cleaning intention, and by determining the area that the user wants to clean, the cleaning area of ​​the cleaning device can be more in line with the user's requirements, thereby improving the user's satisfaction. In addition, by controlling the cleaning device to clean only the area that the user wants to clean, the cleaning efficiency of the cleaning device can be improved.

[0060] Referring to FIG. 2, in some embodiments, step 01 includes: 012, recognizing whether a first cleaning target is present within an area of ​​a predetermined distance from the cleaning device;

[013] If the first cleaning target exists, the method further includes determining an area in which the first cleaning target exists as a target cleaning area.

[0061] In this embodiment, the predetermined distance refers to a distance that can be recognized by the sensing component of the cleaning device, and can be a custom distance. It can be understood that the custom distance needs to be within the range that can be recognized by the sensing component of the cleaning device, and the first cleaning target refers to dirt or dust that is not easy to clean.

[0062] Specifically, the process of determining the target cleaning area of ​​the cleaning device is as follows: the cleaning device first recognizes whether there is a first cleaning target in an area at a predetermined distance from the cleaning device. Specifically, if the cleaning device is provided with a sensing component that can be used to sense the situation near the location of the cleaning device, the cleaning device can recognize the situation near the location of the cleaning device through the sensing component, that is, recognize whether there is the first cleaning target near the cleaning device.

[0063] By way of example, the sensing component may be one of the following types: an AI (image) recognition device, a camera, a video camera, etc., but is not limited thereto.

[0064] Furthermore, when the cleaning device determines that a first cleaning target exists within an area at a predetermined distance from the cleaning device, the cleaning device determines the area in which the first cleaning target is located as the area the user desires to clean, i.e., determines the area in which the first cleaning target is located as the target cleaning area.

[0065] Specifically, when the cleaning device recognizes that a first cleaning target exists within an area at a predetermined distance from the cleaning device, the user's cleaning intention is to clean the first cleaning target, and at this time, the cleaning device determines the area where the first cleaning target is located as the target cleaning area, and in a subsequent step, the cleaning device simply controls the cleaning device to clean the area where the first cleaning target is located, so that the cleaning area of ​​the cleaning device is more in line with the user's requirements.

[0066] In this embodiment, when the cleaning device recognizes that a first cleaning target exists within an area at a predetermined distance from the cleaning device, the cleaning device determines the area where the first cleaning target is located as the target cleaning area, i.e., determines the area where the first cleaning target is located as the area that the user wants to clean. Then, in a subsequent step, the cleaning device only controls the cleaning device to clean the area where the first cleaning target is located, so that the cleaning area of ​​the cleaning device meets the user's requirements, thereby improving the cleaning efficiency of the cleaning device and improving the user's satisfaction.

[0067] Referring to FIG. 3, in some embodiments, step 01 includes: 014, obtaining a semantic map of all scenes constructed by the cleaning device; 015, determining a target cleaning area by a semantic map.

[0068] In this embodiment, the semantic map refers to a semantic map of all scenes that the cleaning device builds after cleaning all scenes, where all scenes refer to scenes of all rooms in the user's home, and the semantic map specifically includes room types and the locations and types of items such as home appliances and furniture.

[0069] Specifically, in the process of determining the target cleaning area of ​​the cleaning device, the cleaning device first obtains the semantic map that the cleaning device has constructed after cleaning in all scenes. Specifically, the cleaning device first constructs a semantic map of all scenes in the user's house after cleaning all rooms in the user's house for the first time, and then updates the semantic map according to the actual situation after each cleaning, and stores the updated semantic map in the storage unit of the cleaning device. The cleaning device can obtain the semantic map through the storage unit. In addition, the cleaning device obtains the updated semantic map after the latest cleaning, so as to ensure the accuracy of the target cleaning area determined by the semantic map in the subsequent steps.

[0070] Furthermore, the cleaning device determines the target cleaning area through the semantic map, i.e., determines the area the user wishes to clean through the semantic map. Specifically, the semantic map can determine the room type where the cleaning device is currently located and whether furniture or home appliances are present at the location where the cleaning device is currently located, and the user's cleaning intention can be analyzed based on the information, and the user's cleaning intention can determine the area the user wishes to clean. Therefore, the cleaning device can determine the target cleaning area through the semantic map.

[0071] In this embodiment, when it is determined that the cleaning device is in a state to be activated in a non-charging position, i.e., when the cleaning device activates a local cleaning mode, the cleaning apparatus can obtain the semantic map constructed by the cleaning device, and use the semantic map to analyze the situation near the current location of the cleaning device, i.e., analyze the user's cleaning intention, and determine the target cleaning area according to the user's cleaning intention. In a subsequent step, the area cleaned by the cleaning device is ensured to be the area desired by the user to be cleaned, so that the cleaning area of ​​the cleaning device meets the user's requirements, improves the cleaning efficiency of the cleaning device, and improves user satisfaction.

[0072] Referring to FIG. 4, in some embodiments, step 014 includes:

[0151] If the semantic map determines that the cleaning device is in a room of a predetermined type, determining the room of a predetermined type as the target cleaning area.

[0073] In this embodiment, the predetermined type of room refers to a room that is not easy to clean or is easily soiled, such as a toilet or a kitchen.

[0074] Specifically, in the process of determining the target cleaning area by the semantic map, the cleaning device first analyzes whether the room type where the cleaning device is currently located is a predetermined room type by the semantic map.

[0075] Furthermore, if the cleaning device determines that the room type of the location where the cleaning device is currently located is a predetermined room type, the cleaning device determines the room as the area the user wishes to clean, that is, determines the room as the target cleaning area.

[0076] Specifically, when the cleaning device analyzes through the semantic map that the room type where the cleaning device is currently located is a predetermined room type, the user's cleaning intention is to clean a room of a predetermined type, and at this time, the cleaning device determines the room of the predetermined type (i.e., the room where the cleaning device is currently located) as the target cleaning area, and in a subsequent step, the cleaning device simply controls the cleaning device to clean the room of the predetermined type, so that the cleaning area of ​​the cleaning device better meets the user's requirements.

[0077] In this embodiment, if the cleaning device analyzes that the current location of the cleaning device is a certain type of room through the semantic map, the cleaning device determines the certain type of room as the area that the user wants to clean, i.e., determines the certain type of room as the target cleaning area, so that in a subsequent step, the cleaning device only controls the cleaning device to clean the certain type of room, so that the cleaning area of ​​the cleaning device is more in line with the user's requirements, improving the cleaning efficiency of the cleaning device and improving the user's satisfaction.

[0078] Referring to FIG. 5, in some embodiments, step 015 includes: and if the semantic map determines that the cleaning device is not present in a room of a specified type and that a second cleaning target is present within an area at a specified distance from the cleaning device, determining the area in which the second cleaning target is located as the target cleaning area.

[0079] In this embodiment, the second cleaning target refers to the appliances or furniture in the user's room that are likely to interrupt the cleaning process of the cleaning device, such as a bed, a tea table, and a washing machine.

[0080] Specifically, in the process of determining the target cleaning area by the semantic map, when the cleaning device analyzes that the room type of the current location of the cleaning device is not a predetermined room type by the semantic map, the cleaning device analyzes whether there is a second cleaning target within an area of ​​a predetermined distance from the cleaning device by the semantic map. Specifically, by clarifying the second cleaning target by the semantic map, the cleaning device does not need to call the detection component of the cleaning device to detect the situation near the cleaning device, and can determine whether there is a second cleaning target within an area of ​​a predetermined distance from the cleaning device by the semantic map.

[0081] Furthermore, when the cleaning device analyzes that a second cleaning target exists within an area within a predetermined distance from the cleaning device through the semantic map, the cleaning device determines the area where the second cleaning target is located as the area the user wishes to clean, i.e., determines the area where the second cleaning target is located as the target cleaning area.

[0082] Specifically, when the cleaning device analyzes through the semantic map that the room type where the cleaning device is currently located is not the specified room type and there is a second cleaning target within an area at a specified distance from the cleaning device, the user's cleaning intention is to clean the area where the second cleaning target is located. At this time, the cleaning device determines the area where the second cleaning target is located as the target cleaning area, and in a subsequent step, the cleaning device simply controls the cleaning device to clean the area where the second cleaning target is located, so that the cleaning area of ​​the cleaning device is more in line with the user's requirements.

[0083] In this embodiment, if the cleaning device analyzes through the semantic map that the location of the cleaning device is not a predetermined type of room and there is a second cleaning target within an area at a predetermined distance from the cleaning device, the cleaning device determines the area where the second cleaning target is located as the area that the user wants to clean, i.e., determines the area where the second cleaning target is located as the target cleaning area. Then, in a subsequent step, the cleaning device only controls the cleaning device to clean the area where the second cleaning target is located, so that the cleaning area of ​​the cleaning device meets the user's requirements, thereby improving the cleaning efficiency of the cleaning device and improving the user's satisfaction.

[0084] Referring to FIG. 6, in some embodiments, step 015 includes:

[0153] If the semantic map determines that the cleaning device is not present in a room of a predetermined type and that a second cleaning target is not present within an area of ​​a predetermined distance from the cleaning device, determining an area of ​​a predetermined distance from the cleaning device as a target cleaning area.

[0085] In this embodiment, the process of determining the target cleaning area using the semantic map is as follows: the cleaning device first analyzes, using the semantic map, whether the room type where the cleaning device is currently located is a predetermined room type, and whether there is a second cleaning target within an area within a predetermined distance from the cleaning device.

[0086] Specifically, if the cleaning device analyzes through the semantic map that the room type of the location where the cleaning device is currently located is not the specified room type and a second cleaning target exists within an area within a specified distance from the cleaning device, the cleaning device determines the area within a specified distance from the cleaning device as the area the user wants to clean, i.e., determines the area within a specified distance from the cleaning device as the target cleaning area.

[0087] Specifically, if the cleaning device analyzes through the semantic map that the room type of the location where the cleaning device is currently located is not the specified room type and there is no second cleaning target within the area at a specified distance from the cleaning device, it indicates that the user's cleaning intention is to clean the area at a specified distance from the cleaning device. At this time, the cleaning device determines the area at a specified distance from the cleaning device as the target cleaning area, and in a subsequent step, the cleaning device simply controls the cleaning device to clean the area at a specified distance from the cleaning device, so that the cleaning area of ​​the cleaning device better meets the user's requirements.

[0088] In this embodiment, if the cleaning device analyzes through the semantic map that the location of the cleaning device is not a predetermined type of room and there is no second cleaning target within the area at a predetermined distance from the cleaning device, it determines the area at a predetermined distance from the cleaning device as the area the user wants to clean, i.e., determines the area at a predetermined distance from the cleaning device as the target cleaning area, so that in a subsequent step, the cleaning device only controls the cleaning device to clean the area at a predetermined distance from the cleaning device, so that the cleaning area of ​​the cleaning device meets the user's requirements, improves the cleaning efficiency of the cleaning device, and improves the user's satisfaction.

[0089] Referring to FIG. 7, in some embodiments, prior to step 015, step 01 includes: 016, recognizing whether a first cleaning target is present within an area of ​​a predetermined distance from the cleaning device; 017, further including the step of confirming that if the first cleaning target does not exist, the step of determining the target cleaning area by the semantic map is performed.

[0090] In this embodiment, before the cleaning device determines the target cleaning area through the semantic map, the cleaning device also needs to recognize whether there is a first cleaning target in an area at a predetermined distance from the cleaning device. Specifically, a sensing component is provided on the cleaning device, which can be used to sense the situation near the location of the cleaning device, and the cleaning device can recognize the situation near the location of the cleaning device through the sensing component, i.e., recognize whether there is the first cleaning target near the cleaning device.

[0091] Further, it is determined that when it is determined that the cleaning device recognizes that the first cleaning target is not present within an area at a predetermined distance from the cleaning device, the cleaning device can perform a step of determining the target cleaning area by the semantic map.

[0092] Specifically, when the cleaning device recognizes that the first cleaning target is not present within an area of ​​a predetermined distance from the cleaning device, the user's cleaning intention is expressed as not dirt or debris near the cleaning device, and at this time, the cleaning device can perform steps such as analyzing the user's specific cleaning intention through the semantic map to clean a certain type of room, clean near the second cleaning target, or clean an area near the cleaning device.

[0093] In this embodiment, before the cleaning device performs the step of determining the target cleaning area of ​​the cleaning device through the semantic map, it also needs to recognize whether the first target cleaning area exists within an area of ​​a predetermined distance from the cleaning device, so that the user's cleaning intention can be more accurately determined, and by determining the more accurate target cleaning area, the cleaning area of ​​the cleaning device can be more in line with the user's requirements, improving the cleaning efficiency of the cleaning device and improving the user's experience effect.

[0094] In some embodiments, step 03 includes: 031, cleaning the target cleaning area using a cleaning method in which cleaning is performed along the edge and then in a bow shape, and after completion, returning to the charging position and charging.

[0095] In this embodiment, the process of the cleaning apparatus controlling the cleaning device to clean the target cleaning area is that the cleaning apparatus controls the cleaning device to clean the boundary of the target cleaning area in an edge-wise manner and then clean an area of ​​the target cleaning area in an arcuate manner.

[0096] Specifically, since the above-mentioned target cleaning area is the area that the user wants to clean primarily, the cleaning device needs to adopt the cleaning methods of edge cleaning and bow-shaped cleaning, so as to ensure that the required cleaning effect can be achieved.

[0097] In addition, after determining that the cleaning device has completed cleaning the target cleaning area, the cleaning device controls the cleaning device to return to the charging position and charge. Specifically, the cleaning device can determine whether to complete cleaning of the target cleaning area according to the cleaning time of the cleaning device, and detect the cleaning situation near the cleaning device through the detection component of the cleaning device to determine whether to complete cleaning of the target cleaning area, but is not limited to this.

[0098] In this embodiment, the cleaning device adopts cleaning methods of edge cleaning and bow-shaped cleaning to control the cleaning device to clean the target cleaning area, thereby ensuring cleaning of the target cleaning area, achieving the cleaning effect, and improving user satisfaction.

