Carpet cleaning method and device for cleaning robot

Abstract

The present application provides a carpet cleaning method and device for a cleaning robot. The method detects the ground type of a preset area in front of the preset area when performing a cleaning task for a preset area to be cleaned, and when it is determined that the ground type is a carpet type, adds the cleaning grid corresponding to the preset area in the initial cleaning grid sequence to the carpet grid sequence to be cleaned, and removes the target cleaning grid from the initial cleaning grid sequence to form an updated cleaning grid sequence. Furthermore, when it is determined that all target cleaning grids in the carpet grid sequence to be cleaned form a closed area, the closed area is determined as the carpet area to be cleaned. After the cleaning robot completes cleaning the updated cleaning grid sequence, the working mode of the cleaning robot is switched to a carpet cleaning mode to clean each target cleaning grid in the carpet grid sequence to be cleaned, thereby ensuring that the cleaning robot uses a cleaning mode suitable for the area with different ground types.

Description

Technical Field

[0001] The present application relates to data processing technology, and more particularly to a carpet cleaning method and device for a cleaning robot. Background Art

[0002] With the rapid development of smart homes and robotics, cleaning robots have gradually penetrated into people's daily lives and become an important auxiliary tool for household cleaning. However, as the home environment becomes increasingly complex, cleaning robots face more diverse floor types and cleaning challenges.

[0003] Traditional cleaning robots, when initially designed, often employ fixed cleaning patterns and strategies, making them incapable of handling the varying cleaning methods and strength requirements of different floor types (e.g., wood, tile, marble, carpet, etc.). Specifically, hard surfaces (e.g., wood and tile) are generally smooth and hard, suited to higher rotational speeds and forward speeds for rapid cleaning. Soft surfaces like carpet, on the other hand, have complex fiber structures that easily harbor dirt and grime, requiring a gentler and deeper cleaning method to ensure thorough removal of dust and debris.

[0004] Most traditional cleaning robots only have a few fixed cleaning modes and are unable to automatically switch cleaning methods and strategies based on different floor types. This results in the risk of damaging carpet fibers due to excessive cleaning force in standard cleaning modes when cleaning soft surfaces such as carpets, or leaving blind spots due to insufficient cleaning depth. Summary of the Invention

[0005] The present application provides a carpet cleaning method and device for a cleaning robot, so as to ensure that the cleaning robot uses a cleaning mode suitable for a carpet area.

[0006] In a first aspect, the present application provides a carpet cleaning method using a cleaning robot, comprising:

[0007] When the cleaning robot performs a cleaning task for a preset area to be cleaned, the cleaning robot detects the ground type in a preset area range in front, where the preset area range in front is the area in front of the current moving direction of the cleaning robot, and the cleaning task includes an initial cleaning path, and the initial cleaning path includes an initial cleaning grid sequence;

[0008] If it is determined that the floor type is a carpet type, at least one target cleaning grid corresponding to the front preset area in the initial cleaning grid sequence is added to the carpet grid sequence to be cleaned, and the cleaning robot removes the target cleaning grid from the initial cleaning grid sequence to form an updated cleaning grid sequence;

[0009] If it is determined that all target cleaning grids in the carpet grid sequence to be cleaned form a closed area, the closed area is determined as the carpet area to be cleaned;

[0010] After the cleaning robot completes cleaning the updated cleaning grid sequence, the working mode of the cleaning robot is switched to the carpet cleaning mode to clean each target cleaning grid in the carpet grid sequence to be cleaned, wherein the carpet cleaning mode includes at least one of switching the working brush head of the cleaning robot to a preset carpet brush head, reducing the brush head rotation speed of the cleaning robot, and reducing the forward speed of the cleaning robot.

[0011] In the above scheme, by detecting the ground type in the preset area ahead and identifying the carpet area, the cleaning robot can specifically add the carpet area to the carpet grid sequence to be cleaned, and remove the grid marked as the target cleaning from the initial cleaning grid sequence, so that after cleaning the non-carpet area, it can switch to the carpet cleaning mode to clean the carpet area, thereby avoiding the need for the cleaning robot to frequently switch between the carpet cleaning mode and the non-carpet cleaning mode when cleaning according to the pre-planned cleaning path, ensuring that the cleaning robot uses the cleaning mode suitable for the area in areas with different ground types, effectively reducing unnecessary energy consumption and mechanical wear, and also improving the integrity and efficiency of cleaning.

[0012] Optionally, the detecting of the ground type in the preset area ahead includes:

[0013] An ultrasonic transmitter on the cleaning robot transmits an ultrasonic signal, and an ultrasonic receiver on the cleaning robot receives an echo signal of the ultrasonic signal, wherein a calibrated transmission area of ​​the ultrasonic transmitter is the preset area range in front;

[0014] A model is determined using a preset ground type, and the ground type corresponding to the preset area ahead is determined according to the echo signal.

[0015] In this solution, an ultrasonic transmitter and receiver are used to precisely identify the floor type within a preset area ahead by transmitting and receiving ultrasonic signals, improving both accuracy and reliability. Furthermore, a preset floor type determination model, combined with the echo characteristics of ultrasonic signals, can effectively distinguish between various floor types, such as carpet and hard surfaces, enabling the cleaning robot to better adapt to diverse home or office environments. By accurately identifying whether the floor ahead is carpet, this method avoids unnecessary switching between cleaning modes, reduces resource waste, and improves cleaning efficiency and accuracy.

[0016] Optionally, the using a preset ground type determination model and determining the ground type corresponding to the preset area ahead according to the echo signal includes:

[0017] Acquire a set of emission characteristic parameters of the ultrasonic signal, wherein the set of emission characteristic parameters includes ultrasonic emission intensity and ultrasonic emission frequency;

[0018] Acquiring an echo characteristic parameter set of the echo signal, wherein the echo characteristic parameter set includes ultrasonic echo intensity and ultrasonic echo frequency;

[0019] determining an ultrasonic attenuation characteristic parameter set according to the emission characteristic parameter set and the echo characteristic parameter set, wherein the ultrasonic attenuation characteristic parameter set includes an ultrasonic intensity attenuation characteristic parameter and an ultrasonic frequency attenuation characteristic parameter;

[0020] generating an ultrasonic feature parameter vector set according to the emission feature parameter set, the echo feature parameter set, and the ultrasonic attenuation feature parameter;

[0021] The ultrasonic feature parameter vector set is input into a preset first support vector machine to output the ground type corresponding to the preset area range in front.

[0022] In the above scheme, by obtaining the emission and echo characteristic parameters of the ultrasonic signal and calculating the ultrasonic attenuation characteristic parameters, the process of ground type identification is further refined, and the accuracy and stability of identification are improved. In addition, the ultrasonic characteristic parameter vector set is processed using a support vector machine, which can quickly output the ground type identification results and improve the real-time response capability of the cleaning robot.

[0023] Optionally, before determining the model by using a preset ground type and determining the ground type corresponding to the preset area ahead according to the echo signal, the method further includes:

[0024] Acquire a ground image of the preset area ahead by a visual sensor on the cleaning robot, and preprocess the ground image to generate a ground feature image, wherein the preprocessing includes grayscale processing and noise reduction processing;

[0025] Extracting an image feature parameter vector set of the ground feature image using a gray level co-occurrence matrix, wherein the parameters in the image feature vector set include at least one of image energy, image contrast, image entropy, and image inverse moment difference;

[0026] Correspondingly, the method of determining a model using a preset ground type and determining the ground type corresponding to the preset area ahead according to the echo signal further includes:

[0027] The ultrasound feature parameter vector set and the image feature parameter vector set are input into a preset second support vector machine to output the ground type corresponding to the front preset area range.

[0028] In the above scheme, the data of ultrasonic sensors and visual sensors are combined, the image feature parameters are extracted through the gray-level co-occurrence matrix, and then multi-sensor fusion recognition is performed, which further improves the accuracy and robustness of ground type recognition.

