Cleaning system, method and device

Abstract

The invention provides a cleaning system, method and device. The method comprises the following steps: acquiring cleaning request information from terminal equipment, wherein the cleaning request information comprises information of a to-be-cleaned target or information and a target position of the to-be-cleaned target; the cleaning request information is input into a trained large language model, a cleaning instruction output by the trained large language model is obtained, and the cleaning instruction comprises information of target cleaning equipment, a cleaning task and information of a task position; and sending a cleaning instruction to the target cleaning equipment according to the information of the target cleaning equipment. The cleaning instruction which can be recognized by the cleaning equipment is generated according to the cleaning requirement by utilizing the trained large language model, the cleaning equipment is indicated to execute the specified task, the requirement for manpower is reduced, and the labor cost is reduced.

Description

Technical Field

[0001] The present application belongs to the field of computer technology, and in particular, relates to a cleaning system, method, and device. Background Art

[0002] Cleaning work in large places, such as shopping malls, office buildings, and airports, including taking out garbage and sweeping dirt, is a task with a large workload and high labor demand. For example, a shopping mall of 100,000 square meters requires at least 20 cleaners every day, and the annual labor cost can reach millions of yuan. This kind of cleaning task requires a large number of cleaning staff, resulting in high labor costs. Summary of the invention

[0003] The embodiments of the present application provide a cleaning system and method, an apparatus, a computer-readable storage medium, and a computer program product, which can solve the problem of high labor costs for cleaning large places.

[0004] In a first aspect, an embodiment of the present application provides a cleaning method, comprising:

[0005] Receiving cleaning request information from a terminal device, wherein the cleaning request information includes information of a target to be cleaned or information of the target to be cleaned and a target position;

[0006] Inputting the cleaning request information into a trained large language model to obtain a cleaning instruction output by the trained large language model, wherein the cleaning instruction includes information of a target cleaning device, a cleaning task, and information of a task location;

[0007] Sending the cleaning instruction to the target cleaning device according to the information of the target cleaning device, wherein the cleaning instruction is used to instruct the target cleaning device to perform the cleaning task after moving to the task position;

[0008] Wherein, when the cleaning request information includes information of a target to be cleaned, the trained large language model is used to determine the target cleaning device, the cleaning task and the task location according to the information of the target to be cleaned, and output the cleaning instruction;

[0009] When the cleaning request information includes the information of the target to be cleaned and the target position, the trained large language model is used to determine the target cleaning equipment and the cleaning task according to the information of the target to be cleaned, determine the task position according to the target position, and output the cleaning instruction.

[0010] In one embodiment, the cleaning request information is information generated by the terminal device after receiving the work content input by the user and obtaining the user's location, wherein the work content includes information of the target to be cleaned;

[0011] The cleaning request information is information generated by the terminal device after receiving an image captured by a first camera device and obtaining the position of the first camera device, wherein the image includes an imaging area of ​​the target to be cleaned, and each of the first camera devices is installed at various locations of the site;

[0012] The cleaning request information is generated by the terminal device according to a task automatically triggered at a preset time, and the task includes information of the target to be cleaned;

[0013] The target location includes the location of the user and the location of the first camera device.

[0014] In one embodiment, the method further comprises:

[0015] Inputting the cleaning request information into a trained large language model to obtain an unprocessable instruction output by the trained large language model;

[0016] Sending the unprocessable instruction to the terminal device, where the unprocessable instruction is used to instruct the terminal device to send the cleaning request information to a device of a cleaning staff;

[0017] The trained large language model is used to determine that the size of the target to be cleaned is greater than a preset size according to the information of the target to be cleaned, or to determine an empty position according to the target position, and to output the unprocessable instruction.

[0018] In one embodiment, the cleaning device includes a first cleaning robot and a second cleaning robot, the first cleaning robot includes a mobile base and a clamp, a container, and a second camera device arranged on the mobile base, and the second cleaning robot is a fully automatic cleaning robot.

[0019] In one embodiment, the task location information includes target area information and identification information representing the target location.

[0020] In one embodiment, before inputting the cleaning request information into the trained large language model, the method further includes:

[0021] Acquire training data and labels, the training data including first data, second data and third data, the first data including the location and work content of the user, the second data including the location and captured images of the first camera device, and the third data including tasks that are automatically triggered at a preset time;

[0022] Using the training data, fine-tuning the large language model until a value of a loss function of the large language model is less than a preset loss value, the loss function being used to calculate a difference value between an output result of the large language model and the label;

[0023] When the value of the loss function is less than the preset loss value, the trained large language model is obtained.