[0099] In some embodiments, prior to step 03, the cleaning method includes: 04. The method further includes the step of presenting a cleaning method of the cleaning device for the target cleaning area by voice.

[0100] In this embodiment, before controlling the cleaning device to clean the target cleaning area, the cleaning device needs to play audio prompting information to inform the user of how the cleaning device will control the cleaning device to clean the target cleaning area.

[0101] Specifically, before the cleaning device cleans the target cleaning area, the cleaning device plays audio prompting information so that the user can understand the area to be cleaned. For example, the area to be cleaned by the cleaning device is the area where the tea table is located, and the cleaning device plays audio prompting information such as "The area under the tea table will be cleaned soon", so that the user can confirm whether the area to be cleaned by the cleaning device is the area the user wants to clean.

[0102] In this embodiment, before controlling the cleaning device to clean the target cleaning area, the cleaning device plays voice prompting information of the area to be cleaned, so that the user can confirm whether the area to be cleaned by the cleaning device is the area the user wants to clean, and if the cleaning area of ​​the played voice prompting information is not the area the user wants to clean, the cleaning area of ​​the cleaning device can be adjusted in a timely manner.

[0103] In the above embodiment, the first cleaning target includes any one of the following types of debris: dirt, small particles, and hair-like debris.

[0104] In this embodiment, the first cleaning target may be, but is not limited to, dirt that is not easy to clean, small granular dust that is not easy to clean, and hair-like dust that is not easy to clean.

[0105] In the above example, the second cleaning target is an appliance or furniture.

[0106] In this embodiment, the second cleaning target may be, but is not limited to, an appliance or furniture.

[0107] It should be noted that FIG. 2 is merely an illustrative example, and the method of determining the target area is not limited to the method described in the above embodiment.

[0108] 8 is an exemplary flowchart for generating a cleaning trajectory according to some embodiments of the present disclosure. In some embodiments, step 02 of FIG. 1 can be performed based on steps 021 and 022 of FIG.

[0109] 021, in response to the path planning command, determine the current clean state of the mop and the start point and end point corresponding to the path planning command.

[0110] 022, the cleaning trajectory from the starting point to the end point is determined according to the geographical situation map, the cleaning situation map and the cleanliness state.

[0111] In a possible embodiment, before step 021, the target cleaning area is divided into multiple areas, and a cleaning order for each area is determined. The control device of the cleaning device divides the cleaning task into multiple serialized subtasks, and issues corresponding subtasks according to the serialized order, and each subtask has a corresponding subarea.

[0112] In a possible embodiment, in step 021, the step of determining a start point and an end point corresponding to the path planning command includes a step of determining a start point and an end point corresponding to the path planning command according to a current cleaning stage of the cleaning device and a cleaning sequence of each area.

[0113] In a possible embodiment, in step 021, determining the current cleanliness state of the mop comprises: determining that the current cleanliness state of the mop is clean if the current cleaning stage of the cleaning device is a starting stage after self-cleaning at the water station; and determining that the current cleanliness state of the mop is dirty if the current cleaning stage of the cleaning device is moving from an area where the current cleaning is completed to an area waiting to be cleaned, or returning to the water station to wash the mop.

[0114] In a possible embodiment, in step 022, the step of determining a cleaning trajectory from the starting point to the end point according to a geographical situation map, a cleaning situation map and the cleanliness state includes: determining an obstacle-free area according to the geographical situation map; determining a cleaning trajectory from the starting point to the end point according to the obstacle-free area and the cleaning status map, according to a predetermined traffic strategy corresponding to the cleanliness state.

[0115] In a possible embodiment, the step of determining a cleaning trajectory from the starting point to the end point according to the obstacle-free area and the cleaning status map and according to a predetermined traffic strategy corresponding to the clean state includes: The method includes a step of determining an area that has currently been cleaned from the cleaning status map based on the clean state being clean; a step of determining all areas from the starting point to the end point that are free of obstacles and have been cleaned based on the areas free of obstacles and the currently cleaned areas; a step of determining that at least one continuous movement trajectory can be configured between the starting point, all areas that have been cleaned that are free of obstacles, and the end point, and determining the shortest movement trajectory among the at least one movement trajectory as the cleaning trajectory from the starting point to the end point.

[0116] In a possible embodiment, the step of determining a cleaning trajectory from the starting point to the end point according to the obstacle-free area and the cleaning status map and according to a predetermined traffic strategy corresponding to the clean state includes: The method includes a step of determining all areas waiting to be cleaned from the cleaning status map based on the cleanliness state being dirty; a step of determining all areas between the starting point and the end point that are free of obstacles and waiting to be cleaned based on the areas free of obstacles and all areas waiting to be cleaned; a step of determining that at least one continuous movement trajectory can be configured between the starting point, all areas free of obstacles and waiting to be cleaned, and the end point, and determining the shortest movement trajectory among the at least one movement trajectory as the cleaning trajectory from the starting point to the end point.

[0117] In a possible embodiment, the method further comprises the step of determining that the continuous movement trajectory cannot be constructed and determining a cleaning trajectory from the start point to the end point according to the obstacle-free area.

[0118] In a possible embodiment, before step 021, the user remotely controls the cleaning device through an application or a remote control, or directly controls the cleaning device through buttons, to start the cleaning device to perform cleaning.

[0119] In a possible embodiment, before step 021, the method further includes a step in which the control device of the cleaning device performs multi-layer map initialization to obtain information of each map of the multi-layer map, and periodically updates the map information in real time during the cleaning process.

[0120] By using the accurate geographical information obtained by the multi-layer map, different cleaning trajectories can be changed according to the geographical situation map where the cleaning device is located, the cleaning situation map and the cleanliness state, and the change of the cleaning trajectory is self-adaptive, and the subsequent cleaning trajectory can be adjusted according to the position and stage of the cleaning device, thereby reducing secondary contamination of the floor surface and achieving high-efficiency cleaning.

[0121] In a possible embodiment, the geographical situation of each of the subareas is determined by the situation of an obstacle layer of the raster layer of each of the subareas, and the cleaning situation of each of the subareas is determined by the situation of a cleaning situation layer of the raster layer of each of the subareas.

[0122] When planning a cleaning path, the geographical situation and cleaning status of an area in the raster layer are determined based on the obstacle layer status and cleaning status layer status of the raster layer.

[0123] At the stage where the mop of the cleaning device has just been washed, the mop passes preferentially through areas that are free of obstacles and have already been cleaned according to the obstacle layer state and cleaning status layer state, and at the stage of normal cleaning or at the stage of interruption or completion where the mop of the cleaning device is highly likely to be dirty, the mop passes preferentially through uncleaned areas according to the obstacle layer state and cleaning status layer state.

[0124] During path planning, the stage at which the cleaning device is located should be fully considered, and different planning strategies should be adopted at different stages to more efficiently avoid cross-pollution.

[0125] In a possible embodiment, the geographical situation map and the cleaning situation map are updated in real time according to respective set intervals.

[0126] The real-time updated geographical map and cleaning status map can be set to different intervals according to the user's needs, or they can be set to the same interval.

[0127] The need for updating the obstacle situation arises from fluctuations due to moving or temporary obstacles.

[0128] The so-called update can be seen as updating the attribute values ​​corresponding to a layer in a raster map.

[0129] In a possible embodiment, the method further comprises the steps of: If the cleaning path planning for a sub-area fails, the cleaning path planning for the next sub-area is performed.

[0130] If the cleaning path planning for a sub-area fails, the sub-area is deemed unreachable, the sub-area is ignored, and the cleaning subtask switches to the next sub-area and continues to replan the cleaning path for the next sub-area. If there is no next sub-area, cleaning ends.

[0131] In the present embodiment, the path planning algorithms used in the cleaning device path planning process include, but are not limited to, DijkStra, A*, JPS, etc., and the purpose is to find a path that is the shortest and has the least pollution.

[0132] Referring to FIG. 9, FIG. 9 shows a cleaning method provided in the embodiment of the present application, where a user remotely controls the cleaning device through an application or a remote control, or directly controls the cleaning device through a button, and starts the cleaning device to perform cleaning. The task management module of the control device of the cleaning device prepares to start cleaning by planning a cleaning path according to the received cleaning command, where the cleaning command has information such as a cleaning task type and a cleaning area. The cleaning task is divided into multiple serialized subtasks, and corresponding subtasks are sent according to the serialization order, and each subtask has a corresponding subarea. And the map module of the control device of the cleaning device performs multi-layer map initialization to obtain map information of each layer of the multi-layer map, and periodically updates the map information in real time during the cleaning process.

[0133] During the process of path planning for each sub-area, if the plan is successful, execute the cleaning of the sub-area according to the planned path. If the plan fails, re-plan by adjusting the conditions of the considered map layer, and if the plan is successful this time, execute it. If the plan fails again, skip the path planning for this sub-area, switch the cleaning subtask to the next sub-area, and continue to re-plan the cleaning path for the next sub-area. If there is no next sub-area, the cleaning ends.

[0134] Determine a current clean state of the mop and a start point and an end point corresponding to the path planning command according to a path planning command, and further determine a cleaning trajectory from the start point to the end point according to a geographical situation map, a cleaning situation map and the clean state.

[0135] When determining the cleanliness status of the mop, if the current cleaning stage of the cleaning device is the departure stage after self-cleaning at the water station, the current cleanliness status of the mop is determined to be clean, and if the current cleaning stage of the cleaning device is the stage of moving from the area where the current cleaning is completed to the waiting area for cleaning, or the stage of returning to the water station to wash the mop, the current cleanliness status of the mop is determined to be dirty.

[0136] An obstacle-free area is determined according to the geographical situation map, and a cleaning trajectory from the start point to the end point is determined according to the obstacle-free area and the cleaning situation map according to a predetermined traffic strategy corresponding to the cleanliness state.

[0137] If the cleanliness state is clean, a currently cleaned area is determined from the cleaning status map, and all areas between the starting point and the end point that are free of obstacles and have been cleaned are determined based on the areas free of obstacles and the currently cleaned areas, it is determined that at least one continuous movement trajectory can be configured between the starting point, all areas free of obstacles and that have been cleaned, and the end point, and the shortest movement trajectory among the at least one movement trajectory is determined as the cleaning trajectory from the starting point to the end point.

[0138] Since the clean state is dirty, all areas waiting to be cleaned are determined from the cleaning status map, and all areas between the starting point and the end point that are free of obstacles and waiting to be cleaned are determined using the areas free of obstacles and all areas waiting to be cleaned, it is determined that at least one continuous movement trajectory can be formed between the starting point, all areas free of obstacles and waiting to be cleaned, and the end point, and the shortest movement trajectory among the at least one movement trajectory is determined as the cleaning trajectory from the starting point to the end point.

[0139] If it is determined that the continuous moving trajectory cannot be constructed, a cleaning trajectory from the starting point to the end point is determined according to the obstacle-free area.

[0140] Figure 10 shows layer 1 of a raster map, which stores obstacle and prohibited area (equivalently regarded as obstacle) information of the map integrated with multiple sensor records. As shown in Figure 10, the area types in the layer include obstacle, free area and map boundary.

[0141] 11 shows Layer 2 of the raster map, where the cleaning track information of the cleaning device is stored. As shown in FIG. 11, the area types in the layer include uncleaned, cleaned and map boundary.

[0142] In a possible embodiment, the path planning stages for the entire cleaning process include a stage where the cleaning device washes the mop and departs from the water station (hereinafter referred to as stage 1), a stage where during the normal cleaning process, the cleaning device moves from the cleaned area to other areas waiting to be cleaned (hereinafter referred to as stage 2), and a stage where cleaning is interrupted or completed and the cleaning device needs to return to the water station to wash the mop (hereinafter referred to as stage 3).

[0143] In each stage, it is necessary to plan a reasonable route for the cleaning device to move, using a multi-layer raster map, and adopting different planning strategies according to the different geographic stages in which the cleaning device is located, to ensure that the device preferentially passes through the areas that have been cleaned in stage 1, and preferentially passes through the areas that have not been cleaned in stages 2 and 3, thereby reducing cross-contamination of the floor surface.

[0144] 12 shows a schematic diagram of an embodiment of path planning, and it can be seen that in the path planning of the cleaning task of the sub-area, there are two selectable paths, namely, path 1 and path 2. The path 2 may be selected and passed through an area that is free of obstacles and has already been cleaned after the cleaning device has cleaned the starting point, and the path 2 may be selected and passed through an area that is free of obstacles and will not be cleaned when the cleaning device returns from the end point to the starting point.

[0145] During the cleaning process of the cleaning device, two layers of raster maps are maintained in real time, and each layer of raster maps stores a kind of environmental features, where layer 1 stores obstacle and prohibited area (equivalent to obstacle) information of the map integrated with the records of multiple sensors, and layer 2 stores the cleaning trajectory information of the cleaning device. In the path planning of the cleaning device, the robot adopts different planning strategies according to different cleaning stages, and prioritizes passing through non-cleaned areas during normal cleaning and return mop cleaning navigation, and prioritizes passing through cleaned areas when continuing cleaning after mopping, thereby reducing secondary contamination of the floor surface.

[0146] The embodiments of the present application take as examples the target information that is most commonly considered in the application scenarios of cleaning devices, but in addition to obstacle information and information on the path cleaned by the cleaning device, there are also layers of user personalized setting information, and the embodiments of the present application are not limited thereto.

[0147] In the present embodiment, the method of acquiring the geographical situation includes, but is not limited to, a raster map.

[0148] In some embodiments, step 021 includes: 0211, when the current cleaning stage of the cleaning device is the starting stage after self-cleaning in the water station, determining that the current clean state of the mop is clean; 0212, if the current cleaning stage of the cleaning device is moving from a currently completed cleaning area to a waiting cleaning area, or returning to the water station to wash the mop, determining that the current cleanliness state of the mop is dirty.

[0149] In a possible embodiment, before step 0211, the user remotely controls the cleaning device through an application or a remote control, or directly controls the cleaning device through buttons, to start the cleaning device and perform cleaning.