[0029] Optionally, after inputting the ultrasonic feature parameter vector set into a preset first support vector machine to output the ground type corresponding to the preset front area, the method further includes:

[0030] If the ground type corresponding to the preset area in front outputted by the preset first support vector machine is an undetermined ground type, a ground image of the preset area in front is acquired by a visual sensor on the cleaning robot, and the ground image is preprocessed to generate a ground feature image, wherein the preprocessing includes grayscale processing and noise reduction processing;

[0031] Extracting an image feature parameter vector set of the ground feature image using a gray level co-occurrence matrix, wherein the parameters in the image feature vector set include at least one of image energy, image contrast, image entropy, and image inverse moment difference;

[0032] Correspondingly, the method of determining a model using a preset ground type and determining the ground type corresponding to the preset area ahead according to the echo signal further includes:

[0033] The image feature parameter vector set is input into a preset third support vector machine to output the ground type corresponding to the preset area range in front.

[0034] In the above scheme, when the ultrasonic signal recognition result is uncertain, supplementary recognition is performed through the visual sensor, so that the ground type can be further identified through the visual sensor, which can effectively avoid misjudgment and missed judgment, improve the recognition accuracy, and enable the cleaning robot to maintain high recognition performance and stability in complex environments, and thus can effectively cope with complex and changeable home or office environments, ensuring that the cleaning robot can accurately identify the ground type.

[0035] In addition, it is worth mentioning that since the amount of data processing for ultrasonic signals is small, while the amount of data processing for image signals is large, in order to adapt to the rapid data processing during the movement of the cleaning robot, the above-mentioned method first uses the ultrasonic feature parameter vector to identify the ground type, and only uses the image feature parameter vector for identification when the ground type cannot be accurately identified using the ultrasonic feature parameter vector. This can ensure both the speed and accuracy of recognition.

[0036] Optionally, determining that all target cleaning grids in the to-be-cleaned carpet grid sequence form a closed area includes:

[0037] Acquire a preset area boundary of the preset area to be cleaned, wherein the preset area boundary includes a boundary grid sequence, and the initial cleaning grid sequence includes the boundary grid sequence;

[0038] The closed area is formed according to the carpet grid sequence to be cleaned and the boundary grid sequence, wherein grids on the area boundary of the closed area belong to the carpet grid sequence to be cleaned and / or the boundary grid sequence.

[0039] In this solution, by obtaining the preset area boundaries and the grid sequence of the carpet to be cleaned, the carpet area to be cleaned can be accurately delineated, ensuring that the cleaning robot can perform targeted cleaning. Furthermore, based on the identification of the enclosed area, the cleaning robot can formulate a more reasonable cleaning strategy, such as prioritizing or deprioritizing the cleaning of carpets within the enclosed area, further improving overall cleaning efficiency.

[0040] Optionally, after determining to form the closed area according to the carpet grid sequence to be cleaned and the boundary grid sequence, the method further includes:

[0041] The cleaning robot determines a set of carpet grids to be cleaned according to the closed area, where the set of carpet grids to be cleaned includes all grids on the boundary of the closed area and all grids within the boundary of the closed area;

[0042] Correspondingly, after the cleaning robot completes cleaning the updated cleaning grid sequence, the method further includes:

[0043] The cleaning robot obtains a current remaining cleaning area margin, and determines a current remaining carpet cleaning area margin according to the current remaining cleaning area margin and a preset carpet cleaning characteristic coefficient, wherein the current remaining carpet cleaning area margin is a ratio of the current remaining cleaning area margin to the preset carpet cleaning characteristic coefficient, and the preset carpet cleaning characteristic coefficient is a constant greater than 1;

[0044] The cleaning robot determines an estimated carpet cleaning area based on the carpet grid set to be cleaned;

[0045] If the current remaining carpet cleaning area is greater than the estimated required carpet cleaning area, the cleaning robot is switched to the carpet cleaning mode and cleans each target cleaning grid in the sequence of carpet grids to be cleaned, wherein the current remaining carpet cleaning area is associated with the remaining garbage capacity or the remaining power of the cleaning robot;

[0046] If the current remaining carpet cleaning area is less than the estimated required carpet cleaning area, the cleaning robot executes a cleaning pause instruction and executes a self-cleaning instruction or a charging instruction, so that after completing the self-cleaning task or the charging task, the cleaning robot switches to the carpet cleaning mode and cleans each target cleaning grid in the carpet grid sequence to be cleaned.

[0047] In this solution, by calculating the current remaining cleaning area and the estimated carpet cleaning area required, the cleaning robot can intelligently determine cleaning timing, avoiding cleaning carpets when the battery is low or the trash container is full, ensuring the smooth completion of cleaning tasks. Furthermore, when the battery is low or the trash container is full, the cleaning robot can automatically perform self-cleaning or charging tasks, and then resume carpet cleaning tasks after the tasks are completed, effectively managing the cleaning robot's resource allocation and improving work efficiency.

[0048] In a second aspect, the present application provides a carpet cleaning device of a cleaning robot, comprising:

[0049] a detection module, configured to detect the ground type in a preset area ahead of the cleaning robot when the cleaning robot performs a cleaning task, wherein the preset area ahead is the area ahead of the current moving direction of the cleaning robot, the cleaning task includes an initial cleaning path, and the initial cleaning path includes an initial cleaning grid sequence;

[0050] a processing module configured to, when determining that the floor type is a carpet type, add at least one target cleaning grid corresponding to the front preset area in the initial cleaning grid sequence to the carpet grid sequence to be cleaned, and the cleaning robot removes the target cleaning grid from the initial cleaning grid sequence to form an updated cleaning grid sequence;

[0051] The processing module is further configured to determine that all target cleaning grids in the carpet grid sequence to be cleaned form a closed area, and determine the closed area as the carpet area to be cleaned;

[0052] A control module is used to switch the working mode of the cleaning robot to a carpet cleaning mode after the cleaning robot completes cleaning the updated cleaning grid sequence, so as to clean each target cleaning grid in the carpet grid sequence to be cleaned, wherein the carpet cleaning mode includes at least one of switching the working brush head of the cleaning robot to a preset carpet brush head, reducing the brush head rotation speed of the cleaning robot, and reducing the forward speed of the cleaning robot.

[0053] Optionally, the detection module is specifically configured to:

[0054] An ultrasonic transmitter on the cleaning robot transmits an ultrasonic signal, and an ultrasonic receiver on the cleaning robot receives an echo signal of the ultrasonic signal, wherein a calibrated transmission area of ​​the ultrasonic transmitter is the preset area range in front;

[0055] A model is determined using a preset ground type, and the ground type corresponding to the preset area ahead is determined according to the echo signal.

[0056] Optionally, the processing module is specifically configured to:

[0057] Acquire a set of emission characteristic parameters of the ultrasonic signal, wherein the set of emission characteristic parameters includes ultrasonic emission intensity and ultrasonic emission frequency;

[0058] Acquiring an echo characteristic parameter set of the echo signal, wherein the echo characteristic parameter set includes ultrasonic echo intensity and ultrasonic echo frequency;

[0059] determining an ultrasonic attenuation characteristic parameter set according to the emission characteristic parameter set and the echo characteristic parameter set, wherein the ultrasonic attenuation characteristic parameter set includes an ultrasonic intensity attenuation characteristic parameter and an ultrasonic frequency attenuation characteristic parameter;

[0060] generating an ultrasonic feature parameter vector set according to the emission feature parameter set, the echo feature parameter set, and the ultrasonic attenuation feature parameter;

[0061] The ultrasonic feature parameter vector set is input into a preset first support vector machine to output the ground type corresponding to the preset area range in front.