[0024] In a second aspect, an embodiment of the present application provides a cleaning device, comprising:

[0025] A communication module, configured to receive cleaning request information from a terminal device, wherein the cleaning request information includes information of a target to be cleaned or information of the target to be cleaned and a target position;

[0026] It is also used to send the cleaning instruction to the target cleaning device according to the information of the target cleaning device, and the cleaning instruction is used to instruct the target cleaning device to perform the cleaning task after moving to the task position;

[0027] A prediction module, used for inputting the cleaning request information into a trained large language model to obtain a cleaning instruction output by the trained large language model, wherein the cleaning instruction includes information of a target cleaning device, a cleaning task, and information of a task location;

[0028] Wherein, when the cleaning request information includes information of a target to be cleaned, the trained large language model is used to determine the target cleaning device, the cleaning task and the task location according to the information of the target to be cleaned, and output the cleaning instruction;

[0029] When the cleaning request information includes the information of the target to be cleaned and the target position, the trained large language model is used to determine the target cleaning equipment and the cleaning task according to the information of the target to be cleaned, determine the task position according to the target position, and output the cleaning instruction.

[0030] In one embodiment, the device further comprises:

[0031] an acquisition module, used to acquire training data and labels, wherein the training data includes first data, second data and third data, wherein the first data includes the location and work content of the user, the second data includes the location and captured images of the first camera device, and the third data includes tasks that are automatically triggered at a preset time;

[0032] A training module, used to fine-tune the large language model using the training data until a value of a loss function of the large language model is less than a preset loss value, wherein the loss function is used to calculate a difference value between an output result of the large language model and the label;

[0033] It is also used to obtain the trained large language model when the value of the loss function is less than the preset loss value.

[0034] In a third aspect, an embodiment of the present application provides a cleaning system, including a server, a terminal device, and at least one cleaning device;

[0035] The terminal device is used to send the cleaning request information to the server after obtaining the cleaning request information;

[0036] The server is configured to receive the cleaning request information from the terminal device, wherein the cleaning request information includes information of a target to be cleaned or information of the target to be cleaned and a target location; input the cleaning request information into a trained large language model to obtain a cleaning instruction output by the trained large language model; and send the cleaning instruction to the target cleaning device according to the information of the target cleaning device, wherein the cleaning instruction includes information of the target cleaning device, a cleaning task, and information of a task location;

[0037] The target cleaning device is used to receive the cleaning instruction from the server, move to the task position according to the cleaning instruction, and perform the cleaning task;

[0038] Wherein, when the cleaning request information includes information of a target to be cleaned, the trained large language model is used to determine the target cleaning device, the cleaning task and the task location according to the information of the target to be cleaned, and output the cleaning instruction;

[0039] When the cleaning request information includes the information of the target to be cleaned and the target position, the trained large language model is used to determine the target cleaning equipment and the cleaning task according to the information of the target to be cleaned, determine the task position according to the target position, and output the cleaning instruction.

[0040] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method as described in any one of the above-mentioned first aspects is implemented.

[0041] In a fifth aspect, an embodiment of the present application provides a computer program product, which, when executed on a server, enables the server to execute any of the methods described in the first aspect.

[0042] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0043] The embodiment of the present application obtains cleaning request information from a terminal device, where the cleaning request information includes information about a target to be cleaned or information about a target to be cleaned and a target position; inputs the cleaning request information into a trained large language model to obtain a cleaning instruction output by the trained large language model, where the cleaning instruction includes information about a target cleaning device, a cleaning task, and information about a task position; sends a cleaning instruction to the target cleaning device based on the information about the target cleaning device, where the cleaning instruction is used to instruct the target cleaning device to move to the task position and then perform the cleaning task; wherein, when the cleaning request information includes information about the target to be cleaned, the trained large language model is used to determine the target cleaning device, the cleaning task, and the task position based on the information about the target to be cleaned, and output the cleaning instruction; when the cleaning request information includes information about the target to be cleaned and a target position, the trained large language model is used to determine the target cleaning device and the cleaning task based on the information about the target to be cleaned, and determine the task position based on the target position, and output the cleaning instruction. The trained large language model can be used to generate cleaning instructions recognizable to the cleaning device based on cleaning requirements, thereby instructing the cleaning device to perform a specified task, reducing the demand for manpower, and thus reducing labor costs.