[0150] In a possible embodiment, in step 0211, the control device of the cleaning device prepares to initiate cleaning by performing cleaning path planning according to the received cleaning command, where the cleaning command has information such as cleaning task type and cleaning area.

[0151] In a possible embodiment, the geographical situation map of each of the subareas is determined by the situation of the obstacle layer of the raster layer of each of the subareas, and the cleaning situation map of each of the subareas is determined by the situation of the cleaning situation layer of the raster layer of each of the subareas.

[0152] When planning a cleaning path, the obstacle state and cleaning status of an area in the raster layer are determined based on the obstacle layer state and the cleaning status layer state of the raster layer.

[0153] At the stage where the mop of the cleaning device has just been washed, the mop passes preferentially through areas that are free of obstacles and have already been cleaned according to the obstacle layer state and cleaning status layer state, and at the stage of normal cleaning or at the stage of interruption or completion where the mop of the cleaning device is highly likely to be dirty, the mop passes preferentially through uncleaned areas according to the obstacle layer state and cleaning status layer state.

[0154] During path planning, the stage at which the cleaning device is located should be fully considered, and different planning strategies should be adopted at different stages to more efficiently avoid cross-pollution.

[0155] In a possible embodiment, the current cleanliness status of the mop of the cleaning device, the geographical situation map of each sub-area, and the cleaning status map are updated in real time according to respective set intervals.

[0156] The real-time updated geographical map and cleaning status map can be set to different intervals according to the user's needs, or they can be set to the same interval.

[0157] The need for updating the geographical situation arises from changes due to movement or temporary obstacles. The so-called updates can be regarded as updating the attribute values ​​corresponding to the layers in the raster map.

[0158] In the present embodiment, the path planning algorithms used in the process of cleaning device path planning include, but are not limited to, DijkStra, A*, JPS, etc., and the purpose is to find a path that is the shortest and has the least pollution.

[0159] In some embodiments, step 021 includes: determining an area free of obstacles according to the geographical situation map;

[0214] The method further includes: determining a cleaning track from a starting point to an end point according to the obstacle-free area and the cleaning status map, according to a predetermined travel strategy corresponding to the clean state.

[0160] When planning a cleaning path, the obstacle state and cleaning status of an area in the raster layer are determined based on the obstacle layer state and the cleaning status layer state of the raster layer.

[0161] At the stage where the mop of the cleaning device has just been washed, the mop passes preferentially through areas that are free of obstacles and have already been cleaned according to the obstacle layer state and cleaning status layer state, and at the stage of normal cleaning or at the stage of interruption or completion where the mop of the cleaning device is highly likely to be dirty, the mop passes preferentially through uncleaned areas according to the obstacle layer state and cleaning status layer state.

[0162] During path planning, the stage at which the cleaning device is located should be fully considered, and different planning strategies should be adopted at different stages to more efficiently avoid cross-pollution.

[0163] In a possible embodiment, the current cleanliness status of the mop of the cleaning device, the geographical situation map of each sub-area, and the cleaning status map are updated in real time according to respective set intervals.

[0164] The real-time updated geographical map and cleaning status map can be set to different intervals according to the user's needs, or they can be set to the same interval.

[0165] The need for updating the geographic situation arises from changes due to movement or temporary obstructions.

[0166] The so-called update can be seen as updating the attribute values ​​corresponding to a layer in a raster map.

[0167] In the embodiment of the present application, the path planning algorithm used in the cleaning device path planning process includes, but is not limited to, DijkStra, A*, JPS, etc., and the purpose is to find a path that is the shortest and has the least pollution.

[0168] In some embodiments, step 02 includes: Step a: in response to a path planning command, determine a current clean state of the mop and a start point and an end point corresponding to the path planning command, and if the current cleaning stage of the cleaning device is a departure stage after self-cleaning at the water station, determine that the current clean state of the mop is clean; if the current cleaning stage of the cleaning device is a stage of moving from the area where the current cleaning is completed to the waiting area for cleaning, or a stage of returning to the water station to wash the mop, determine that the current clean state of the mop is dirty; Step b: determining an obstacle-free area according to a geographical situation map; and determining a cleaning track from the starting point to the end point according to the obstacle-free area and the cleaning situation map according to a predetermined traffic strategy corresponding to the clean state.

[0169] In a possible embodiment, in step b, the cleaning state is determined to be clean, thereby determining a currently cleaned area from the cleaning status map, and based on the area free of obstacles and the currently cleaned area, determining all areas between the starting point and the end point that are free of obstacles and have been cleaned, determining that at least one continuous movement trajectory can be constructed between the starting point, all areas free of obstacles and that have been cleaned, and the end point, and determining the shortest movement trajectory among the at least one movement trajectory as the cleaning trajectory from the starting point to the end point.

[0170] In a possible embodiment, in step b, the clean state is determined to be dirty, thereby determining all areas waiting to be cleaned from the cleaning status map, and determining all areas between the starting point and the end point that are free of obstacles and waiting to be cleaned by the areas free of obstacles and all areas waiting to be cleaned, determining that at least one continuous movement trajectory can be constructed between the starting point, all areas free of obstacles and waiting to be cleaned, and the end point, and determining the shortest movement trajectory among the at least one movement trajectory as the cleaning trajectory from the starting point to the end point.

[0171] In a possible embodiment, before step a, the user remotely controls the cleaning device through an application or a remote control, or directly controls the cleaning device through a button, to start the cleaning device and perform cleaning.

[0172] In a possible embodiment, in step a, the control device of the cleaning device prepares to initiate cleaning by planning a cleaning path according to the received cleaning command, where the cleaning command has information such as a cleaning task type and a cleaning area.

[0173] In a possible embodiment, the geographical situation map of each of the subareas is determined by the situation of the obstacle layer of the raster layer of each of the subareas, and the cleaning situation map of each of the subareas is determined by the situation of the cleaning situation layer of the raster layer of each of the subareas.

[0174] When planning a cleaning path, the obstacle state and cleaning status of an area in the raster layer are determined based on the obstacle layer state and the cleaning status layer state of the raster layer.

[0175] At the stage where the mop of the cleaning device has just been washed, the mop passes preferentially through areas that are free of obstacles and have already been cleaned according to the obstacle layer state and cleaning status layer state, and at the stage of normal cleaning or at the stage of interruption or completion where the mop of the cleaning device is highly likely to be dirty, the mop passes preferentially through uncleaned areas according to the obstacle layer state and cleaning status layer state.

[0176] During path planning, the stage at which the cleaning device is located should be fully considered, and different planning strategies should be adopted at different stages to more efficiently avoid cross-pollution.

[0177] In a possible embodiment, the current cleanliness status of the mop of the cleaning device, the geographical situation map of each sub-area, and the cleaning status map are updated in real time according to respective set intervals.

[0178] The real-time updated geographical map and cleaning status map can be set to different intervals according to the user's needs, or they can be set to the same interval.

[0179] The need for updating the geographic situation arises from changes due to movement or temporary obstructions.

[0180] The so-called update can be seen as updating the attribute values ​​corresponding to a layer in a raster map.

[0181] In the embodiment of the present application, the path planning algorithm used in the cleaning device path planning process includes, but is not limited to, DijkStra, A*, JPS, etc., and the purpose is to find a path that is the shortest and has the least pollution.

[0182] In some embodiments, step 02 includes: Step c: in response to the path planning command, determine the current clean state of the mop and a start point and an end point corresponding to the path planning command, and if the current cleaning stage of the cleaning device is a departure stage after self-cleaning at the water station, determine that the current clean state of the mop is clean; if the current cleaning stage of the cleaning device is a stage of moving from the area where the current cleaning is completed to the waiting area for cleaning, or a stage of returning to the water station to wash the mop, determine that the current clean state of the mop is dirty; Step d: determining an obstacle-free area according to a geographical situation map; and determining a cleaning track from the starting point to the end point according to the obstacle-free area and the cleaning situation map according to a predetermined traffic strategy corresponding to the clean state.

[0183] In a possible embodiment, in step d, the cleaning state is determined to be clean, thereby determining a currently cleaned area from the cleaning status map; determining all areas between the starting point and the end point that are free of obstacles and have been cleaned based on the areas free of obstacles and the currently cleaned area; determining that at least one continuous movement trajectory can be constructed between the starting point, all areas that are free of obstacles and have been cleaned, and the end point, and determining the shortest movement trajectory among the at least one movement trajectory as the cleaning trajectory from the starting point to the end point.

[0184] In a possible embodiment, in step d, the method includes the steps of: determining all areas waiting to be cleaned from the cleaning status map by the cleanliness state being dirty; determining all areas between the starting point and the end point that are free of obstacles and waiting to be cleaned by the areas free of obstacles and all areas waiting to be cleaned; determining that at least one continuous movement trajectory can be constructed between the starting point, all areas free of obstacles and waiting to be cleaned, and the end point, and determining the shortest movement trajectory among the at least one movement trajectory as the cleaning trajectory from the starting point to the end point.

[0185] In a possible embodiment, the method further comprises the step of determining that the continuous movement trajectory cannot be constructed and determining a cleaning trajectory from the start point to the end point according to the obstacle-free area.

[0186] In a possible embodiment, before step c, the user remotely controls the cleaning device via an application or a remote control, or directly controls the cleaning device via buttons, to start the cleaning device and perform cleaning.

[0187] In a possible embodiment, in step c, the control device of the cleaning device prepares to initiate cleaning by performing cleaning path planning according to the received cleaning command, where the cleaning command has information such as cleaning task type and cleaning area.

[0188] In a possible embodiment, the geographical situation map of each of the subareas is determined by the situation of the obstacle layer of the raster layer of each of the subareas, and the cleaning situation map of each of the subareas is determined by the situation of the cleaning situation layer of the raster layer of each of the subareas.

[0189] When planning a cleaning path, the obstacle state and cleaning status of an area in the raster layer are determined based on the obstacle layer state and the cleaning status layer state of the raster layer.

[0190] At the stage where the mop of the cleaning device has just been washed, the mop passes preferentially through areas that are free of obstacles and have already been cleaned according to the obstacle layer state and cleaning status layer state, and at the stage of normal cleaning or at the stage of interruption or completion where the mop of the cleaning device is highly likely to be dirty, the mop passes preferentially through uncleaned areas according to the obstacle layer state and cleaning status layer state.

[0191] During path planning, the stage at which the cleaning device is located should be fully considered, and different planning strategies should be adopted at different stages to more efficiently avoid cross-pollution.

[0192] In a possible embodiment, the current cleanliness status of the mop of the cleaning device, the geographical situation map of each sub-area, and the cleaning status map are updated in real time according to respective set intervals.

[0193] The real-time updated geographical map and cleaning status map can be set to different intervals according to the user's needs, or they can be set to the same interval.

[0194] The need for updating the geographic situation arises from changes due to movement or temporary obstructions.

[0195] The so-called update can be seen as updating the attribute values ​​corresponding to a layer in a raster map.

[0196] In the embodiment of the present application, the path planning algorithm used in the cleaning device path planning process includes, but is not limited to, DijkStra, A*, JPS, etc., and the purpose is to find a path that is the shortest and has the least pollution.

[0197] In some embodiments, step 02 includes: Step e: in response to the path planning command, determine the current clean state of the mop and a start point and an end point corresponding to the path planning command, and if the current cleaning stage of the cleaning device is a departure stage after self-cleaning at the water station, determine that the current clean state of the mop is clean; if the current cleaning stage of the cleaning device is a stage of moving from the area where the current cleaning is completed to the waiting area for cleaning, or a stage of returning to the water station to wash the mop, determine that the current clean state of the mop is dirty; Step f further includes determining an obstacle-free area according to a geographical situation map, and determining a cleaning trajectory from the starting point to the end point according to the obstacle-free area and the cleaning situation map according to a predetermined traffic strategy corresponding to the clean state.

[0198] In a possible embodiment, the method further includes the step of dividing the target cleaning area into a plurality of areas and determining a cleaning sequence for each area before responding to the path planning command in step e.

[0199] In a possible embodiment, the step of determining a start point and an end point corresponding to the path planning command includes a step of determining a start point and an end point corresponding to the path planning command according to a current cleaning stage of the cleaning device and a cleaning order of each area.

[0200] In a possible embodiment, in step f, the cleaning state includes a step of determining a currently cleaned area from the cleaning status map by being clean; a step of determining all areas between the starting point and the end point that are free of obstacles and have been cleaned based on the areas free of obstacles and the currently cleaned area; a step of determining that at least one continuous movement trajectory can be constructed between the starting point, all areas that are free of obstacles and have been cleaned, and the end point, and a step of determining the shortest movement trajectory among the at least one movement trajectory as the cleaning trajectory from the starting point to the end point.

[0201] In a possible embodiment, in step f, the cleanliness state is dirty, and the method includes the steps of determining all areas waiting to be cleaned from the cleaning status map; determining all areas between the starting point and the end point that are free of obstacles and waiting to be cleaned by the areas free of obstacles and all areas waiting to be cleaned; determining that at least one continuous movement trajectory can be constructed between the starting point, all areas free of obstacles and waiting to be cleaned, and the end point, and determining the shortest movement trajectory among the at least one movement trajectory as the cleaning trajectory from the starting point to the end point.

[0202] In a possible embodiment, the method further comprises the step of determining that the continuous movement trajectory cannot be constructed and determining a cleaning trajectory from the start point to the end point according to the obstacle-free area.

[0203] In a possible embodiment, before step e, the user remotely controls the cleaning device via an application or a remote control, or directly controls the cleaning device via buttons, to start the cleaning device and perform cleaning.

[0204] In a possible embodiment, in step e, the control device of the cleaning device prepares to initiate cleaning by performing cleaning path planning according to the received cleaning command, where the cleaning command has information such as cleaning task type and cleaning area.

[0205] In a possible embodiment, the geographical situation map of each of the subareas is determined by the situation of the obstacle layer of the raster layer of each of the subareas, and the cleaning situation map of each of the subareas is determined by the situation of the cleaning situation layer of the raster layer of each of the subareas.