[0062] Optionally, the detection module is further specifically used to

[0063] Acquire a ground image of the preset area ahead by a visual sensor on the cleaning robot, and preprocess the ground image to generate a ground feature image, wherein the preprocessing includes grayscale processing and noise reduction processing;

[0064] Extracting an image feature parameter vector set of the ground feature image using a gray level co-occurrence matrix, wherein the parameters in the image feature vector set include at least one of image energy, image contrast, image entropy, and image inverse moment difference;

[0065] Correspondingly, optionally, the processing module is further specifically configured to:

[0066] The ultrasound feature parameter vector set and the image feature parameter vector set are input into a preset second support vector machine to output the ground type corresponding to the front preset area range.

[0067] Optionally, the processing module is further configured to:

[0068] If the ground type corresponding to the preset area in front outputted by the preset first support vector machine is an undetermined ground type, a ground image of the preset area in front is acquired by a visual sensor on the cleaning robot, and the ground image is preprocessed to generate a ground feature image, wherein the preprocessing includes grayscale processing and noise reduction processing;

[0069] Extracting an image feature parameter vector set of the ground feature image using a gray level co-occurrence matrix, wherein the parameters in the image feature vector set include at least one of image energy, image contrast, image entropy, and image inverse moment difference;

[0070] Correspondingly, the method of determining a model using a preset ground type and determining the ground type corresponding to the preset area ahead according to the echo signal further includes:

[0071] The image feature parameter vector set is input into a preset third support vector machine to output the ground type corresponding to the preset area range in front.

[0072] Optionally, the processing module is further configured to:

[0073] Acquire a preset area boundary of the preset area to be cleaned, wherein the preset area boundary includes a boundary grid sequence, and the initial cleaning grid sequence includes the boundary grid sequence;

[0074] The closed area is formed according to the carpet grid sequence to be cleaned and the boundary grid sequence, wherein grids on the area boundary of the closed area belong to the carpet grid sequence to be cleaned and / or the boundary grid sequence.

[0075] Optionally, the processing module is specifically configured to:

[0076] The cleaning robot determines a set of carpet grids to be cleaned according to the closed area, where the set of carpet grids to be cleaned includes all grids on the boundary of the closed area and all grids within the boundary of the closed area;

[0077] The cleaning robot obtains a current remaining cleaning area margin, and determines a current remaining carpet cleaning area margin according to the current remaining cleaning area margin and a preset carpet cleaning characteristic coefficient, wherein the current remaining carpet cleaning area margin is a ratio of the current remaining cleaning area margin to the preset carpet cleaning characteristic coefficient, and the preset carpet cleaning characteristic coefficient is a constant greater than 1;

[0078] The cleaning robot determines an estimated carpet cleaning area based on the carpet grid set to be cleaned;

[0079] If the current remaining carpet cleaning area is greater than the estimated required carpet cleaning area, the cleaning robot is switched to the carpet cleaning mode and cleans each target cleaning grid in the sequence of carpet grids to be cleaned, wherein the current remaining carpet cleaning area is associated with the remaining garbage capacity or the remaining power of the cleaning robot;

[0080] If the current remaining carpet cleaning area is less than the estimated required carpet cleaning area, the cleaning robot executes a cleaning pause instruction and executes a self-cleaning instruction or a charging instruction, so that after completing the self-cleaning task or the charging task, the cleaning robot switches to the carpet cleaning mode and cleans each target cleaning grid in the carpet grid sequence to be cleaned.

[0081] In a third aspect, the present application provides an electronic device, comprising:

[0082] processor; and,

[0083] a memory for storing executable instructions of the processor;

[0084] The processor is configured to perform any possible method described in the first aspect by executing the executable instructions.

[0085] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement any possible method described in the first aspect.

[0086] The carpet cleaning method and device of the cleaning robot provided in the present application detect the ground type of the preset area in front when performing the cleaning task of the preset area to be cleaned, and when determining that the ground type is a carpet type, at least one target cleaning grid corresponding to the preset area in front in the initial cleaning grid sequence is added to the carpet grid sequence to be cleaned, and the target cleaning grid is removed from the initial cleaning grid sequence to form an updated cleaning grid sequence, and when it is determined that all the target cleaning grids in the carpet grid sequence to be cleaned form a closed area, the closed area is determined as the carpet area to be cleaned, so that when the cleaning robot performs the cleaning task of the updated cleaning grid sequence After the grid sweeping sequence is completed, the working mode of the cleaning robot is switched to the carpet cleaning mode to clean each target cleaning grid in the carpet grid sequence to be cleaned. After cleaning the non-carpet area, it is switched to the carpet cleaning mode to clean the carpet area. This avoids the need for the cleaning robot to frequently switch between the carpet cleaning mode and the non-carpet cleaning mode when cleaning according to the pre-planned cleaning path, ensuring that the cleaning robot uses the cleaning mode suitable for the area with different floor types, effectively reducing unnecessary energy consumption and mechanical wear, and also improving the integrity and efficiency of cleaning. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0088] Figure 1 This is a flowchart of a carpet cleaning method of a cleaning robot according to an exemplary embodiment of the present application;

[0089] Figure 2 This is a flowchart of a carpet cleaning method of a cleaning robot according to another exemplary embodiment of the present application;

[0090] Figure 3 1 is a schematic structural diagram of a carpet cleaning device of a cleaning robot according to an exemplary embodiment of the present application;

[0091] Figure 4 It is a structural diagram of an electronic device according to an exemplary embodiment of the present application.

[0092] The above drawings illustrate specific embodiments of the present application, which will be described in more detail below. These drawings and the textual description are not intended to limit the scope of the present application in any way, but rather to illustrate the concepts of the present application to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0093] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

[0094] To address the aforementioned issues, the present invention provides an embodiment in which an ultrasonic transmitter and receiver on a cleaning robot transmit and receive ultrasonic signals to detect the surface type in a preset area in front of the cleaning robot. Using a preset surface type determination model, the ultrasonic signal's emission characteristic parameters (such as emission intensity and frequency) and echo characteristic parameters (such as echo intensity and frequency) are combined to calculate ultrasonic attenuation characteristic parameters, generating a vector set of ultrasonic characteristic parameters. This vector set is then input into a preset support vector machine to output the surface type in the preset area in front of the robot.

[0095] To further improve the accuracy and robustness of recognition, the embodiments provided in the application also introduce visual sensors as an auxiliary recognition method. When the ultrasonic signal recognition result is uncertain, the visual sensor obtains the ground image of the preset area in front, performs preprocessing such as grayscale conversion and noise reduction, and uses the gray-level co-occurrence matrix to extract the image feature parameter vector set, which is then input into another preset support vector machine for recognition. This multi-sensor fusion recognition method effectively avoids the problems of misjudgment and missed judgment that may be caused by a single sensor.

[0096] After identifying the carpet area, the embodiment provided in the application removes the target cleaning grid corresponding to the carpet area from the initial cleaning grid sequence and adds it to the carpet grid sequence to be cleaned. When the cleaning robot completes cleaning the non-carpet area, it automatically switches to carpet cleaning mode to perform targeted cleaning of the carpet area. Carpet cleaning mode includes switching to a preset carpet brush head, reducing the brush head speed, and reducing the forward speed to adapt to carpet cleaning needs and reduce mechanical wear and energy consumption.

[0097] The embodiment provided by the application also proposes a method for identifying closed areas. By obtaining the boundary grid sequence of the preset area to be cleaned and the grid sequence of the carpet to be cleaned, the closed area is determined. On this basis, the cleaning robot can formulate a more reasonable cleaning strategy, such as giving priority to cleaning the carpet in the closed area. In addition, the embodiment provided by the application also takes into account the resource management capabilities of the cleaning robot. By calculating the current remaining cleaning area and the expected required carpet cleaning area, it intelligently determines the cleaning time to avoid cleaning the carpet when the battery is low or the garbage container is full.