[0044] It can be understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0046] Figure 1 is a schematic diagram of a cleaning system provided by an embodiment of the present application;

[0047] Figure 2 is a schematic diagram of a first cleaning robot provided in an embodiment of the present application;

[0048] Figure 3 This is a first flow chart of a cleaning method provided by an embodiment of the present application;

[0049] Figure 4 This is a second flow chart of the cleaning method provided by an embodiment of the present application;

[0050] Figure 5 This is a third flow chart of a cleaning method provided in one embodiment of the present application;

[0051] Figure 6It is a schematic diagram of the structure of a cleaning device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0052] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0053] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0054] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0055] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0056] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0057] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0058] Cleaning work in large places, such as shopping malls, office buildings, and airports, including taking out garbage and sweeping dirt, is a task with a large workload and high labor demand. For example, a shopping mall of 100,000 square meters requires at least 20 cleaners every day, and the annual labor cost can reach millions of yuan. This kind of cleaning task requires a large number of cleaning staff, resulting in high labor costs.

[0059] In addition, various emergencies may occur in large venues, such as spilled drinks on the ground, falling wall paint, etc., which require cleaning staff to deal with them in a timely manner, but manual cleaning methods may not be able to clean them up in time. In addition, manual cleaning methods vary from person to person, and the cleaning quality may be inconsistent.

[0060] In response to the above problems, the embodiments of the present application propose a cleaning system and method, an apparatus, a readable storage medium, and a computer program product.

[0061] In one embodiment, Figure 1 Schematic diagram of a cleaning system provided by an embodiment of the present application. Figure 1 As shown, the system includes: a terminal device 10, a server 11 and at least one cleaning device 12.

[0062] The cleaning device includes a first cleaning robot and a second cleaning robot. Figure 2 Schematic diagram of a first cleaning robot provided in one embodiment of the present application. Figure 2 As shown, the first cleaning robot includes a mobile base 20 and a fixture 21 (not shown) disposed on the mobile base 20, a container 22, and a second camera device 23. The fixture 21 includes a clamp 211 and a mechanical arm 212, one end of the mechanical arm 212 is connected to the mobile base 20, and the other end of the mechanical arm 212 is connected to the clamp 211. The container 22 can be a trash can. It can be understood that the first cleaning robot is provided with a signal receiver integrated inside.

[0063] The second cleaning robot is a fully automatic cleaning robot. The fully automatic cleaning robot can be a remote-controlled sweeping and mopping robot. It can be understood that the second cleaning robot is provided with a signal receiver integrated inside.

[0064] The terminal device 10 is used to send the cleaning request information to the server after obtaining the cleaning request information.

[0065] The cleaning request information includes information of the target to be cleaned or information of the target to be cleaned and the target position.

[0066] In applications, terminal devices can install software systems to realize information processing.

[0067] In a possible implementation, the cleaning request information is information generated by the terminal device after receiving the work content input by the user and obtaining the user's location, and the work content includes information of the target to be cleaned.

[0068] For example, each terminal device is installed at each user's workstation. The user finds sewage on the ground and enters the work content on the terminal device: There is sewage on the ground in area 1 of the venue, please assign cleaning equipment to clean it. The work content includes the target to be cleaned: sewage. The terminal device receives the work content entered by the user and obtains the user's location: workstation coordinates through the software system.

[0069] The user finds that the trash can is full of trash, and enters the work content on the terminal device: The trash can is full, please assign cleaning equipment to clean it. The work content includes the target to be cleaned: the trash can. The terminal device receives the work content entered by the user and obtains the user's location through the software system: the work station coordinates.

[0070] The terminal device generates cleaning request information according to the work content and the user's location through the software system. The cleaning request information may include the work content and the user's location, and the cleaning request information further includes the information of the target to be cleaned and the target location.

[0071] Since the cleaning equipment includes the first cleaning robot and the second cleaning robot, the work content input by the user may also be: there is sewage on the ground in area 1 of the site, please assign the second cleaning robot to clean it, and the trash can is full, please assign the first cleaning robot to clean it.

[0072] In a possible implementation, the cleaning request information is information generated by the terminal device after receiving an image captured by the first camera device and obtaining the position of the first camera device. The image includes an imaging area of ​​the target to be cleaned, and each first camera device is installed at various locations in the venue.

[0073] For example, the first camera device finds sewage on the ground and collects an image containing the sewage. The image includes the target to be cleaned: the imaging area of ​​the sewage. After the terminal device receives the image, the position of the first camera device is obtained through the software system. The position of the first camera device is used to determine the position of the stain or the area to be cleaned.

[0074] The terminal device generates cleaning request information according to the image and the position of the first camera device through a software system. The cleaning request information includes the image and the position of the first camera device, and the cleaning request information includes information of the target to be cleaned and the target position.

[0075] Correspondingly, the target location includes the location of the user and the location of the first camera device.