[0206] When planning a cleaning path, the obstacle state and cleaning status of an area in the raster layer are determined based on the obstacle layer state and the cleaning status layer state of the raster layer.

[0207] At the stage where the mop of the cleaning device has just been washed, the mop passes preferentially through areas that are free of obstacles and have already been cleaned according to the obstacle layer state and cleaning status layer state, and at the stage of normal cleaning or at the stage of interruption or completion where the mop of the cleaning device is highly likely to be dirty, the mop passes preferentially through uncleaned areas according to the obstacle layer state and cleaning status layer state.

[0208] During path planning, the stage at which the cleaning device is located should be fully considered, and different planning strategies should be adopted at different stages to more efficiently avoid cross-pollution.

[0209] In a possible embodiment, the current cleanliness status of the mop of the cleaning device, the geographical situation map of each sub-area, and the cleaning status map are updated in real time according to respective set intervals.

[0210] The real-time updated geographical map and cleaning status map can be set to different intervals according to the user's needs, or they can be set to the same interval.

[0211] The need for updating the geographic situation arises from changes due to movement or temporary obstructions.

[0212] The so-called update can be seen as updating the attribute values ​​corresponding to a layer in a raster map.

[0213] In the embodiment of the present application, the path planning algorithm used in the cleaning device path planning process includes, but is not limited to, DijkStra, A*, JPS, etc., and the purpose is to find a path that is the shortest and has the least pollution.

[0214] In summary, by responding to a path planning command, determine the current clean state of the mop and the start point and end point corresponding to the path planning command, and determine the cleaning trajectory from the start point to the end point according to the geographical situation map, the cleaning situation map and the clean state. In the process of the cleaning task of the cleaning device, the path is reasonably planned, so as to reduce secondary pollution of the road surface as much as possible and improve the cleaning effect.

[0215] It should be noted that FIGS. 8-12 are merely illustrative, and the method of generating the cleaning trajectory is not limited to the contents described in the above embodiment.

[0216] 13 is an exemplary flow chart of a cleaning method according to some embodiments of the present disclosure. In some embodiments, the cleaning method includes: 01. Step of determining a target cleaning area; 05. Marking obstacles in a target cleaning area, avoiding the obstacles in the target cleaning area, and performing cleaning operations; 06. Detecting the state of an obstacle in the target cleaning area and determining the obstacle whose state has changed; 07. performing a cleaning operation at the location of the obstacle whose state has changed.

[0217] In actual dynamic application scenarios, such as people walking, pets moving, doors opening and closing, and furniture moving, the execution terminal of the method according to the embodiment of the present invention may be a cleaning device, an automatic cleaning device, a smart vacuum cleaner, a robot vacuum cleaner, and other devices, and is not limited thereto. The following description will be given with the cleaning device as the execution terminal.

[0218] It should be noted that, with respect to step 01, when performing a cleaning task, the cleaning device first determines a target cleaning area of ​​the cleaning task in the cleaning scene, and if an obstacle is encountered when determining the target cleaning area, the cleaning device avoids the detected obstacle and marks the detected obstacle. In the subsequent cleaning process, if the obstacle detected when determining the target cleaning area is a dynamic obstacle and its position moves, the cleaning device will clean the area again.

[0219] It should be noted that, with regard to step 05, the cleaning device itself has the functions of target recognition, target segmentation and target tracking, which allow the robot to mark obstacles in the target cleaning area when performing the cleaning task, save the current obstacle in the obstacle collection after marking it, and then avoid the current obstacle to clean the remaining target cleaning area.

[0220] Regarding step 06, during each cleaning process, the cleaning device detects the status of nearby obstacles in real time, and if it detects that the status of an obstacle has changed, it marks the obstacle whose status has changed so that it can be checked the next time it cleans.

[0221] With regard to step 07, the cleaning device can execute steps 01 to 06 in sequence in one cleaning task until cleaning of the current target cleaning area is completed. After the current cleaning task is completed, the cleaning device executes step 07, traverses the set of obstacles collectively, checks whether the state of any obstacle in the set of obstacles has changed, and if any obstacle has changed state, moves to the position of the obstacle whose state has changed and performs a cleaning operation.

[0222] In addition, after the cleaning device detects that the state of an obstacle has changed in the current cleaning task, it immediately moves to the position of the obstacle to perform the cleaning operation. For example, after the cleaning device avoids the current obstacle, it immediately detects that the state of the obstacle has changed and immediately cleans the initial position of the obstacle, without needing to additionally clean the position after the cleaning of the entire room is completed.

[0223] The cleaning method of the present embodiment can detect movable obstacles in a target cleaning area during the cleaning process, and determine whether the state of the moving obstacles has changed. The cleaning method can move to the position of the obstacle whose state has changed and perform the cleaning operation, thereby effectively preventing cleaning omissions caused by moving objects and improving the coverage rate of automatic cleaning of dynamic scenes.

[0224] It can be understood that the steps of marking obstacles in the target cleaning area are: Marking the location, type and shape of obstacles within the target cleaning area.

[0225] In addition, when the cleaning device encounters an obstacle, it can first mark the location of the obstacle in the target cleaning area, and then obtain the type and shape of the obstacle based on the visual sensing device and radar device installed on the cleaning device. The types of obstacles are divided into static obstacles and dynamic obstacles, and the embodiment of the present application mainly focuses on the impact of dynamic obstacles on the cleaning coverage rate. The category of dynamic obstacles has many subcategories, such as furniture, pets, shoes, socks, doors, trash cans, etc.

[0226] Specifically, the types include dynamic obstacles and static obstacles, and the types are used to determine whether the obstacles need to be detected. The position of a static obstacle generally does not change, and once the position is determined, the area it covers will not be the target cleaning area for cleaning, so the static obstacles are also marked when marking, but the robot does not need to detect the state of the static obstacles, which saves computation power.

[0227] The type of obstacle is generally acquired through processing after being collected by a visual sensing device, where the visual sensing device may be an RGB camera, the cleaning device collects RGB image data of obstacles on the travel route, marks the location of the obstacles, and outputs the type of obstacle data currently collected through a target recognition method. A dynamic obstacle refers to an obstacle whose position or shape constantly moves in the scene, and a static obstacle refers to an obstacle whose position constantly does not change in the scene, such as a wall, a pillar, etc.

[0228] The shape of the obstacle is generally collected by a radar device and then processed to obtain it, the radar device may be a rotating laser lidar, the rotating laser lidar can collect point cloud information of the obstacle, and the point cloud information includes the 3D shape of the obstacle. In addition, some obstacle information needs to be combined with a visual sensing device and a laser device to comprehensively judge the type and shape of the obstacle, for example, in the case of a pet, the change in its movement will affect the range of the waiting area for cleaning.

[0229] Specifically, the embodiment of the present invention constructs a target recognition method and adopts a general deep learning-based target detection model as a basic model, such as Convolutional Neural Network (CNN), R-CNN, Fast R-CNN, YOLO network model, etc. The target recognition algorithm constructs a variety of training sets in the field of household cleaning to train the target detection model, and the labels in the training set include types of obstacles commonly found in household scenes, such as furniture, people, pets, shoes, chairs, electric wires, socks, doors, trash cans, etc. By fetching and training the depth features of obstacles in real time, finally, the target recognition method in the embodiment of the present invention can output the type of obstacle according to the visual data.

[0230] In addition, the door, which is an obstacle, needs to be combined and comprehensively processed with information obtained from the visual sensing device and the laser device, and the visual sensing device detects whether the door is open or closed, and the laser device judges whether the opening and closing size of the door is suitable for the cleaning device to pass through, and obtains the door position and door state. Here, the door open state indicates that the cleaning device can enter, and the door closed state indicates that the cleaning device cannot enter. The reason why the door is particularly indicated among the obstacles is that the opening and closing of the door affects the range of the target cleaning area, and therefore, when it is recognized in 06 that the door state has changed from closed to open, the cleaning device needs to move to the door position and execute the target cleaning area determination task again from step 01.

[0231] According to the cleaning method of the present embodiment, the position, type and shape of the obstacles in the target cleaning area can be marked during the cleaning process, and the marked obstacles can be stored in the obstacle set. Not all obstacle types are known in the obstacle set, but if some obstacles change state during subsequent detection, the types of those obstacles can be marked in the obstacle set as dynamic obstacles. By marking dynamic obstacles, the method of the present embodiment can allow the cleaning device to process the position of dynamic obstacles in a dynamic scene, and avoid missing them.

[0232] It can be understood that the steps of detecting the state of obstacles in the target cleaning area and determining which obstacles have changed state are as follows: Detecting the location and shape of a moving obstacle within a target cleaning area; determining a dynamic obstacle whose state has changed based on the dynamic obstacle whose position or shape has changed.

[0233] It should be noted that the embodiment of the present invention mainly detects dynamic obstacles, and the state of the obstacle includes a state of existence and a state of nonexistence, particularly for a door, the state of existence corresponds to a closed state, and the state of nonexistence corresponds to an open state. When the cleaning device performs a cleaning task in the target cleaning area, it collects the obstacle state in the sensing area of ​​its visual sensing device and radar device in real time. The change of the obstacle state includes a change of position or a change of shape, and these two changes may cause the obstacle to move from its original position, and the state of the obstacle may change from a state of existence to a state of nonexistence. If the state of the obstacle no longer exists, the cleaning device can re-clean the area occupied by the obstacle.

[0234] It can be understood that the cleaning operation includes a first cleaning operation and a second cleaning operation, wherein the first cleaning trajectory determined by performing the first cleaning operation is a closed loop, and the second cleaning trajectory determined by performing the second cleaning operation can fill the target cleaning area.

[0235] The first cleaning motion may be an edge cleaning motion, specifically including cleaning along a wall or along the edge of an obstacle, starting from a starting point and finally returning to the starting point, so that the cleaning trajectory forms a closed loop. The second cleaning motion may be a bow-shaped cleaning motion, specifically the cleaning device starts from an origin position, enters the target cleaning area, starts moving in a forward direction, and when the cleaning device reaches the rightmost end of the cleaning area, turns around and moves in the opposite direction, and when the cleaning device reaches the leftmost end of the cleaning area, turns around and moves in the forward direction. Finally, when the bow-shaped cleaning motion is completed, the formed second cleaning trajectory fills the first cleaning trajectory formed by the edge cleaning motion.

[0236] With reference to FIG. 14, it can be seen that what the drawing shows is a first cleaning trajectory and a second cleaning trajectory that are ultimately formed when, in an ideal state and without any obstacles, the first cleaning motion is cleaning along the edge and the second cleaning motion is bow-shaped cleaning.

[0237] It can be understood that the steps to determine the target cleaning area are: performing a first cleaning operation and determining a target cleaning area based on a cleaning trajectory of the first cleaning operation.

[0238] In addition, the cleaning device cleans along the wall of the environment in the first cleaning operation, obtains a target cleaning area, and the target cleaning area becomes the map created in this cleaning task, and the subsequent obstacle positions are stored in the map in coordinate format. For example, the cleaning device cleans along the edge of a rectangular room to obtain a rectangular target cleaning area.

[0239] It can be understood that the steps of avoiding obstacles and performing cleaning operations within a target cleaning area include: performing a second cleaning operation within the target cleaning area; performing a first cleaning action when it is determined that an obstacle has been encountered, and continuing to perform a second cleaning action after avoiding the obstacle.

[0240] After checking the target cleaning area, the cleaning device performs bow-shaped cleaning within the target cleaning area, and when it encounters an obstacle, it cleans along the edge of the obstacle in a counterclockwise direction, and then continues to perform the bow-shaped cleaning motion. After repeatedly performing the first cleaning motion and the second cleaning motion, it can finally clean the area other than the area covered by the obstacle in the target cleaning area.

[0241] It can be understood that the step of performing a cleaning action at the location of the obstacle whose state has changed includes three situations: a first situation, a second situation, and a third situation.

[0242] In the first situation, the obstacle is removed and a second cleaning operation is performed in the area occupied by the obstacle, specifically, by performing additional cleaning in response to real-time changes in the obstacle through bow-shaped cleaning, thereby eliminating the need to re-clean the entire room and improving cleaning efficiency.

[0243] In the second situation, the obstacle moves slightly, and a first cleaning operation is performed in the area occupied by the obstacle, and if the position of the obstacle moves slightly, the area covered by the original obstacle is not completely exposed, at this time, the first cleaning operation is performed to additionally clean the exposed area.

[0244] In a third situation, if the original obstacle is removed but another obstacle exists, then the first cleaning action needs to be combined with the second cleaning action, for example, first cleaning along the edge to determine the boundary of the new obstacle, and then cleaning in an arc to clean the uncleaned target cleaning area between the original boundary and the current boundary.

[0245] A second cleaning operation is then performed in the area occupied by the obstacle whose state has changed.

[0246] It can be understood that when an obstacle is encountered, the present invention utilizes a path planning algorithm based on the first cleaning action and the second cleaning action to optimize the first cleaning trajectory and the second cleaning trajectory, and avoids repeated cleaning as much as possible.

[0247] Exemplarily, the operating sequence in which the cleaning device performs the cleaning method includes the following steps:

[0248] Step 201, a cleaning device performs edge cleaning at an edge of a target cleaning area.

[0249] Step 202, the cleaning device performs arcuate cleaning within the target cleaning area and performs edge cleaning when an obstacle is encountered.

[0250] In steps 203, 201 and 202, the movable obstacles oi are recorded in the obstacle set 0.

[0251] Movable obstacles fall into three categories: a. If the obstacle is visible to the laser lidar, record the obstacle's position and point cloud information. b. If the obstacle is recognized by the RGB camera, the position and type of the obstacle are recorded. c. If it is a room door, record the open / close state of the room door.

[0252] Step 204: Repeat steps 202 and 203 until cleaning of the target cleaning area is completed.

[0253] Step 205, perform navigation traversal for obstacles in 0 to detect whether there are obstacles, and the detection method is as follows:

[0254] e. If an obstacle is detected by the laser lidar, check whether point cloud information exists.