[0098] To further improve cleaning efficiency, the embodiments provided in the application also introduce the concept of a cleaning task scheduling platform. After identifying the carpet area to be cleaned, the cleaning robot sends the carpet boundary grid sequence to the cleaning task scheduling platform. Based on the carpet boundary grid sequence and the preset grid array of the area to be cleaned, the platform determines the carpet marking grid set and generates a special cleaning path for the carpet area in the next cleaning task. This remote task scheduling and collaborative method makes the cleaning robot's cleaning work more orderly and efficient.

[0099] In summary, the embodiments provided in the application realize the intelligence and efficiency of cleaning robots in carpet cleaning through innovations such as precise ground type recognition, differentiated cleaning strategies, closed area recognition and intelligent scheduling, and remote task scheduling and collaboration, providing a new solution for cleaning work in home or office environments.

[0100] Figure 1 FIG. 1 is a flow chart of a carpet cleaning method of a cleaning robot according to an exemplary embodiment of the present application. Figure 1 As shown, the carpet cleaning method of the cleaning robot provided in this embodiment includes:

[0101] S101. When the cleaning robot performs a cleaning task on a preset area to be cleaned, it detects the ground type in the preset area ahead.

[0102] In this step, when the cleaning robot performs the cleaning task of the preset area to be cleaned, it detects the ground type of the preset area in front. The preset area in front is the area in front of the current movement direction of the cleaning robot. The cleaning task includes an initial cleaning path, and the initial cleaning path includes an initial cleaning grid sequence.

[0103] In a possible implementation, detecting the ground type in a preset area ahead may include the following steps:

[0104] Step 1: Start the detection process:

[0105] After the cleaning robot starts the cleaning task, it first activates the ultrasonic transmitter on it and sets its calibrated emission area to the preset area range in front of the current movement direction of the cleaning robot.

[0106] Step 2: Transmit and receive ultrasonic signals:

[0107] The ultrasonic transmitter transmits ultrasonic signals to a preset area, and the ultrasonic receiver then receives echo signals from the ultrasonic signals. These signals carry information about the physical characteristics of the ground type ahead.

[0108] Step 3: Calculate ultrasonic characteristic parameters:

[0109] The system acquires a set of ultrasonic signal emission characteristic parameters (including ultrasonic emission intensity and ultrasonic emission frequency) and a set of echo characteristic parameters (including ultrasonic echo intensity and ultrasonic echo frequency). Based on these parameters, it calculates a set of ultrasonic attenuation characteristic parameters, including ultrasonic intensity attenuation characteristic parameters and ultrasonic frequency attenuation characteristic parameters.

[0110] Step 4: Generate ultrasound feature parameter vector set:

[0111] An ultrasonic feature parameter vector set is generated by combining the emission feature parameter set, the echo feature parameter set, and the ultrasonic attenuation feature parameter.

[0112] Step 5: Identify the ground type:

[0113] The ultrasonic feature parameter vector set is input into a preset first support vector machine to output the ground type corresponding to the preset area ahead. If the ground type output by the first support vector machine is undetermined or uncertain, further auxiliary recognition is performed using a visual sensor.

[0114] In addition, visually assisted recognition can be performed. The cleaning robot's visual sensor acquires ground images of a preset area ahead. After preprocessing with grayscale conversion and noise reduction, it uses a gray-level co-occurrence matrix to extract a set of image feature parameter vectors (including image energy, image contrast, image entropy, and image inverse moment). These parameter vectors are then fed into another support vector machine to output the final ground type.

[0115] S102: Add at least one target cleaning grid corresponding to the front preset area in the initial cleaning grid sequence to the carpet grid sequence to be cleaned, and remove the target cleaning grid from the initial cleaning grid sequence to form an updated cleaning grid sequence.

[0116] If the floor type is determined to be carpet type, at least one target cleaning grid corresponding to the front preset area in the initial cleaning grid sequence is added to the carpet grid sequence to be cleaned, and the cleaning robot removes the target cleaning grid from the initial cleaning grid sequence to form an updated cleaning grid sequence.

[0117] Specifically, based on the floor type identified in the first step, if it is determined to be a carpet type, the next step is performed. At least one target cleaning grid corresponding to the preset front area in the initial cleaning grid sequence is added to the carpet grid sequence to be cleaned. At the same time, these target cleaning grids are removed from the initial cleaning grid sequence to form an updated cleaning grid sequence.

[0118] S103: Determine that all target cleaning grids in the carpet to be cleaned grid sequence form a closed area, and then determine the closed area as the carpet to be cleaned area.

[0119] In this step, the system obtains the preset area boundary of the preset area to be cleaned, which includes a boundary grid sequence. Based on the grid sequence of carpets to be cleaned and the boundary grid sequence, it determines whether a closed area is formed. If all target cleaning grids in the grid sequence of carpets to be cleaned form a closed area, the closed area is determined as the carpet area to be cleaned.

[0120] S104: Switch the working mode of the cleaning robot to the carpet cleaning mode.

[0121] In this step, after the cleaning robot completes cleaning the updated cleaning grid sequence, the working mode of the cleaning robot is switched to the carpet cleaning mode to clean each target cleaning grid in the carpet grid sequence to be cleaned, wherein the carpet cleaning mode includes at least one operation of switching the working brush head of the cleaning robot to a preset carpet brush head, reducing the brush head rotation speed of the cleaning robot, and reducing the forward speed of the cleaning robot.

[0122] Specifically, the cleaning robot completes cleaning tasks for non-carpet areas according to the updated cleaning grid sequence. After cleaning is complete, the cleaning robot automatically switches to carpet cleaning mode. This mode involves at least one of switching the working brush head to a preset carpet brush head, reducing the brush head rotation speed, and reducing the forward speed to meet carpet cleaning requirements. In carpet cleaning mode, the cleaning robot cleans each target cleaning grid in the carpet grid sequence to ensure a thorough cleaning of the carpet area.

[0123] In this embodiment, when performing a cleaning task for a preset area to be cleaned, the ground type of the front preset area is detected, and when it is determined that the ground type is a carpet type, at least one target cleaning grid corresponding to the front preset area in the initial cleaning grid sequence is added to the carpet grid sequence to be cleaned, and the target cleaning grid is removed from the initial cleaning grid sequence to form an updated cleaning grid sequence, and when it is determined that all the target cleaning grids in the carpet grid sequence to be cleaned form a closed area, the closed area is determined as the carpet area to be cleaned, so that when the cleaning robot completes the updated cleaning grid sequence After cleaning, the working mode of the cleaning robot is switched to the carpet cleaning mode to clean each target cleaning grid in the carpet grid sequence to be cleaned, so that after cleaning the non-carpet area, it is switched to the carpet cleaning mode to clean the carpet area, thereby avoiding the need for the cleaning robot to frequently switch between the carpet cleaning mode and the non-carpet cleaning mode when cleaning according to the pre-planned cleaning path, ensuring that the cleaning robot uses the cleaning mode suitable for the area with different floor types, effectively reducing unnecessary energy consumption and mechanical wear, and also improving the integrity and efficiency of cleaning.

[0124] Figure 2FIG. 1 is a flow chart of a carpet cleaning method of a cleaning robot according to another exemplary embodiment of the present application. Figure 2 As shown, the carpet cleaning method of the cleaning robot provided in this embodiment includes:

[0125] S201. When the cleaning robot performs a cleaning task on a preset area to be cleaned, it detects the ground type in the preset area ahead.

[0126] In this step, when the cleaning robot performs the cleaning task of the preset area to be cleaned, it detects the ground type of the preset area in front. The preset area in front is the area in front of the current movement direction of the cleaning robot. The cleaning task includes an initial cleaning path, and the initial cleaning path includes an initial cleaning grid sequence.