[0076] In a possible implementation, the cleaning request information is generated by the terminal device according to a task automatically triggered at a preset time.

[0077] For example, the preset time is set to 5 pm, and the task is to clean all trash cans in the place. The terminal device generates a cleaning request information based on the automatic triggering of cleaning all trash cans in the place at 5 pm through the software system. The cleaning request information includes the task information, and the cleaning request information includes the information of the target to be cleaned.

[0078] The server 11 is used to receive cleaning request information from a terminal device; input the cleaning request information into a trained large language model, obtain a cleaning instruction output by the trained large language model, and send a cleaning instruction to a target cleaning device according to information of the target cleaning device.

[0079] Among them, when the cleaning request information includes information about the target to be cleaned, the trained large language model is used to determine the target cleaning equipment, cleaning task and task location according to the information about the target to be cleaned, and output cleaning instructions; when the cleaning request information includes information about the target to be cleaned and the target location, the trained large language model is used to determine the target cleaning equipment and cleaning task according to the information about the target to be cleaned, and determine the task location according to the target position, and output cleaning instructions; the cleaning instructions include information about the target cleaning equipment, cleaning task and task location.

[0080] For example, the cleaning task may be to empty a trash can, sweep the floor, clean liquid off the floor, or pick up trash off the floor.

[0081] In a possible implementation manner, the task location information includes target area information and identification information representing the target location.

[0082] In the application, the server integrates a trained large language model, which can be set to the LLAVA model (Large Language and Vision Assistant, a new type of large multimodal model).

[0083] For example, the cleaning request information may include the work content and the user's location. Work content: The trash can is full, please assign the first cleaning robot to clean it, and the user's location is station 23. The trained large language model determines the target cleaning device: the first cleaning robot and the cleaning task: clean the trash can according to the trash can, and determines the task location: station 23 in area 1 according to the user's location. The cleaning instruction includes the number of the first cleaning robot, station 23 in area 1, and clean the trash can.

[0084] It can be understood that the task location information includes information of the target area and identification information representing the target location, which can provide accurate positioning for the target robot.

[0085] In order for the target cleaning device to better identify the cleaning instruction, the server processes the cleaning instruction to obtain a new cleaning instruction: robot: 1; position: block1, no45; task: clean the trash can.

[0086] The target cleaning device 12 is used to receive the cleaning instruction from the server, move to the task position according to the cleaning instruction, and perform the cleaning task.

[0087] For example, the cleaning instruction is: robot: 1; position: block 1, no 45; task: clean the trashcan. After receiving the cleaning instruction, the first cleaning robot moves to the 23rd station in zone 1, uses the second camera device to locate, uses the clamp to clamp the trashcan at the station, and pours the trash in the trashcan into the container.

[0088] It can be understood that the cleaning request information is input into the trained large language model through the server to obtain the cleaning instructions output by the trained large language model; according to the information of the target cleaning device, the cleaning instruction is sent to the target cleaning device, and after the target cleaning device receives the cleaning instruction from the server, it moves to the task position according to the cleaning instruction and performs the cleaning task, so that the cleaning device can be assigned to clean in time. At the same time, because the cleaning equipment is used to perform the task, the consistency of the cleaning quality can be guaranteed.

[0089] And through the trained large language model, it can generate accurate cleaning tasks, realize the scheduling of cleaning equipment, and cope with complex and changing cleaning needs.

[0090] After obtaining the cleaning request information through the terminal device, this embodiment sends the cleaning request information to the server; the server receives the cleaning request information from the terminal device, the cleaning request information including the information of the target to be cleaned or the information of the target to be cleaned and the target position; the cleaning request information is input into the trained large language model to obtain the cleaning instruction output by the trained large language model; according to the information of the target cleaning device, the cleaning instruction is sent to the target cleaning device, wherein the cleaning instruction includes the information of the target cleaning device, the cleaning task and the task location information; after the target cleaning device receives the cleaning instruction from the server, it moves to the task location according to the cleaning instruction and performs the cleaning task, and can use the trained large language model to generate the cleaning instruction recognizable by the cleaning device according to the cleaning requirements, so as to instruct the cleaning device to perform the specified task, reduce the demand for manpower, and thus reduce labor costs.

[0091] In one embodiment, Figure 3 This is a schematic diagram of the first process of the cleaning method provided by an embodiment of the present application. Figure 3 As shown, the method, applied to a server, comprises:

[0092] S11: receiving cleaning request information from a terminal device.

[0093] The cleaning request information includes information of the target to be cleaned or information of the target to be cleaned and the target position.