[0255] f. If an obstacle is detected by the RGB camera, check whether the obstacle's position moves.

[0256] In step 206, if there is no obstacle detected in step 205, a single arcuate cleaning is performed on the area occupied by the obstacle.

[0257] In addition, during the above driving process, the cleaning task is divided into two parts. The first cleaning is the traditional normal cleaning, but it needs to mark the location and status of dynamic obstacles. The second cleaning is to inspect the movable obstacle area and perform additional cleaning. The above method can not only improve the efficiency and obstacle avoidance success rate of the first cleaning, but also improve the coverage rate of the second cleaning.

[0258] It should be noted that FIGS. 13 and 14 are merely illustrative, and the method of generating the cleaning trajectory is not limited to the contents described in the above embodiment.

[0259] 15 is an exemplary flow chart for performing a cleaning operation according to some embodiments of the present disclosure. In some embodiments, step 03 of FIG. 1 can be performed based on steps 032-034 of FIG.

[0260] 032. Monitor the amount of dust collected in the dust box of the cleaning device.

[0261] 034. Determine a target operation mode of the cleaning device based on the magnitude relationship between the dust collection amount and a predetermined dust amount threshold, where the target operation mode is used to change the cleaning strength and cleaning area of ​​the cleaning device.

[0262] 035, performing a cleaning operation by driving and operating the cleaning device in a target operation mode.

[0263] Furthermore, the present application relates to adjusting the width of adjacent cleaning paths of the cleaning device, the number of cleaning times of adjacent cleaning paths, and the rotation speed of the motor according to the degree of dirt in the current cleaning area, thereby representing the cleaning function of the operating mode.

[0264] In one embodiment, for example, the more the current area is determined to be dirty, the more the rotation speed of the suction motor is increased; and / or the more the current area is determined to be dirty, the narrower the width of the adjacent cleaning path is determined, the wider the overlapping area of ​​the paths is, and the better the cleaning effect is; and / or the more the current area is determined to be dirty, the more the adjacent paths are repeatedly cleaned, so that the cleaning effect is further improved.

[0265] Specifically, when the present application detects that the dust amount of the dust box is greater than the predetermined dust amount threshold, the current cleaning area of ​​the cleaning device is deemed to be a dirty area, and the cleaning device needs to be adjusted to operate in a target operation mode of high cleaning intensity and large cleaning area. When the application detects that the dust amount of the dust box is less than the predetermined dust amount threshold, the current cleaning area of ​​the cleaning device is deemed to be a non-dirty area, and the cleaning device needs to be adjusted to operate in a target operation mode of medium-low cleaning intensity and small cleaning area.

[0266] In addition, the present application is not specifically limited to the predetermined dust amount threshold value, and may be, for example, 5 g or 10 g.

[0267] Furthermore, the present application determines a target operation mode used to represent cleaning strength and cleaning area, and then drives the cleaning device to operate the cleaning function according to the target operation mode.

[0268] For example, the rotation speed of the suction motor of the cleaning device is controlled according to the degree of dirt on the floor surface. It can be understood that the rotation speed of the suction motor is set low for areas with light dirt, and set high for areas with heavy dirt. This method not only ensures the cleaning effect, but also reduces the power consumption and extends the operating time of the cleaning device.

[0269] Alternatively, the present invention can control the overlap width of the travel path of the cleaning device and the number of cleaning passes according to the degree of dirt on the floor surface, which can enhance the cleaning effect and further guarantee the cleaning quality.

[0270] In one embodiment, the operation mode of the general cleaning strength and cleaning area can be used for the areas where the floor surface is not very dirty, so that the cleaning effect and cleaning efficiency can be guaranteed.

[0271] In this embodiment, the dust collection amount of the dust box in the cleaning device is monitored, and the target operation mode of the cleaning device is determined according to the magnitude relationship between the dust collection amount and a predetermined dust amount threshold, where the target operation mode is used to change the cleaning strength and cleaning area of ​​the cleaning device, and the cleaning device is driven in the target operation mode. By applying the technical solution of the present application, the cleaning strength and cleaning area of ​​the cleaning device can be adjusted correspondingly according to the detected dust collection amount of the dust box of the cleaning device. This realizes that the robot operates with high intensity cleaning in dirty areas and low intensity cleaning in normal areas. In addition, the cleaning effect on dirty areas can be guaranteed, and the problem of low cleaning efficiency for certain areas caused by always cleaning with the same cleaning strength and cleaning area in the related art is avoided.

[0272] Optionally, in a possible embodiment of the present application, step 033 further comprises: 0331, detecting that the dust amount is greater than a predetermined dust amount threshold, and determining to control the cleaning device to operate a target operation mode of the cleaning function at a first cleaning intensity and a first cleaning area; or 0332, detecting that the dust collection amount is not greater than a predetermined dust amount threshold and determining to control the cleaning device to operate the target operating mode of the cleaning function at a second cleaning intensity and a second cleaning area, where the second cleaning intensity is lower than the first cleaning intensity and the second cleaning area is smaller than the first cleaning area.

[0273] First, the cleaning intensity in this application can correspond to many kinds of cleaning operation parameters, such as the rotation speed of the suction motor of the cleaning device, which can be understood as the higher the rotation speed of the suction motor, the higher the cleaning intensity of the cleaning device.

[0274] Alternatively, the cleaning intensity may include the number of cleaning repetitions of the cleaning device. It can be understood that the more the number of cleaning repetitions, the higher the cleaning intensity of the cleaning device. Alternatively, the cleaning intensity may include the travel interval of the cleaning path of the cleaning device. It can be understood that the shorter the travel interval, the higher the cleaning intensity of the cleaning device.

[0275] In addition, the cleaning area of ​​the cleaning device can be controlled according to the degree of dirt on the floor surface. It can be understood that the dirtier the floor surface, the larger the predetermined cleaning area will be, and it is only necessary to set the general cleaning area for areas where the floor surface is not severely dirty.

[0276] Take an example, for example, the predetermined dust amount threshold is 5g, and when the cleaning device determines that the current dust amount is greater than 5g by monitoring the dust box, the current cleaning area can be determined as a dirty floor area, and the area can be recognized as a high cleaning area. Furthermore, the present application can select a target operation mode for operating the cleaning function with high cleaning intensity (i.e., the first cleaning intensity) and large cleaning area (i.e., the first cleaning area) through a predetermined control strategy.

[0277] Here, the target operation mode corresponding to high cleaning strength and large cleaning area may be that the rotation speed of the suction motor reaches 500 revolutions per minute, the distance between two adjacent running routes is 1 cm, the number of cleaning repetitions is 5, and the cleaning area is 1 square meter, so that the subsequent cleaning device selects the target operation mode and then operates the cleaning function according to the rotation speed of the suction motor, the running interval, the cleaning area, and the number of cleaning repetitions represented by the target operation mode.

[0278] Also, for example, the predetermined dust amount threshold is 5g, and when the cleaning device monitors the dust box to determine that the current dust amount is less than 5g (e.g., 3g), it can determine that the current cleaning area is an area with normal floor dirt, and recognize the area as a medium cleaning area.Furthermore, the present application can select a target operation mode for operating the cleaning function with a medium cleaning intensity (i.e., the second cleaning intensity) and a medium cleaning area (i.e., the second cleaning area) through a predetermined control strategy.

[0279] Here, the target operation mode corresponding to medium cleaning intensity and medium cleaning area may be that the rotation speed of the suction motor reaches 300 revolutions per minute, the distance between two adjacent travel routes is 2 cm, the number of cleaning repetitions is 3, and the cleaning area is 0.8 square meters. Therefore, after the subsequent cleaning device selects the target operation mode, it operates the cleaning function according to the rotation speed of the suction motor, the travel interval, the cleaning area, and the number of cleaning repetitions represented by the target operation mode.

[0280] Optionally, for example, the predetermined dust amount threshold is 5g, and when the cleaning device monitors the dust box to determine that the current dust amount is less than 5g (e.g., 0.5g), it can determine that the current cleaning area is a clean area and recognize the area as a low cleaning area.Furthermore, the present application can select a target operation mode for operating the cleaning function with a low cleaning intensity (i.e., the second cleaning intensity) and a low cleaning area (i.e., the second cleaning area) through a predetermined control strategy.

[0281] Here, the target operation mode corresponding to low cleaning intensity and small cleaning area may be that the rotation speed of the suction motor reaches 100 revolutions per minute, the distance between two adjacent running routes is 5 cm, the number of cleaning repetitions is 1, and the cleaning area is 0.3 square meters. Therefore, after the subsequent cleaning device selects the target operation mode, it operates the cleaning function according to the rotation speed of the suction motor, the running interval, the cleaning area, and the number of cleaning repetitions represented by the target operation mode.

[0282] Optionally, in a possible embodiment of the present application, the step of detecting that the dust collection amount is greater than a predetermined dust amount threshold and determining to control the cleaning device to operate a target operating mode of the cleaning function at a first cleaning intensity and a first cleaning area may include: determining that the dust amount is greater than the predetermined dust amount threshold and recording the location of the cleaning device; The method further includes the steps of marking an area within a predetermined range from the location as a high cleaning area, and controlling the cleaning device within the high cleaning area to operate the cleaning function at a first cleaning intensity and a first cleaning area.

[0283] In one embodiment, when the cleaning device determines that the current dust collection amount is greater than a predetermined dust amount threshold by monitoring the dust box, the present application first needs to determine the location of the dirty area to ensure that the operating range for operating the cleaning function with a first cleaning strength and a first cleaning area acts on the dirty area.

[0284] Specifically, the current location of the cleaning device needs to be recorded first. For example, the coordinate point is the center point of bedroom A. Furthermore, that is, the area within a certain range (for example, within a radius of 50 cm) from the center point position can be marked as a high-cleaning area. Then, the cleaning function is operated with a first cleaning intensity and a first cleaning area in the high-cleaning area (i.e., the area within a radius of 50 cm from the center point of bedroom A as the origin). This ensures that the target operation mode of the cleaning device acts on the dirty area.

[0285] It should be noted that the present application does not specifically limit the predetermined range, and for example, the radius may be 50 cm or 100 cm.

[0286] Optionally, in a possible embodiment of the present application, in S103 (driving and operating the cleaning device in the target operating mode), determining at least one of the rotation speed of the dust suction motor, the travel interval, the cleaning area, and the number of cleaning repetitions represented by the target operation mode; The method further includes a step of driving the cleaning device to operate a cleaning function with at least one of a rotation speed of the dust suction motor, a running interval, a cleaning area, and a number of cleaning repetitions, wherein the running interval is used to reflect the distance between two adjacent running routes of the cleaning device.

[0287] Furthermore, the present application relates to a cleaning device that adjusts the width of adjacent cleaning paths, the number of cleanings of adjacent cleaning paths, and the motor rotation speed, etc. (one or more of which) according to the degree of dirt in the current cleaning area, thereby representing the cleaning function of the operating mode.

[0288] In one embodiment, for example, the more the current area is determined to be dirty, the more the rotation speed of the dust suction motor is increased; and / or the more the current area is determined to be dirty, the narrower the width of the adjacent cleaning path is determined, and the wider the overlapping area of ​​the paths is, the better the cleaning effect is; and / or the more the current area is determined to be dirty, the more the adjacent paths can be repeatedly cleaned multiple times, and the cleaning effect is further improved.

[0289] Specifically, when the present application detects that the dust amount of the dust box is greater than a predetermined dust amount threshold, the current cleaning area of ​​the cleaning device is deemed to be a dirty area, and the cleaning device needs to be driven to operate in a target operation mode of high cleaning intensity and large cleaning area. When the application detects that the dust amount of the dust box is less than a predetermined dust amount threshold, the current cleaning area of ​​the cleaning device is deemed to be a non-dirty area, and the cleaning device needs to be driven to operate in a target operation mode of medium / low cleaning intensity and small cleaning area.

[0290] Also, for example, when it is detected that the dust collection amount is greater than a predetermined dust amount threshold 1, it can be determined to control the cleaning device to operate the target operating mode of the cleaning function at a first cleaning intensity and a first cleaning area.

[0291] When it is detected that the dust collection amount is not greater than the predetermined dust amount threshold 1 but not less than the predetermined dust amount threshold 2, it can be determined that the cleaning device is controlled to operate the target operation mode of the cleaning function at a second cleaning intensity and a second cleaning area.When it is detected that the dust collection amount is not greater than the predetermined dust amount threshold 3, it can be determined that the cleaning device is controlled to operate the target operation mode of the cleaning function at a third cleaning intensity and a third cleaning area.

[0292] It can be understood that, in this case, the predetermined dust amount threshold 1 is greater than the predetermined dust amount threshold 2, and the predetermined dust amount threshold 2 is greater than the predetermined dust amount threshold 3. The specific values ​​can be set according to actual conditions, and the present application is not limited thereto.

[0293] Optionally, the step of driving the cleaning device to operate the cleaning function at the travel interval includes: After determining the initial interval between two adjacent travel routes of the current cleaning device, adjusting the initial interval to a target interval having a smaller distance; Specifically, the method includes a step of driving the cleaning device and operating the cleaning function so that the distance between two adjacent travel routes is a target interval.

[0294] FIG. 16 shows the initial interval between two adjacent travel routes of the current cleaning device in the initial operation mode. It can be seen that it is a normal bow-shaped travel. And the initial interval between each two adjacent travel routes is large. This makes the cleaning overlap area of ​​the cleaning device small, and as a result, the cleaning intensity is not high.

[0295] To address the above problem, the present invention can adjust the initial interval to a smaller target interval, so that when the cleaning device is moving in a bow-shaped manner, the initial interval between each two adjacent moving routes is small, and the cleaning overlap area is increased, thereby achieving the purpose of strengthening the cleaning intensity.

[0296] Optionally, in a possible embodiment of the present application, in step 033: Detecting that the dust amount is greater than a predetermined dust amount threshold, and activating a photographing device of the cleaning device;

[0334] The method further includes using an imaging device to capture an area image in the travel direction of the cleaning device.