[0127] Specifically, an ultrasonic signal may be emitted by an ultrasonic transmitter on the cleaning robot, and an echo signal of the ultrasonic signal may be received by an ultrasonic receiver on the cleaning robot. The calibrated emission area of ​​the ultrasonic transmitter is a preset area range in front. Then, the preset ground type is used to determine the model, and the ground type corresponding to the preset area range in front is determined according to the echo signal.

[0128] Among them, using a preset ground type to determine a model and determining the ground type corresponding to a preset area ahead according to the echo signal may include:

[0129] Acquire an emission characteristic parameter set of an ultrasonic signal, the emission characteristic parameter set including ultrasonic emission intensity and ultrasonic emission frequency;

[0130] Acquire an echo characteristic parameter set of the echo signal, where the echo characteristic parameter set includes ultrasonic echo intensity and ultrasonic echo frequency;

[0131] Determining an ultrasonic attenuation characteristic parameter set according to the emission characteristic parameter set and the echo characteristic parameter set, the ultrasonic attenuation characteristic parameter set including an ultrasonic intensity attenuation characteristic parameter and an ultrasonic frequency attenuation characteristic parameter;

[0132] generating an ultrasonic characteristic parameter vector set according to the emission characteristic parameter set, the echo characteristic parameter set, and the ultrasonic attenuation characteristic parameter;

[0133] The ultrasonic feature parameter vector set is input into a preset first support vector machine to output the ground type corresponding to the preset area range in front.

[0134] First, the system reads the ultrasonic signal's emission characteristic parameters from the ultrasonic transmitter, including ultrasonic emission intensity and ultrasonic emission frequency. These parameters reflect the energy and frequency characteristics of the ultrasonic signal when it is emitted.

[0135] The system then extracts echo characteristic parameters from the echo signal received by the ultrasonic receiver, including ultrasonic echo intensity and ultrasonic echo frequency. These parameters reflect the energy and frequency changes of the ultrasonic signal after interacting with the ground during propagation.

[0136] Then, using the emission characteristic parameters and echo characteristic parameters, the system further calculates a set of ultrasonic attenuation characteristic parameters. These parameters include ultrasonic intensity attenuation characteristic parameters and ultrasonic frequency attenuation characteristic parameters, which describe the energy and frequency attenuation of ultrasonic signals during ground propagation.

[0137] The emission characteristic parameter set, echo characteristic parameter set, and ultrasonic attenuation characteristic parameter set are then combined to generate a comprehensive ultrasonic characteristic parameter vector set, which contains all the key information for identifying the ground type.

[0138] Next, the generated ultrasonic feature parameter vector set is input into a preset first support vector machine. The first support vector machine is a classifier that can output the type of ground surface corresponding to a preset area ahead based on the input feature parameter vector set. After the first support vector machine processes the ultrasonic feature parameter vector set, it outputs a classification result, namely, the type of ground surface corresponding to the preset area ahead. This result may be carpet, hard surface, other types of ground, or the result is yet to be determined.

[0139] It is worth noting that the training of the first support vector machine can be carried out through the following specific steps:

[0140] Step 1.1. Data collection:

[0141] Ultrasonic signal data: The ultrasonic transmitter and receiver on the cleaning robot transmit and receive ultrasonic signals on different floor types (such as carpets, hard floors, etc.), and collect the emission and echo data of the ultrasonic signals.

[0142] Label data: Label each collected ultrasonic signal dataset and clearly identify the corresponding ground type label.

[0143] Step 1.2, data preprocessing:

[0144] Feature extraction: Extract emission feature parameters (such as ultrasonic emission intensity and frequency) and echo feature parameters (such as ultrasonic echo intensity and frequency) from each ultrasonic signal dataset.

[0145] Feature calculation: Calculate ultrasonic attenuation feature parameters (such as ultrasonic intensity attenuation and frequency attenuation) based on emission and echo feature parameters.

[0146] Generate a feature vector set: combine the emission feature parameters, the echo feature parameters, and the attenuation feature parameters into a feature vector set.

[0147] Step 2.1. Select kernel function:

[0148] Select an appropriate kernel function based on the data characteristics and task requirements. Common kernel functions include linear kernels, polynomial kernels, and radial basis function (RBF) kernels. For complex terrain type recognition problems, the RBF kernel is often used due to its ability to map to high-dimensional spaces.

[0149] Step 2.2, configure parameters:

[0150] Set other key parameters of the support vector machine, such as the penalty coefficient C, related parameters of the RBF kernel function, etc. These parameters have a significant impact on the complexity and generalization ability of the model.

[0151] Step 3.1, split the dataset:

[0152] The preprocessed dataset is split into training and test sets. Cross-validation is often used to ensure the generalization ability of the model.

[0153] Step 3.2, training model:

[0154] The support vector machine is trained using the training set data. During the training process, the support vector machine attempts to find a hyperplane that maximizes the interval between sample points of different categories.

[0155] Step 4.1. Evaluate model performance:

[0156] Use the test set to evaluate the model's accuracy, recall, F1 score and other indicators to ensure that the model has good recognition ability.

[0157] Step 4.2: Model tuning:

[0158] Based on the evaluation results, adjust the parameters of the support vector machine, such as the kernel function parameters or penalty coefficients, to improve model performance. You can also try using different feature selection methods or feature engineering techniques to improve the model.

[0159] Step 5.1, model deployment:

[0160] The trained support vector machine is deployed into the cleaning robot system so that it can be called in real time during runtime.

[0161] Step 5.2, online identification:

[0162] When performing cleaning tasks, the cleaning robot collects ultrasonic signal data in real time, extracts feature vectors, and inputs them into the support vector machine for ground type recognition.

[0163] In another possible implementation, before determining the model using the preset ground type and determining the ground type corresponding to the preset area ahead according to the echo signal, the method further includes:

[0164] The visual sensor on the cleaning robot acquires a ground image of a preset area ahead and pre-processes the ground image to generate a ground feature image. The pre-processing includes grayscale processing and noise reduction processing.

[0165] Extracting an image feature parameter vector set of a ground feature image using a gray level co-occurrence matrix, wherein the parameters in the image feature vector set include at least one of image energy, image contrast, image entropy, and image inverse moment difference;

[0166] Correspondingly, the model is determined using a preset ground type, and the ground type corresponding to the preset area ahead is determined according to the echo signal, further comprising:

[0167] The ultrasound feature parameter vector set and the image feature parameter vector set are input into a preset second support vector machine to output the ground type corresponding to the preset area range in front.

[0168] Combining the data of ultrasonic sensors and visual sensors, image feature parameters are extracted through gray-level co-occurrence matrix, and then multi-sensor fusion recognition is performed to further improve the accuracy and robustness of ground type recognition.

[0169] In addition, it is worth noting that the generation process of the above-mentioned preset second support vector machine can refer to the generation process of the preset first support vector machine, and will not be repeated here.

[0170] In another possible implementation, after inputting the ultrasonic feature parameter vector set into a preset first support vector machine to output a ground type corresponding to a preset area ahead, the method further includes:

[0171] If the ground type corresponding to the preset area in front outputted by the preset first support vector machine is the undetermined ground type, a ground image of the preset area in front is acquired by a visual sensor on the cleaning robot, and the ground image is preprocessed to generate a ground feature image, wherein the preprocessing includes grayscale processing and noise reduction processing;

[0172] Extracting an image feature parameter vector set of a ground feature image using a gray level co-occurrence matrix, wherein the parameters in the image feature vector set include at least one of image energy, image contrast, image entropy, and image inverse moment difference;

[0173] Correspondingly, the model is determined using a preset ground type, and the ground type corresponding to the preset area ahead is determined according to the echo signal, further comprising:

[0174] The image feature parameter vector set is input into a preset third support vector machine to output the ground type corresponding to the preset area range in front.