[0297] A target operation mode for the cleaning device is determined based on the area image recognition result and the magnitude relationship between the dust collection amount and a predetermined dust amount threshold.

[0298] Furthermore, the present application can also use a photographing device to collect area images in the running direction of the cleaning device in order to determine a target operating mode that is more suitable for the dirty area, and then assist in determining the target operating mode of the cleaning device based on the recognition result of the area image.

[0299] It can be understood that by recognizing the area image, the type of the object to be cleaned (for example, the object to be cleaned is dust or oil stain) existing in the current location area of ​​the cleaning device can be determined, and different cleaning strengths and target cleaning area operation modes can be selected according to different types of the object to be cleaned.

[0300] Optionally, after using the image capture device to capture an area image in the direction of travel of the cleaning device, Recognizing area object features in the area image using a predetermined image detection model, the area object features including at least one of size features, color features and contour features; The method further includes a step of determining a recognition result of the area image based on the area characteristics, where the recognition result is used to reflect objects waiting to be cleaned that exist in the area where the cleaning device is currently located.

[0301] Specifically, the present application uses an image capturing device installed on the robot to capture an image of an area in the forward running direction, obtains an area image, and inputs the area into a pre-generated image detection model to determine the object features present in the area image, so as to subsequently select an object to be cleaned that matches the object features in a set of predetermined object features.

[0302] Here, the present application does not specifically limit the image detection model. For example, it can be a Convolutional Neural Network (CNN). A Convolutional Neural Network is a type of Feedforward Neural Network with a deep structure that includes calculations, and is one of the representative algorithms of deep learning. A Convolutional Neural Network has the ability of representation learning, and can classify input information in a translation-invariant manner according to its hierarchical structure. Thanks to the strong feature representation ability of CNN (Convolutional Neural Network) for images, it has achieved notable results in fields such as image classification, target detection, and semantic segmentation.

[0303] Furthermore, the present invention uses a CNN neural network model to fetch feature information (i.e., area object features) of the object waiting to be cleaned that exists in the area image, and identifies features in the area image (e.g., including size features, color features, contour features, texture features, etc. corresponding to the object waiting to be cleaned), thereby determining which object it belongs to.

[0304] Specifically, in an embodiment of the present invention, the area image is input to a predetermined convolutional neural network model, and the output of a fully connected layer (FC), which is the final layer of the convolutional neural network model, can be used as the identification result of feature data corresponding to the area object features of the area image.

[0305] Optionally, in a possible embodiment of the present application, in step 032: 0321, includes a step of monitoring the amount of dust collected at the dust box inlet for a predetermined period of time, or a step of monitoring the weight change of the dust box for a predetermined period of time.

[0306] First, the present application does not specifically limit the predetermined period, and may be, for example, one minute or 30 seconds. Specifically, the cleaning device measures the increase in the weight of the dust box over a certain period of time, and the larger the weight, the more severe the dirt.

[0307] In one embodiment, as shown in FIG. 18, a schematic diagram of a process for operating a cleaning method of a cleaning device as presented in the present application, wherein: The cleaning device includes a step of monitoring the amount of dust collected in a dust box in the cleaning device, and determining a target operating mode of the cleaning device based on the relationship between the amount of dust collected and a predetermined dust amount threshold, wherein the target operating mode is used to change the cleaning strength and cleaning area of ​​the cleaning device, and driving the operation of the cleaning device in the target operating mode.

[0308] Here, by applying the technical means of the present application, the cleaning strength and cleaning area of ​​the cleaning device can be adjusted according to the detected amount of dust collected in the dust box of the cleaning device, so that the robot can be operated with high cleaning strength in dirty areas and low cleaning strength in normal areas, and the cleaning effect on dirty areas can be guaranteed, and the problem of low cleaning efficiency in certain areas caused by always cleaning with the same cleaning strength and cleaning area in the related art can be avoided.

[0309] FIG. 19 shows a schematic block diagram of a cleaning device according to an embodiment of the present invention. The cleaning device 1000 includes: A location planning module 1002 used to determine the target cleaning area; a trajectory planning module 1003 used to generate a cleaning trajectory within the target cleaning area in response to a path planning command; A first cleaning module 1004 is included which is adapted to perform cleaning operations along the cleaning trajectory.

[0310] In some embodiments, the location planning module 1002 is used to determine a target cleaning area for the cleaning device after activating the cleaning device if it recognizes that the cleaning device is activated in a non-charging location.

[0311] In this embodiment, the cleaning device is activated in the non-charging position when the user carries the cleaning device to a certain position and manually activates it, at which time the cleaning device enters an operating mode for spot cleaning.

[0312] Specifically, after the cleaning device is started, the position planning module 1002 first recognizes whether the starting position of the cleaning device is a non-charging position. Specifically, after the cleaning device is in an awake state, the position recognition component in the cleaning device is also started accordingly, the position recognition component can detect whether the cleaning device is in a charging position, and the position planning module 1002 can determine whether the starting position of the cleaning device is a non-charging position according to the recognition result of the position recognition component.

[0313] Furthermore, when the location planning module 1002 determines that the cleaning device is to be activated in a non-charging location, i.e., when the cleaning device is transported to a location and manually activated, the location planning module 1002 needs to determine a target cleaning area where the cleaning device needs to clean.

[0314] Specifically, when the position planning module 1002 determines that the cleaning device is activated in a non-charging position, it indicates that the cleaning device is in a local cleaning mode, and at this time, it can be determined that the user's cleaning intention is to clean a certain area, and therefore the position planning module 1002 needs to determine the target cleaning area, and in subsequent steps, clean accurately according to the user's cleaning intention.

[0315] Specifically, a sensing component is provided in the cleaning device, which is used to sense situations near the location where the cleaning device is located, and a semantic map of all scenes is stored in the cleaning device, and the location planning module 1002 can analyze the user's cleaning intention based on the sensing result of the sensing component and / or the semantic map, and further determine a target cleaning area for the cleaning device based on the user's cleaning intention.

[0316] In this embodiment, the position planning module 1002 first determines whether the cleaning device is in a state to be activated in a non-charging position, and when the cleaning device is in a state to be activated in a non-charging position, determines the area that the user wants to clean (i.e., the above-mentioned target cleaning area). In an embodiment of the present invention, when the position planning module 1002 determines that the cleaning device is in a state to be activated in a non-charging position, it can automatically analyze the user's cleaning intention and determine the area that the user wants to clean.

[0317] In the above embodiment, the location planning module 1002 is used to recognize whether a first cleaning target exists within an area within a predetermined distance from the cleaning device, and if the first cleaning target exists, to determine the area where the first cleaning target is located as the target cleaning area.

[0318] In this embodiment, the predetermined distance refers to a distance that can be recognized by the sensing component of the cleaning device, and can be a custom distance. It can be understood that the custom distance needs to be within the range that can be recognized by the sensing component of the cleaning device, and the first cleaning target refers to dirt or dust that is not easy to clean.

[0319] Specifically, in the process of determining the target cleaning area of ​​the cleaning device, the location planning module 1002 first recognizes whether a first cleaning target exists in an area within a predetermined distance from the cleaning device. Specifically, a sensing component is provided on the cleaning device, which can be used to sense the situation near the location where the cleaning device is located, and the location planning module 1002 can recognize the situation near the location where the cleaning device is located through the sensing component, that is, recognize whether the first cleaning target exists near the cleaning device.

[0320] By way of example, the sensing component may be one of the following: an AI (image) recognition device, a camera, a video camera, etc., but is not limited thereto.

[0321] Furthermore, when the location planning module 1002 determines that a first cleaning target exists within an area within a predetermined distance from the cleaning device, the location planning module 1002 determines the area in which the first cleaning target is located as the area the user desires to clean, i.e., determines the area in which the first cleaning target is located as the target cleaning area.

[0322] Specifically, when the position planning module 1002 recognizes that a first cleaning target exists within an area at a predetermined distance from the cleaning device, the user's cleaning intention is to clean the first cleaning target, and at this time, the position planning module 1002 determines the area where the first cleaning target is located as the target cleaning area, so that in a subsequent step, the cleaning device only controls the cleaning device to clean the area where the first cleaning target is located, so that the cleaning area of ​​the cleaning device is more in line with the user's requirements.

[0323] In this embodiment, when the location planning module 1002 recognizes that a first cleaning target exists within an area within a predetermined distance from the cleaning device, it determines the area in which the first cleaning target is located as the target cleaning area, i.e., determines the area in which the first cleaning target is located as the area that the user wants to clean.

[0324] In the above embodiment, the cleaning device 1000 of the above cleaning device includes an acquisition module, which is used to acquire a semantic map of all scenes constructed by the cleaning device, and a position planning module 1002 is used to determine a target cleaning area according to the semantic map.

[0325] In this embodiment, the semantic map refers to a semantic map of all scenes that the cleaning device builds after cleaning all scenes, where all scenes refer to scenes of all rooms in the user's home, and the semantic map specifically includes room types and the locations and types of items such as home appliances and furniture.

[0326] Specifically, the process of determining the target cleaning area of ​​the cleaning device is to first obtain the semantic map constructed by the cleaning device after cleaning all scenes through the acquisition module. Specifically, the cleaning device constructs a semantic map of all scenes in the user's house after cleaning all rooms in the user's house for the first time, and updates the semantic map according to the actual situation after each cleaning, and stores the updated semantic map in the storage unit of the cleaning device. The acquisition module can obtain the semantic map through the storage unit. In addition, the acquisition module obtains the updated semantic map after the latest cleaning, and then in the subsequent steps, the location planning module 1002 ensures the accuracy of the target cleaning area determined by the semantic map.

[0327] Furthermore, the location planning module 1002 determines the target cleaning area according to the semantic map, i.e., determines the area the user wishes to clean according to the semantic map. Specifically, the semantic map can determine the room type of the current location of the cleaning device and whether furniture or home appliances are present at the current location of the cleaning device, and the user's cleaning intention can be analyzed according to the information, and the user's cleaning intention can determine the area the user wishes to clean. Therefore, the location planning module 1002 can determine the target cleaning area according to the semantic map.

[0328] In this embodiment, when it is determined that the cleaning device is in a state to be activated in a non-charging position, i.e., when the cleaning device activates a local cleaning mode, the acquisition module can acquire the semantic map constructed by the cleaning device, and the location planning module 1002 analyzes the situation near the current location of the cleaning device using the semantic map, i.e., analyzes the user's cleaning intention, and determines the target cleaning area according to the user's cleaning intention, and in a subsequent step ensures that the area to be cleaned by the cleaning device is the area desired by the user to be cleaned, so that the cleaning area of ​​the cleaning device meets the user's requirements, improves the cleaning efficiency of the cleaning device, and improves user satisfaction.

[0329] In the above embodiment, the location planning module 1002 is used to determine the predetermined type of room as the target cleaning area if the semantic map determines that the cleaning device is in a predetermined type of room.

[0330] In this embodiment, the predetermined type of room refers to a room that is not easy to clean or is easily soiled, such as a toilet or a kitchen.

[0331] Specifically, in the process of determining the target cleaning area through the semantic map, the location planning module 1002 first analyzes whether the room type of the location where the cleaning device is currently located is a predetermined room type through the semantic map.

[0332] Furthermore, if the location planning module 1002 determines that the room type of the location where the cleaning device is currently located is a predetermined room type, the location planning module 1002 determines the room as the area the user wishes to clean, i.e., determines the room as the target cleaning area.

[0333] Specifically, when the location planning module 1002 analyzes that the room type of the location where the cleaning device is currently located is a predetermined room type through the semantic map, it indicates that the user's cleaning intention is to clean a room of the predetermined type, and in this case, the cleaning device determines the room of the predetermined type (i.e., the room where the cleaning device is currently located) as the target cleaning area.

[0334] In this embodiment, if the location planning module 1002 analyzes that the current location of the cleaning device is a room of a certain type based on the semantic map, it determines the room of the certain type as the area that the user wants to clean, i.e., determines the room of the certain type as the target cleaning area.

[0335] In the above embodiment, when the semantic map determines that the cleaning device is not present in a room of a specified type and a second cleaning target is present within an area at a specified distance from the cleaning device, the location planning module 1002 is further used to determine the area where the second cleaning target is located as the target cleaning area.

[0336] In this embodiment, the second cleaning target refers to the appliances or furniture in the user's room that are likely to interrupt the cleaning process of the cleaning device, such as a bed, a tea table, and a washing machine.

[0337] Specifically, the process of determining the target cleaning area by the semantic map is as follows: if the location planning module 1002 analyzes through the semantic map that the room type of the location where the cleaning device is currently located is not a predetermined room type, the semantic map analyzes whether the second cleaning target exists within an area of ​​a predetermined distance from the cleaning device. Specifically, the second cleaning target can be defined by the semantic map, so that the location planning module 1002 does not need to call the detection component of the cleaning device to detect the situation near the cleaning device, and can determine whether the second cleaning target exists within an area of ​​a predetermined distance from the cleaning device by the semantic map.

[0338] Furthermore, when the location planning module 1002 analyzes that a second cleaning target exists within an area within a predetermined distance from the cleaning device based on the semantic map, the location planning module 1002 determines the area where the second cleaning target is located as the area the user desires to clean, i.e., determines the area where the second cleaning target is located as the target cleaning area.

[0339] Specifically, when the location planning module 1002 analyzes through the semantic map that the room type of the location where the cleaning device is currently located is not the specified room type and there is a second cleaning target within an area at a specified distance from the cleaning device, it indicates that the user's cleaning intention is to clean the area where the second cleaning target is located, and at this time, the cleaning device determines the area where the second cleaning target is located as the target cleaning area, and in a subsequent step, the first cleaning module 1004 controls only the cleaning device to clean the area where the second cleaning target is located, so that the cleaning area of ​​the cleaning device is more in line with the user's requirements.

[0340] In this embodiment, if the location planning module 1002 analyzes through the semantic map that the location of the cleaning device is not a predetermined type of room and there is a second cleaning target within an area at a predetermined distance from the cleaning device, it determines the area where the second cleaning target is located as the area that the user wants to clean, that is, determines the area where the second cleaning target is located as the target cleaning area. Then, in a subsequent step, the first cleaning module 1004 only controls the cleaning device to clean the area where the second cleaning target is located, so that the cleaning area of ​​the cleaning device meets the user's requirements, improves the cleaning efficiency of the cleaning device, and improves the user's satisfaction.

[0341] In the above embodiment, if the semantic map determines that the cleaning device is not present in a room of a predetermined type and that there is no second cleaning target within an area within a predetermined distance from the cleaning device, the location planning module 1002 is further used to determine an area within a predetermined distance from the cleaning device as a target cleaning area.

[0342] In this embodiment, in the process of determining the target cleaning area by the semantic map, the location planning module 1002 first analyzes, by the semantic map, whether the room type of the location where the cleaning device is currently located is a predetermined room type, and whether there is a second cleaning target within an area within a predetermined distance from the cleaning device.

[0343] Specifically, if the location planning module 1002 analyzes through the semantic map that the room type of the location where the cleaning device is currently located is not a predetermined room type and a second cleaning target exists within an area within a predetermined distance from the cleaning device, the location planning module 1002 determines the area within a predetermined distance from the cleaning device as the area the user wants to clean, i.e., determines the area within a predetermined distance from the cleaning device as the target cleaning area.

[0344] Specifically, if the location planning module 1002 analyzes through the semantic map that the room type of the location where the cleaning device is currently located is not the specified room type and there is no second cleaning target within the area at a specified distance from the cleaning device, it indicates that the user's cleaning intention is to clean the area at a specified distance from the cleaning device, and at this time, the cleaning device determines the area at a specified distance from the cleaning device as the target cleaning area, and in a subsequent step, the first cleaning module 1004 only controls the cleaning device to clean the area at a specified distance from the cleaning device, so that the cleaning area of ​​the cleaning device better meets the user's requirements.

[0345] In this embodiment, if the location planning module 1002 analyzes through the semantic map that the location of the cleaning device is not a predetermined type of room and there is no second cleaning target within the area within a predetermined distance from the cleaning device, it determines the area within a predetermined distance from the cleaning device as the area that the user wants to clean, i.e., determines the area within a predetermined distance from the cleaning device as the target cleaning area. Then, in a subsequent step, the first cleaning module 1004 only controls the cleaning device to clean the area within a predetermined distance from the cleaning device, so that the cleaning area of ​​the cleaning device is more in line with the user's requirements, improves the cleaning efficiency of the cleaning device, and improves the user's satisfaction.

[0346] In the above embodiment, the location planning module 1002 is further used to recognize whether a first cleaning target exists within an area at a predetermined distance from the cleaning device, and if the first cleaning target does not exist, to determine the target cleaning area by the semantic map.

[0347] In this embodiment, before the location planning module 1002 determines the target cleaning area through the semantic map, the location planning module 1002 also needs to recognize whether a first cleaning target exists in an area within a predetermined distance from the cleaning device. Specifically, a sensing component is provided on the cleaning device, which can be used to sense a situation near the location where the cleaning device is located, and the location planning module 1002 can recognize the situation near the location where the cleaning device is located through the sensing component, i.e., recognize whether the first cleaning target exists near the cleaning device.

[0348] Further, when the location planning module 1002 determines that the first cleaning target is not present within an area at a predetermined distance from the cleaning device, the location planning module 1002 determines that it can perform a step of determining the target cleaning area by the semantic map.

[0349] Specifically, when the location planning module 1002 recognizes that the first cleaning target does not exist within an area of ​​a predetermined distance from the cleaning device, the user's cleaning intention is not represented as dirt or debris near the cleaning device, and at this time, the location planning module 1002 can perform steps such as analyzing through the semantic map that the user's specific cleaning intention is to clean a certain type of room, to clean near the second cleaning target, or to clean an area near the cleaning device.

[0350] In this embodiment, before the step of determining the target cleaning area of ​​the cleaning device through the semantic map, the location planning module 1002 also needs to perform a step of recognizing whether the first target cleaning area exists within an area of ​​a predetermined distance from the cleaning device, so that the user's cleaning intention can be more accurately determined, and by determining a more accurate target cleaning area, the cleaning area of ​​the cleaning device can be more in line with the user's requirements, improving the cleaning efficiency of the cleaning device and improving the user's experience.

[0351] In this embodiment, the first cleaning module 1004 controls the cleaning device to clean the target cleaning area using a cleaning method of cleaning along the edge and then cleaning in a bow shape, thereby ensuring the cleanliness of the target cleaning area, achieving the required cleaning effect, and improving user satisfaction.

[0352] In the above embodiment, the first cleaning module 1004 is further used to audio-suggest the cleaning method of the cleaning device for the target cleaning area.

[0353] In this embodiment, before controlling the cleaning device to clean the target cleaning area, the first cleaning module 1004 needs to play audio prompting information to inform the user how the first cleaning module 1004 will control the cleaning device to clean the target cleaning area.

[0354] Specifically, before the cleaning device cleans the target cleaning area, the first cleaning module 1004 plays audio prompting information so that the user can understand the area to be cleaned. For example, the area to be cleaned by the cleaning device is the area where the tea table is located, and the first cleaning module 1004 plays audio prompting information such as “The area under the tea table will be cleaned soon”, so that the user can confirm whether the area to be cleaned by the cleaning device is the area the user wants to clean.

[0355] In this embodiment, before controlling the cleaning device to clean the target cleaning area, the first cleaning module 1004 plays voice prompting information of the area to be cleaned, so that the user can confirm whether the area to be cleaned by the cleaning device is the area the user wants to clean, and if the cleaning area played by the voice prompting information is not the area the user wants to clean, the cleaning area of ​​the cleaning device can be adjusted in a timely manner.

[0356] Based on the same technical concept, the trajectory planning module 1003 of the cleaning device in this embodiment of the present application: in response to the path planning command, determining a current cleanliness state of the mop and a start point and an end point corresponding to the path planning command; and determining a cleaning trajectory from the starting point to the end point according to a geographical situation map, a cleaning situation map and the cleanliness status.

[0357] In a possible embodiment, the trajectory planning module 1003 specifically includes: It is used to determine that the current cleanliness status of the mop is clean when the current cleaning stage of the cleaning device is the departure stage after self-cleaning at the water station, and to determine that the current cleanliness status of the mop is dirty when the current cleaning stage of the cleaning device is the stage of moving from the area where the current cleaning is completed to the waiting area for cleaning, or the stage of returning to the water station to wash the mop.

[0358] In a possible embodiment, the trajectory planning module 1003 specifically includes: The geographical situation map is used to determine an obstacle-free area, and the obstacle-free area and the cleaning situation map are used to determine a cleaning trajectory from the start point to the end point according to a predetermined traffic strategy corresponding to the cleanliness state.

[0359] In a possible embodiment, the step of determining a cleaning trajectory from the starting point to the end point according to the obstacle-free area and the cleaning status map and according to a predetermined traffic strategy corresponding to the clean state includes: The method includes the steps of: determining all currently cleaned areas from the cleaning status map based on the cleanliness state being clean; determining all areas between the starting point and the end point that are free of obstacles and have been cleaned based on the areas free of obstacles and the currently cleaned areas; determining that at least one continuous movement trajectory can be constructed between the starting point, all areas that are free of obstacles and have been cleaned, and the end point, and determining the shortest movement trajectory among the at least one movement trajectory as the cleaning trajectory from the starting point to the end point.

[0360] In a possible embodiment, the step of determining a cleaning trajectory from the starting point to the end point according to the obstacle-free area and the cleaning status map and according to a predetermined traffic strategy corresponding to the clean state includes: The method includes the steps of: determining all areas waiting to be cleaned from the cleaning status map based on the clean state being dirty; determining all areas between the starting point and the end point that are free of obstacles and waiting to be cleaned using the areas free of obstacles and all areas waiting to be cleaned; determining that at least one continuous movement trajectory can be constructed between the starting point, all areas free of obstacles and waiting to be cleaned, and the end point; and determining the shortest movement trajectory among the at least one movement trajectory as the cleaning trajectory from the starting point to the end point.

[0361] In a possible embodiment, the trajectory planning module 1003 includes: It may further be used to determine that the continuous moving trajectory cannot be constructed, and determine a cleaning trajectory from the starting point to the end point according to the obstacle-free area.

[0362] In a possible embodiment, the cleaning device comprises: The apparatus further includes a division module, which is used for dividing the target cleaning area into a plurality of areas and determining a cleaning sequence for each area before responding to a path planning command.

[0363] In a possible embodiment, the division module, in particular According to the current cleaning stage of the cleaning device and the cleaning sequence of each area, the starting point and the end point corresponding to the path planning command are determined.

[0364] Referring to FIG. 20, the cleaning device according to the embodiment of the present invention is A second cleaning module 1005 is used to mark obstacles in the target cleaning area, avoid the obstacles in the target cleaning area, and perform cleaning operations; A dynamic detection module 1006 is used for detecting the status of obstacles in the target cleaning area and determining the obstacles whose status has changed; and a third cleaning module 1007 adapted to perform a cleaning action at the location of the obstacle whose state has changed.

[0365] In the cleaning device according to the embodiment of the present invention, the position planning module 301 first determines the target cleaning area that needs to be cleaned, and the second cleaning module 1005 can perform cleaning operations within the target cleaning area while avoiding obstacles and marking the obstacles. The dynamic detection module 1006 can detect movable obstacles within the target cleaning area, and when there are movable objects in the scene, situations in which the user carries objects and situations in which the door is opened and closed, the third cleaning module 1007 is used to make the executing subject move to the position of the obstacle whose state has changed and perform the cleaning operation, thereby avoiding cleaning omissions caused by object movement and improving the coverage rate of automatic cleaning in dynamic scenes.

[0366] It can be understood that the step of marking obstacles in the target cleaning area in the second cleaning module 1005 can include: Marking the location, type and shape of obstacles within the target cleaning area.

[0367] In addition, in order to realize the cleaning device of the embodiment of the present invention, the cleaning device must at least include an RGB camera, a laser lidar, a high-performance GPU, an AI chip, and a memory. For example, high-resolution cameras are installed at 90-degree intervals at the edge positions of the cleaning device to capture real-time visual scenes. A high-performance GPU is installed in the cleaning device to perform feature fetching and image recognition on the visual scene images captured in real time. An AI chip is installed in the high-performance GPU, and a target recognition algorithm is integrated in the chip. A memory is installed in the cleaning device to store a map of the target cleaning area and a set of obstacles.

[0368] The GPU can realize obstacle state management, AI recognition and navigation planning functions.

[0369] Specifically, the RGB camera can recognize low obstacles such as socks and power lines, while the laser lidar can recognize high obstacles such as furniture and trash cans.

[0370] It can be understood that the step of detecting the state of an obstacle in the target cleaning area and determining the obstacle whose state has changed in the dynamic detection module 1006 includes: Detecting the location and shape of an obstacle within a target cleaning area; and determining which obstacles have changed status based on the obstacles whose position or shape have changed.

[0371] It can be understood that the cleaning operation includes a first cleaning operation and a second cleaning operation, wherein the first cleaning trajectory determined by performing the first cleaning operation is a closed loop, and the second cleaning trajectory determined by performing the second cleaning operation is used to traverse the closed loop.

[0372] It can be understood that the step of determining the target cleaning area in the location planning module 301 includes: performing a first cleaning operation and determining a target cleaning area based on a cleaning trajectory of the first cleaning operation.

[0373] It can be understood that the step of avoiding obstacles and performing cleaning operations in the target cleaning area in the third cleaning module 1007 includes: performing a second cleaning operation within the target cleaning area; performing a first cleaning action when it is determined that an obstacle has been encountered, and continuing to perform a second cleaning action after avoiding the obstacle.

[0374] In another embodiment of the present application, as shown in FIG. 21, the first cleaning module 1004 of the present application: a dust amount monitoring unit 10041 configured to monitor a dust amount of a dust box of the cleaning device; A unit 10042 is configured to determine a target operating mode of the cleaning device according to a magnitude relationship between the dust collection amount and a predetermined dust amount threshold, where the target operating mode is used to change the cleaning strength and cleaning area of ​​the cleaning device; and a drive unit 10043 configured to drive and operate the cleaning device in the target operating mode.

[0375] In this application, the dust collection amount of the dust box in the cleaning device is monitored, and the target operation mode of the cleaning device is determined based on the magnitude relationship between the dust collection amount and a predetermined dust amount threshold, where the target operation mode is used to change the cleaning strength and cleaning area of ​​the cleaning device, and the operation of the cleaning device is driven in the target operation mode. By applying the technical means of this application, the cleaning strength and cleaning area of ​​the cleaning device can be adjusted correspondingly according to the detected dust collection amount of the dust box of the cleaning device. This realizes operating the robot with high-intensity cleaning in dirty areas and low-intensity cleaning in normal areas. In addition, the cleaning effect on dirty areas can be guaranteed, and the problem of low cleaning efficiency for certain areas caused by always cleaning with the same cleaning strength and cleaning area in the related art is avoided.

[0376] In another embodiment of the present application, the unit 10042 for determining the operation mode includes: Detecting that the dust amount is greater than the predetermined dust amount threshold, and determining to control the cleaning device to operate a target operating mode of a cleaning function at a first cleaning intensity and a first cleaning area; or and determining to control the cleaning device to operate a target operating mode of a cleaning function at a second cleaning intensity and a second cleaning area when the dust collection amount is detected to be not greater than the predetermined dust amount threshold, where the second cleaning intensity is lower than the first cleaning intensity and the second cleaning area is smaller than the first cleaning area.

[0377] In another embodiment of the present application, the unit 10042 for determining the operation mode includes: determining that the dust amount is greater than the predetermined dust amount threshold and recording the location of the cleaning device; and controlling the cleaning device to mark an area within a predetermined range from the location as a high cleaning area and to operate a cleaning function at a first cleaning intensity and a first cleaning area within the high cleaning area.