[0175] In the above scheme, when the ultrasonic signal recognition result is uncertain, supplementary recognition is performed through the visual sensor, so that the ground type can be further identified through the visual sensor, which can effectively avoid misjudgment and missed judgment, improve the recognition accuracy, and enable the cleaning robot to maintain high recognition performance and stability in complex environments, and thus can effectively cope with complex and changeable home or office environments, ensuring that the cleaning robot can accurately identify the ground type.

[0176] Since the amount of data processing for ultrasonic signals is small, while the amount of data processing for image signals is large, in order to adapt to the rapid data processing during the movement of the cleaning robot, the above method first uses the ultrasonic feature parameter vector to identify the ground type, and only uses the image feature parameter vector for identification when the ground type cannot be accurately identified using the ultrasonic feature parameter vector. This can ensure both the speed and accuracy of recognition.

[0177] In addition, it is worth noting that the generation process of the above-mentioned preset third support vector machine can refer to the generation process of the preset first support vector machine, which will not be described in detail here.

[0178] S202: Add at least one target cleaning grid corresponding to the front preset area in the initial cleaning grid sequence to the carpet grid sequence to be cleaned, and remove the target cleaning grid from the initial cleaning grid sequence to form an updated cleaning grid sequence.

[0179] If the floor type is determined to be carpet type, at least one target cleaning grid corresponding to the front preset area in the initial cleaning grid sequence is added to the carpet grid sequence to be cleaned, and the cleaning robot removes the target cleaning grid from the initial cleaning grid sequence to form an updated cleaning grid sequence.

[0180] Specifically, based on the floor type identified in the first step, if it is determined to be a carpet type, the next step is performed. At least one target cleaning grid corresponding to the preset front area in the initial cleaning grid sequence is added to the carpet grid sequence to be cleaned. At the same time, these target cleaning grids are removed from the initial cleaning grid sequence to form an updated cleaning grid sequence.

[0181] S203: Determine that all target cleaning grids in the carpet to be cleaned grid sequence form a closed area, and then determine the closed area as the carpet to be cleaned area.

[0182] In this step, the system obtains the preset area boundary of the preset area to be cleaned, which includes a boundary grid sequence. Based on the grid sequence of carpets to be cleaned and the boundary grid sequence, it determines whether a closed area is formed. If all target cleaning grids in the grid sequence of carpets to be cleaned form a closed area, the closed area is determined as the carpet area to be cleaned.

[0183] Specifically, a preset area boundary of a preset area to be cleaned is obtained, the preset area boundary including a boundary grid sequence, and the initial cleaning grid sequence including a boundary grid sequence. A closed area is determined based on the carpet grid sequence to be cleaned and the boundary grid sequence, wherein grids on the area boundary of the closed area belong to the carpet grid sequence to be cleaned and / or the boundary grid sequence.

[0184] S204: The cleaning robot determines a set of carpet grids to be cleaned according to the enclosed area.

[0185] In this step, the cleaning robot determines a set of carpet grids to be cleaned according to the closed area. The set of carpet grids to be cleaned includes all grids on the boundary of the closed area and all grids within the boundary of the closed area.

[0186] S205: The cleaning robot obtains the current remaining cleaning area margin, and determines the current remaining carpet cleaning area margin according to the current remaining cleaning area margin and a preset carpet cleaning characteristic coefficient.

[0187] In this step, the cleaning robot obtains the current remaining cleaning area margin, and determines the current remaining carpet cleaning area margin based on the current remaining cleaning area margin and the preset carpet cleaning characteristic coefficient. The current remaining carpet cleaning area margin is the ratio of the current remaining cleaning area margin to the preset carpet cleaning characteristic coefficient. The preset carpet cleaning characteristic coefficient is a constant greater than 1.

[0188] S206: The cleaning robot determines an estimated carpet cleaning area based on the set of carpet grids to be cleaned, and determines a working mode based on the current remaining carpet cleaning area and the estimated carpet cleaning area.

[0189] Specifically, the cleaning robot determines an estimated required carpet cleaning area based on the set of carpet grids to be cleaned.

[0190] If the current remaining carpet cleaning area is greater than the expected required carpet cleaning area, the cleaning robot will switch to carpet cleaning mode and clean each target cleaning grid in the carpet grid sequence to be cleaned. The current remaining carpet cleaning area is associated with the remaining garbage capacity or remaining power of the cleaning robot.

[0191] If the current remaining carpet cleaning area is less than the expected required carpet cleaning area, the cleaning robot executes the cleaning pause command and executes the self-cleaning command or the charging command, so that after completing the self-cleaning task or the charging task, the cleaning robot switches to the carpet cleaning mode and cleans each target cleaning grid in the carpet grid sequence to be cleaned.

[0192] In addition, it is worth mentioning that after the enclosed area is determined as the carpet area to be cleaned, it can also include:

[0193] The cleaning robot generates a carpet boundary grid sequence based on the grids on the boundary of the closed area and sends the carpet boundary grid sequence to the cleaning task scheduling platform;

[0194] The cleaning task scheduling platform determines a carpet marking grid set corresponding to the carpet area to be cleaned from the to-be-cleaned grid array corresponding to the preset to-be-cleaned area based on the carpet boundary grid sequence. The carpet marking grid set includes the carpet boundary grid sequence and all grids located in the closed area of ​​the to-be-cleaned grid array.

[0195] When the cleaning task scheduling platform obtains the cleaning task request for the next cleaning task, a first cleaning path is generated according to the first cleaning grid set, and a second cleaning path is generated according to the carpet marking grid set. The first cleaning grid set is the difference between the initial cleaning grid set and the carpet marking grid set, and the initial cleaning grid set is the grid set corresponding to the initial cleaning grid sequence.

[0196] In the above scheme, by sending the carpet boundary grid sequence to the cleaning task scheduling platform, and generating the first cleaning path according to the difference between the initial cleaning grid set and the carpet mark grid set, and generating the second cleaning path according to the carpet mark grid set, the carpet area and the non-carpet area are automatically separated in the subsequent cleaning tasks to plan the cleaning path, thereby further improving the execution efficiency and accuracy of subsequent cleaning tasks.

[0197] Figure 3 FIG. 1 is a structural diagram of a carpet cleaning device of a cleaning robot according to an exemplary embodiment of the present application. Figure 3 As shown, the carpet cleaning device 300 of the cleaning robot provided in this embodiment includes:

[0198] A detection module 310 is configured to detect the ground type in a preset area ahead of the cleaning robot when the cleaning robot performs a cleaning task, wherein the preset area ahead is the area ahead of the current moving direction of the cleaning robot, and the cleaning task includes an initial cleaning path, and the initial cleaning path includes an initial cleaning grid sequence;

[0199] The processing module 320 is configured to, when determining that the floor type is a carpet type, add at least one target cleaning grid corresponding to the front preset area in the initial cleaning grid sequence to the carpet grid sequence to be cleaned, and the cleaning robot removes the target cleaning grid from the initial cleaning grid sequence to form an updated cleaning grid sequence;

[0200] The processing module 320 is further configured to determine that all target cleaning grids in the carpet grid sequence to be cleaned form a closed area, and determine the closed area as the carpet area to be cleaned;

[0201] The control module 330 is used to switch the working mode of the cleaning robot to the carpet cleaning mode after the cleaning robot completes cleaning the updated cleaning grid sequence, so as to clean each target cleaning grid in the carpet grid sequence to be cleaned, wherein the carpet cleaning mode includes at least one operation of switching the working brush head of the cleaning robot to a preset carpet brush head, reducing the brush head rotation speed of the cleaning robot, and reducing the forward speed of the cleaning robot.

[0202] Optionally, the detection module 310 is specifically configured to:

[0203] An ultrasonic transmitter on the cleaning robot transmits an ultrasonic signal, and an ultrasonic receiver on the cleaning robot receives an echo signal of the ultrasonic signal, wherein a calibrated transmission area of ​​the ultrasonic transmitter is the preset area range in front;

[0204] A model is determined using a preset ground type, and the ground type corresponding to the preset area ahead is determined according to the echo signal.