[0378] In another embodiment of the present application, the unit 10042 for determining the operation mode includes: determining at least one of the rotation speed of a suction motor, a travel interval, a cleaning area, and a number of cleaning repetitions represented by the target operation mode; A step of driving the cleaning device and operating a cleaning function with at least one of a rotation speed of a dust suction motor, a running interval, a cleaning area, and a number of cleaning repetitions is set, and the running interval is used to reflect the distance between two adjacent running routes of the cleaning device.

[0379] In another embodiment of the present application, the unit 10042 for determining the operation mode includes: determining an initial interval between two adjacent travel routes of the current cleaning device; adjusting the initial interval to a target interval whose distance is less than the initial interval; and driving the cleaning device to operate a cleaning function such that the distance between two adjacent travel routes is the target interval.

[0380] In another embodiment of the present application, the unit 10042 for determining the operation mode includes: detecting that the dust collection amount is greater than the predetermined dust amount threshold and activating a photographing device of the cleaning device; Using the photographing device, capture an area image in a direction of travel of the cleaning device; A target operation mode of the cleaning device is determined based on the recognition result of the area image and the magnitude relationship between the dust collection amount and a predetermined dust amount threshold.

[0381] In another embodiment of the present application, the unit 10042 for determining the operation mode includes: Using a predetermined image detection model, recognizing area object features in the area image, the area object features including at least one of size features, color features and contour features; The method further comprises determining a recognition result of the area image based on the area object features, wherein the recognition result is used to reflect objects waiting to be cleaned that exist in the area where the cleaning device is currently located.

[0382] In another embodiment of the present application, the unit 10041 for monitoring the dust collection amount includes: The amount of dust collected at the entrance of the dust box during a predetermined period of time is monitored, or the change in weight of the dust box during a predetermined period of time is monitored.

[0383] 22 is a logic structure block diagram of a cleaning device according to an exemplary embodiment. For example, the cleaning device 1100 may be a cleaning device with a drying function, such as a dryer or a washing machine.

[0384] The cleaning device includes a memory 1102 for storing a program or instruction, and a processor 1104 for executing the program or instruction stored in the memory 1102 to realize the steps of the cleaning method according to the above embodiment of the present invention, and thus has all the beneficial technical effects of the cleaning method according to the above embodiment of the present invention, and detailed description thereof will be omitted here.

[0385] The present invention provides a readable storage medium, which stores a program or instruction, and when the program or instruction is executed by a processor, realizes the cleaning method provided by the above embodiment of the present invention. Therefore, the readable storage medium has all the beneficial technical effects of the cleaning method provided by the above embodiment of the present invention, and detailed description is omitted here.

[0386] Optionally, the instructions are executed by a processor of the cleaning device to complete other steps according to the exemplary embodiment. For example, the non-transitory computing device readable storage medium may be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

Claims

1. A cleaning method for use in a cleaning device, comprising: determining a target cleaning area; generating a cleaning trajectory within the target cleaning area in response to a path planning command; performing a cleaning action along the cleaning locus.

2. The step of determining the target cleaning area includes: The cleaning method of claim 1 , further comprising, after activating the cleaning device, determining a target cleaning area for the cleaning device if it is determined that the cleaning device is activated in a non-charging position.

3. The step of determining a target cleaning area for the cleaning device includes: The cleaning method according to claim 2, further comprising a step of recognizing whether a first cleaning target exists within an area at a predetermined distance from the cleaning device, and, if the first cleaning target exists, determining the area in which the first cleaning target is located as the target cleaning area.

4. The step of determining a target cleaning area for the cleaning device includes: obtaining a semantic map of all scenes constructed by the cleaning device; and determining the target cleaning area by the semantic map.

5. The step of determining the target cleaning area by the semantic map includes:

5. The method of claim 4, further comprising the step of: if the semantic map determines that the cleaning device is in a room of a predetermined type, determining the room of the predetermined type as the target cleaning area.

6. The step of determining the target cleaning area by the semantic map includes: The cleaning method according to claim 4, further comprising a step of determining, when the semantic map determines that the cleaning device is not present in a room of a specified type and that a second cleaning target is present within an area at a specified distance from the cleaning device, the area in which the second cleaning target is located as the target cleaning area.

7. The step of determining the target cleaning area by the semantic map includes:

5. The cleaning method of claim 4, further comprising a step of determining an area a predetermined distance from the cleaning device as the target cleaning area when the semantic map determines that the cleaning device is not present in a room of a predetermined type and that a second cleaning target does not exist within an area a predetermined distance from the cleaning device.

8. Before determining the target cleaning area by the semantic map, The cleaning method according to claim 4, further comprising: recognizing whether a first cleaning target exists within an area of ​​a predetermined distance from the cleaning device; and, if the first cleaning target does not exist, confirming that a step of determining a target cleaning area is performed by the semantic map.

9. The step of performing a cleaning operation along the cleaning locus includes: The cleaning method according to any one of claims 1 to 8, further comprising a step of cleaning the target cleaning area using a cleaning method in which cleaning is performed along the edge and then in a bow-shaped manner, and returning to a charging position after cleaning is completed and charging.

10. Before performing a cleaning operation along the cleaning locus, The cleaning method according to any one of claims 1 to 8, further comprising a step of audibly instructing the cleaning device on a cleaning strategy for the target cleaning area.

11. 9. The cleaning method according to claim 3 or 8, wherein the first cleaning target includes any one of dirt, small granular dust, and hair-like dust.

12. The cleaning method according to claim 6 or 7, wherein the second cleaning target is an electrical appliance or furniture.

13. The step of generating a cleaning trajectory within the target cleaning area in response to a path planning command includes: in response to a path planning command, determining a current cleanliness state of the mop and a start point and an end point corresponding to said path planning command; The cleaning method according to claim 1 , further comprising: determining a cleaning trajectory from the starting point to the end point according to a geographical situation map, a cleaning situation map and the cleanliness state.

14. The step of determining a current cleanliness state of the mop comprises: determining that the current cleanliness state of the mop is clean if the current cleaning stage of the cleaning device is a starting stage after self-cleaning at the water station; and determining that the current cleanliness state of the mop is dirty if the current cleaning stage of the cleaning device is moving from a currently cleaned area to a waiting area for cleaning or returning to the water station to wash the mop.

15. determining a cleaning trajectory from the starting point to the end point according to a geographical situation map, a cleaning situation map, and the cleanliness state, determining an obstacle-free area according to the geographical situation map; and determining a cleaning trajectory from the starting point to the end point according to the obstacle-free area and the cleaning status map, according to a predetermined traffic strategy corresponding to the clean state.

16. determining a cleaning trajectory from the starting point to the end point according to the obstacle-free area and the cleaning status map, according to a predetermined traffic strategy corresponding to the clean state; determining a currently cleaned area from the cleaning status map based on the cleanliness being clean; determining all areas between the start point and the end point that are free of obstacles and have been cleaned based on the obstacle-free areas and the currently cleaned areas; The cleaning method according to claim 15, further comprising a step of determining that at least one continuous movement trajectory can be constructed between the starting point, all areas that are free of obstacles and have been cleaned, and the end point, and determining the shortest movement trajectory among the at least one movement trajectory as the cleaning trajectory from the starting point to the end point.

17. determining a cleaning trajectory from the starting point to the end point according to the obstacle-free area and the cleaning status map, according to a predetermined traffic strategy corresponding to the clean state; determining all areas waiting to be cleaned from the cleaning status map based on the cleanliness status being dirty; determining all areas between the start point and the end point that are free of obstacles and waiting to be cleaned based on the obstacle-free areas and all areas waiting to be cleaned; The cleaning method according to claim 15, further comprising a step of determining that at least one continuous movement trajectory can be constructed between the starting point, all areas free of obstacles and waiting to be cleaned, and the end point, and determining the shortest movement trajectory among the at least one movement trajectory as the cleaning trajectory from the starting point to the end point.

18. The cleaning method includes: The cleaning method according to claim 16 or 17, further comprising the step of determining that the continuous movement trajectory cannot be constructed, and determining a cleaning trajectory from the starting point to the end point based on the obstacle-free area.

19. Before responding to a route planning command, The cleaning method according to any one of claims 13 to 17, further comprising the step of dividing the target cleaning area into a plurality of areas and determining a cleaning order for each area.

20. The step of determining a start point and an end point corresponding to the path planning command includes: The cleaning method of claim 19, further comprising: determining a start point and an end point corresponding to the path planning command according to a current cleaning stage of a cleaning device and a cleaning sequence of each area.

21. The cleaning method includes: marking obstacles within the target cleaning area and performing a cleaning operation to avoid the obstacles within the target cleaning area; Detecting the state of obstacles in the target cleaning area and determining which obstacles have changed state; 2. The method of claim 1, further comprising the step of: performing a cleaning action at the location of the obstacle whose state has changed.

22. The step of marking obstacles within the target cleaning area includes:

22. The method of claim 21, comprising marking the location, type and shape of obstacles within the target cleaning area, the types including dynamic obstacles.

23. The step of detecting the state of an obstacle in the target cleaning area and determining an obstacle whose state has changed includes: detecting a position and shape of the moving obstacle within the target cleaning area; and determining which dynamic obstacles have changed state based on which dynamic obstacles have changed position or shape.

24. 24. The cleaning method according to claim 21, wherein the cleaning action includes a first cleaning action and a second cleaning action, wherein a first cleaning trajectory determined by performing the first cleaning action is a closed loop, and a second cleaning trajectory determined by performing the second cleaning action can fill the target cleaning area.

25. The step of determining the target cleaning area includes:

25. The method of claim 24, further comprising: performing the first cleaning operation; and determining the target cleaning area based on a cleaning trajectory of the first cleaning operation.

26. The step of performing a cleaning operation while avoiding the obstacles in the target cleaning area includes: performing a second cleaning operation within the target cleaning area; 25. The method of claim 24, further comprising the steps of: performing a first cleaning action when it is determined that the obstacle has been encountered; and continuing to perform a second cleaning action after avoiding the obstacle.

27. The step of performing a cleaning operation along the cleaning locus includes: monitoring the amount of dust collected in a dust box of the cleaning device; determining a target operation mode of the cleaning device based on a magnitude relationship between the dust collection amount and a predetermined dust amount threshold, where the target operation mode is used to change a cleaning strength and a cleaning area of ​​the cleaning device; and performing the cleaning operation by driving the cleaning device to operate in the target operating mode.

28. determining a target operation mode of the cleaning device based on a magnitude relationship between the dust collection amount and a predetermined dust amount threshold, Detecting that the dust amount is greater than the predetermined dust amount threshold and determining to control the cleaning device to operate a target operating mode of a cleaning function at a first cleaning intensity and a first cleaning area; or 28. The cleaning method of claim 27, further comprising: detecting that the dust collection amount is not greater than the predetermined dust amount threshold and determining to control the cleaning device to operate a target operating mode of a cleaning function at a second cleaning intensity and a second cleaning area, wherein the second cleaning intensity is lower than the first cleaning intensity and the second cleaning area is smaller than the first cleaning area.

29. The step of detecting that the dust collection amount is greater than the predetermined dust amount threshold and determining to control the cleaning device to operate a target operation mode of a cleaning function at a first cleaning intensity and a first cleaning area includes: determining that the dust amount is greater than the predetermined dust amount threshold and recording the location of the cleaning device; 30. The cleaning method of claim 28, further comprising the steps of: marking an area within a predetermined range from the location as a high cleaning area; and controlling the cleaning device to operate a cleaning function at a first cleaning intensity and the first cleaning area in the high cleaning area.

30. The step of driving and operating the cleaning device in the target operation mode includes: determining at least one of the following items represented by the target operation mode: a rotation speed of a suction motor, a travel interval, a cleaning area, and a number of cleaning repetitions; The cleaning method according to claim 27 or 28, further comprising: driving the cleaning device to operate a cleaning function with at least one of a rotation speed of the dust suction motor, a running interval, a cleaning area, and a number of cleaning repetitions, wherein the running interval is used to reflect the distance between two adjacent running routes of the cleaning device.

31. The step of driving the cleaning device to perform a cleaning function at the travel interval includes: determining an initial interval between two adjacent running routes of the current cleaning device; adjusting the initial interval to a target interval whose distance is less than the initial interval; The cleaning method according to claim 30, further comprising a step of driving the cleaning device to operate a cleaning function such that a distance between two adjacent travel routes becomes the target interval.

32. determining a target operation mode of the cleaning device based on a magnitude relationship between the dust collection amount and a predetermined dust amount threshold, detecting that the dust amount is greater than the predetermined dust amount threshold and activating a photographing device of the cleaning device; Using the photographing device to capture an area image in a travel direction of the cleaning device; The cleaning method according to claim 27, further comprising a step of determining a target operation mode of the cleaning device based on a result of the recognition of the area image and a magnitude relationship between the dust collection amount and a predetermined dust amount threshold.

33. After the step of using the photographing device to capture an area image in the travel direction of the cleaning device, Recognizing area object features in the area image using a predetermined image detection model, the area object features including at least one of a size feature, a color feature, and a contour feature; The cleaning method of claim 32, further comprising: determining a recognition result of the area image based on the area object characteristics, wherein the recognition result is used to reflect objects waiting to be cleaned that exist in an area where a cleaning device is currently located.

34. The step of monitoring the amount of dust collected in the dust box of the cleaning device includes:

28. The cleaning method according to claim 27, further comprising the step of monitoring the amount of dust collected at the inlet of the dust box for a predetermined period of time, or the step of monitoring a change in weight of the dust box for a predetermined period of time.

35. a location planning module used to determine a target cleaning area; a trajectory planning module adapted to generate a cleaning trajectory within the target cleaning area in response to a path planning command; a cleaning module adapted to perform a cleaning operation along the cleaning locus.

36. A cleaning device comprising a memory, a processor and a computer program stored in the memory and executable by the processor, the cleaning device implementing the cleaning method according to any one of claims 1 to 34 when the processor executes the program.

37. A computer readable storage medium having a computer program stored thereon, the computer program implementing the cleaning method according to any one of claims 1 to 34 when executed by a processor.

Images