[0205] Optionally, the processing module 320 is specifically configured to:

[0206] Acquire a set of emission characteristic parameters of the ultrasonic signal, wherein the set of emission characteristic parameters includes ultrasonic emission intensity and ultrasonic emission frequency;

[0207] Acquiring an echo characteristic parameter set of the echo signal, wherein the echo characteristic parameter set includes ultrasonic echo intensity and ultrasonic echo frequency;

[0208] determining an ultrasonic attenuation characteristic parameter set according to the emission characteristic parameter set and the echo characteristic parameter set, wherein the ultrasonic attenuation characteristic parameter set includes an ultrasonic intensity attenuation characteristic parameter and an ultrasonic frequency attenuation characteristic parameter;

[0209] generating an ultrasonic feature parameter vector set according to the emission feature parameter set, the echo feature parameter set, and the ultrasonic attenuation feature parameter;

[0210] The ultrasonic feature parameter vector set is input into a preset first support vector machine to output the ground type corresponding to the preset area range in front.

[0211] Optionally, the detection module 310 is further specifically used to

[0212] Acquire a ground image of the preset area ahead by a visual sensor on the cleaning robot, and preprocess the ground image to generate a ground feature image, wherein the preprocessing includes grayscale processing and noise reduction processing;

[0213] Extracting an image feature parameter vector set of the ground feature image using a gray level co-occurrence matrix, wherein the parameters in the image feature vector set include at least one of image energy, image contrast, image entropy, and image inverse moment difference;

[0214] Correspondingly, optionally, the processing module 320 is further specifically configured to:

[0215] The ultrasound feature parameter vector set and the image feature parameter vector set are input into a preset second support vector machine to output the ground type corresponding to the front preset area range.

[0216] Optionally, the processing module 320 is further configured to:

[0217] If the ground type corresponding to the preset area in front outputted by the preset first support vector machine is an undetermined ground type, a ground image of the preset area in front is acquired by a visual sensor on the cleaning robot, and the ground image is preprocessed to generate a ground feature image, wherein the preprocessing includes grayscale processing and noise reduction processing;

[0218] Extracting an image feature parameter vector set of the ground feature image using a gray level co-occurrence matrix, wherein the parameters in the image feature vector set include at least one of image energy, image contrast, image entropy, and image inverse moment difference;

[0219] Correspondingly, the method of determining a model using a preset ground type and determining the ground type corresponding to the preset area ahead according to the echo signal further includes:

[0220] The image feature parameter vector set is input into a preset third support vector machine to output the ground type corresponding to the preset area range in front.

[0221] Optionally, the processing module 320 is further configured to:

[0222] Acquire a preset area boundary of the preset area to be cleaned, wherein the preset area boundary includes a boundary grid sequence, and the initial cleaning grid sequence includes the boundary grid sequence;

[0223] The closed area is formed according to the carpet grid sequence to be cleaned and the boundary grid sequence, wherein grids on the area boundary of the closed area belong to the carpet grid sequence to be cleaned and / or the boundary grid sequence.

[0224] Optionally, the processing module 320 is specifically configured to:

[0225] The cleaning robot determines a set of carpet grids to be cleaned according to the closed area, where the set of carpet grids to be cleaned includes all grids on the boundary of the closed area and all grids within the boundary of the closed area;

[0226] The cleaning robot obtains a current remaining cleaning area margin, and determines a current remaining carpet cleaning area margin according to the current remaining cleaning area margin and a preset carpet cleaning characteristic coefficient, wherein the current remaining carpet cleaning area margin is a ratio of the current remaining cleaning area margin to the preset carpet cleaning characteristic coefficient, and the preset carpet cleaning characteristic coefficient is a constant greater than 1;

[0227] The cleaning robot determines an estimated carpet cleaning area based on the carpet grid set to be cleaned;

[0228] If the current remaining carpet cleaning area is greater than the estimated required carpet cleaning area, the cleaning robot is switched to the carpet cleaning mode and cleans each target cleaning grid in the sequence of carpet grids to be cleaned, wherein the current remaining carpet cleaning area is associated with the remaining garbage capacity or the remaining power of the cleaning robot;

[0229] If the current remaining carpet cleaning area is less than the estimated required carpet cleaning area, the cleaning robot executes a cleaning pause instruction and executes a self-cleaning instruction or a charging instruction, so that after completing the self-cleaning task or the charging task, the cleaning robot switches to the carpet cleaning mode and cleans each target cleaning grid in the carpet grid sequence to be cleaned.

[0230] Figure 4 FIG. 1 is a schematic diagram of the structure of an electronic device according to an exemplary embodiment of the present application. Figure 4 As shown, this embodiment provides an electronic device 400 including: a processor 401 and a memory 402; wherein:

[0231] The memory 402 is used to store computer programs. The memory may also be a flash memory.

[0232] The processor 401 is configured to execute the execution instructions stored in the memory to implement each step in the above method. For details, please refer to the relevant description in the above method embodiment.

[0233] Optionally, the memory 402 may be independent or integrated with the processor 401 .

[0234] When the memory 402 is a device independent of the processor 401, the electronic device 400 may further include:

[0235] The bus 403 is used to connect the memory 402 and the processor 401 .

[0236] This embodiment further provides a readable storage medium, in which a computer program is stored. When at least one processor of an electronic device executes the computer program, the electronic device executes the methods provided in the various aforementioned embodiments.

[0237] This embodiment further provides a program product, which includes a computer program stored in a readable storage medium. At least one processor of an electronic device can read the computer program from the readable storage medium, and at least one processor can execute the computer program to cause the electronic device to implement the methods provided in the various embodiments described above.

[0238] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of the present application and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered merely as exemplary, and the true scope and spirit of the present application are indicated by the claims.

[0239] It should be understood that the present application is not limited to the exact structure described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A carpet cleaning method using a cleaning robot, characterized in that: include: When the cleaning robot performs a cleaning task for a preset area to be cleaned, the cleaning robot detects the ground type in a preset area range in front, where the preset area range in front is the area in front of the current moving direction of the cleaning robot, and the cleaning task includes an initial cleaning path, and the initial cleaning path includes an initial cleaning grid sequence; If it is determined that the floor type is a carpet type, at least one target cleaning grid corresponding to the front preset area in the initial cleaning grid sequence is added to the carpet grid sequence to be cleaned, and the cleaning robot removes the target cleaning grid from the initial cleaning grid sequence to form an updated cleaning grid sequence; If it is determined that all target cleaning grids in the carpet grid sequence to be cleaned form a closed area, the closed area is determined as the carpet area to be cleaned; After the cleaning robot completes cleaning the updated cleaning grid sequence, the working mode of the cleaning robot is switched to a carpet cleaning mode to clean each target cleaning grid in the carpet grid sequence to be cleaned, wherein the carpet cleaning mode includes at least one of switching the working brush head of the cleaning robot to a preset carpet brush head, reducing the brush head rotation speed of the cleaning robot, and reducing the forward speed of the cleaning robot; The step of determining that all target cleaning grids in the carpet grid sequence to be cleaned form a closed area includes: Acquire a preset area boundary of the preset area to be cleaned, wherein the preset area boundary includes a boundary grid sequence, and the initial cleaning grid sequence includes the boundary grid sequence; Determining the formation of the closed area according to the carpet grid sequence to be cleaned and the boundary grid sequence, wherein grids on the area boundary of the closed area belong to the carpet grid sequence to be cleaned and / or the boundary grid sequence; After determining to form the closed area according to the carpet grid sequence to be cleaned and the boundary grid sequence, the method further includes: The cleaning robot determines a set of carpet grids to be cleaned according to the closed area, where the set of carpet grids to be cleaned includes all grids on the boundary of the closed area and all grids within the boundary of the closed area; Correspondingly, after the cleaning robot completes cleaning the updated cleaning grid sequence, the method further includes: The cleaning robot obtains a current remaining cleaning area margin, and determines a current remaining carpet cleaning area margin according to the current remaining cleaning area margin and a preset carpet cleaning characteristic coefficient, wherein the current remaining carpet cleaning area margin is a ratio of the current remaining cleaning area margin to the preset carpet cleaning characteristic coefficient, and the preset carpet cleaning characteristic coefficient is a constant greater than 1; The cleaning robot determines an estimated carpet cleaning area based on the carpet grid set to be cleaned; If the current remaining carpet cleaning area is greater than the estimated required carpet cleaning area, the cleaning robot is switched to the carpet cleaning mode and cleans each target cleaning grid in the sequence of carpet grids to be cleaned, wherein the current remaining carpet cleaning area is associated with the remaining garbage capacity or the remaining power of the cleaning robot; If the current remaining carpet cleaning area is less than the estimated required carpet cleaning area, the cleaning robot executes a cleaning pause instruction and executes a self-cleaning instruction or a charging instruction, so that after completing the self-cleaning task or the charging task, the cleaning robot switches to the carpet cleaning mode and cleans each target cleaning grid in the carpet grid sequence to be cleaned.

2. The carpet cleaning method of the cleaning robot according to claim 1, characterized in that: The ground type of the preset area ahead of the detection includes: An ultrasonic transmitter on the cleaning robot transmits an ultrasonic signal, and an ultrasonic receiver on the cleaning robot receives an echo signal of the ultrasonic signal, wherein a calibrated transmission area of ​​the ultrasonic transmitter is the preset area range in front; A model is determined using a preset ground type, and the ground type corresponding to the preset area ahead is determined according to the echo signal.

3. The carpet cleaning method of the cleaning robot according to claim 2, characterized in that: The method of determining a model using a preset ground type and determining the ground type corresponding to the preset area ahead according to the echo signal includes: Acquire a set of emission characteristic parameters of the ultrasonic signal, wherein the set of emission characteristic parameters includes ultrasonic emission intensity and ultrasonic emission frequency; Acquiring an echo characteristic parameter set of the echo signal, wherein the echo characteristic parameter set includes ultrasonic echo intensity and ultrasonic echo frequency; determining an ultrasonic attenuation characteristic parameter set according to the emission characteristic parameter set and the echo characteristic parameter set, wherein the ultrasonic attenuation characteristic parameter set includes an ultrasonic intensity attenuation characteristic parameter and an ultrasonic frequency attenuation characteristic parameter; generating an ultrasonic feature parameter vector set according to the emission feature parameter set, the echo feature parameter set, and the ultrasonic attenuation feature parameter; The ultrasonic feature parameter vector set is input into a preset first support vector machine to output the ground type corresponding to the preset area range in front.

4. The carpet cleaning method of the cleaning robot according to claim 3, characterized in that: Before determining the model by using the preset ground type and determining the ground type corresponding to the preset area ahead according to the echo signal, the method further includes: Acquire a ground image of the preset area ahead by a visual sensor on the cleaning robot, and preprocess the ground image to generate a ground feature image, wherein the preprocessing includes grayscale processing and noise reduction processing; Extracting an image feature parameter vector set of the ground feature image using a gray level co-occurrence matrix, wherein parameters in the image feature parameter vector set include at least one of image energy, image contrast, image entropy, and image inverse moment difference; Correspondingly, the method of determining a model using a preset ground type and determining the ground type corresponding to the preset area ahead according to the echo signal further includes: The ultrasound feature parameter vector set and the image feature parameter vector set are input into a preset second support vector machine to output the ground type corresponding to the front preset area range.

5. The carpet cleaning method of the cleaning robot according to claim 3, characterized in that: After inputting the ultrasonic feature parameter vector set into a preset first support vector machine to output the ground type corresponding to the front preset area, the method further includes: If the ground type corresponding to the preset area in front outputted by the preset first support vector machine is an undetermined ground type, a ground image of the preset area in front is acquired by a visual sensor on the cleaning robot, and the ground image is preprocessed to generate a ground feature image, wherein the preprocessing includes grayscale processing and noise reduction processing; Extracting an image feature parameter vector set of the ground feature image using a gray level co-occurrence matrix, wherein parameters in the image feature parameter vector set include at least one of image energy, image contrast, image entropy, and image inverse moment difference; Correspondingly, the method of determining a model using a preset ground type and determining the ground type corresponding to the preset area ahead according to the echo signal further includes: The image feature parameter vector set is input into a preset third support vector machine to output the ground type corresponding to the preset area range in front.

6. A carpet cleaning device for a cleaning robot, characterized in that: include: a detection module, configured to detect the ground type in a preset area ahead when the cleaning robot performs a cleaning task for a preset area to be cleaned, wherein the preset area ahead is the area ahead of the current moving direction of the cleaning robot, the cleaning task includes an initial cleaning path, and the initial cleaning path includes an initial cleaning grid sequence; a processing module configured to, when determining that the floor type is a carpet type, add at least one target cleaning grid corresponding to the front preset area in the initial cleaning grid sequence to the carpet grid sequence to be cleaned, and the cleaning robot removes the target cleaning grid from the initial cleaning grid sequence to form an updated cleaning grid sequence; The processing module is further configured to determine that all target cleaning grids in the carpet grid sequence to be cleaned form a closed area, and determine the closed area as the carpet area to be cleaned; a control module, configured to switch the working mode of the cleaning robot to a carpet cleaning mode after the cleaning robot completes cleaning the updated cleaning grid sequence, so as to clean each target cleaning grid in the carpet grid sequence to be cleaned, wherein the carpet cleaning mode includes at least one of switching the working brush head of the cleaning robot to a preset carpet brush head, reducing the brush head rotation speed of the cleaning robot, and reducing the forward speed of the cleaning robot; The processing module is specifically used to: Acquire a preset area boundary of the preset area to be cleaned, wherein the preset area boundary includes a boundary grid sequence, and the initial cleaning grid sequence includes the boundary grid sequence; Determining the formation of the closed area according to the carpet grid sequence to be cleaned and the boundary grid sequence, wherein grids on the area boundary of the closed area belong to the carpet grid sequence to be cleaned and / or the boundary grid sequence; The processing module is further specifically configured to: The cleaning robot determines a set of carpet grids to be cleaned according to the closed area, where the set of carpet grids to be cleaned includes all grids on the boundary of the closed area and all grids within the boundary of the closed area; The cleaning robot obtains a current remaining cleaning area margin, and determines a current remaining carpet cleaning area margin according to the current remaining cleaning area margin and a preset carpet cleaning characteristic coefficient, wherein the current remaining carpet cleaning area margin is a ratio of the current remaining cleaning area margin to the preset carpet cleaning characteristic coefficient, and the preset carpet cleaning characteristic coefficient is a constant greater than 1; The cleaning robot determines an estimated carpet cleaning area based on the carpet grid set to be cleaned; If the current remaining carpet cleaning area is greater than the estimated required carpet cleaning area, the cleaning robot is switched to the carpet cleaning mode and cleans each target cleaning grid in the sequence of carpet grids to be cleaned, wherein the current remaining carpet cleaning area is associated with the remaining garbage capacity or the remaining power of the cleaning robot; If the current remaining carpet cleaning area is less than the estimated required carpet cleaning area, the cleaning robot executes a cleaning pause instruction and executes a self-cleaning instruction or a charging instruction, so that after completing the self-cleaning task or the charging task, the cleaning robot switches to the carpet cleaning mode and cleans each target cleaning grid in the carpet grid sequence to be cleaned.

7. An electronic device, characterized in that: include: processor; as well as, a memory for storing executable instructions of the processor; The processor is configured to perform the method according to any one of claims 1 to 5 by executing the executable instructions.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 5 when executed by a processor.

Images