[0094] S12: Input the cleaning request information into the trained large language model to obtain the cleaning instruction output by the trained large language model.

[0095] The cleaning instruction includes information of the target cleaning equipment, the cleaning task and the task location.

[0096] S13: Sending a cleaning instruction to the target cleaning device according to the information of the target cleaning device.

[0097] Among them, the cleaning instruction is used to instruct the target cleaning equipment to move to the task position and then perform the cleaning task. When the cleaning request information includes information about the target to be cleaned, the trained large language model is used to determine the target cleaning equipment, cleaning task and task position according to the information about the target to be cleaned, and output the cleaning instruction; when the cleaning request information includes information about the target to be cleaned and the target position, the trained large language model is used to determine the target cleaning equipment and cleaning task according to the information about the target to be cleaned, determine the task position according to the target position, and output the cleaning instruction.

[0098] The embodiment of the present application obtains cleaning request information from a terminal device, where the cleaning request information includes information about a target to be cleaned or information about a target to be cleaned and a target position; inputs the cleaning request information into a trained large language model to obtain a cleaning instruction output by the trained large language model, where the cleaning instruction includes information about a target cleaning device, a cleaning task, and information about a task position; sends a cleaning instruction to the target cleaning device based on the information about the target cleaning device, where the cleaning instruction is used to instruct the target cleaning device to move to the task position and then perform the cleaning task; wherein, when the cleaning request information includes information about the target to be cleaned, the trained large language model is used to determine the target cleaning device, the cleaning task, and the task position based on the information about the target to be cleaned, and output the cleaning instruction; when the cleaning request information includes information about the target to be cleaned and a target position, the trained large language model is used to determine the target cleaning device and the cleaning task based on the information about the target to be cleaned, and determine the task position based on the target position, and output the cleaning instruction, and uses the trained large language model to generate a cleaning instruction recognizable by the cleaning device based on the cleaning requirements, thereby instructing the cleaning device to perform a specified task, reducing the demand for manpower, and thereby reducing labor costs.

[0099] In one embodiment, Figure 4 1 is a schematic diagram of a second flow chart of a cleaning method provided in an embodiment of the present application. Figure 4As shown, the method further includes:

[0100] S14: Inputting the cleaning request information into the trained large language model to obtain the unprocessable instructions output by the trained large language model.

[0101] The trained large language model is used to determine that the size of the target to be cleaned is larger than a preset size according to the information of the target to be cleaned, or to determine an empty position according to the target position, and to output an instruction that cannot be processed.

[0102] In the application, when the cleaning request information represents a complex situation, the trained large language model will output an instruction that cannot be processed. Specifically, the complex situation includes that the size of the target to be cleaned is larger than the preset size, or the task position cannot be determined based on the target position, and an empty position is obtained, making the target cleaning device unable to process the specified task.

[0103] For example: the size of the cleaning target is larger than the preset size because there is too much sewage on the ground or the volume of garbage is too large.

[0104] S15: Sending an unprocessable instruction to the terminal device.

[0105] The unable to process instruction is used to instruct the terminal device to send a cleaning request message to the cleaning staff's device.

[0106] In the application, the terminal device sends cleaning request information to the cleaning staff's device through the software system so that the cleaning staff can complete the request requirements.

[0107] This embodiment inputs the cleaning request information into a trained large language model, obtains the unprocessable instructions output by the trained large language model, and sends the unprocessable instructions to the terminal device. When the cleaning situation is complicated and the cleaning equipment cannot complete the task, the cleaning staff can understand and complete the task, thereby ensuring that the cleaning task can be completed.

[0108] In one embodiment, Figure 5 FIG. 1 is a third flow chart of a cleaning method provided in an embodiment of the present application. Figure 5 As shown, before step S12, it also includes:

[0109] S21: Obtain training data and labels.

[0110] The training data includes first data, second data and third data. The first data includes the user's location and work content, the second data includes the location of the first camera device and the captured images, and the third data includes tasks that are automatically triggered at a preset time.

[0111] For example, the work content may be cleaning a trash can, or the work content may include the location and type of the target.

[0112] S22: Fine-tune the large language model using the training data until the value of the loss function of the large language model is less than a preset loss value.

[0113] Among them, the loss function is used to calculate the difference between the output result of the large language model and the label.

[0114] S23: When the value of the loss function is less than a preset loss value, a trained large language model is obtained.

[0115] This embodiment fine-tunes the large language model by utilizing training data. The first data includes the user's location and work content, the second data includes the location of the first camera device and the captured images, and the third data includes tasks that are automatically triggered at a preset time, thereby ensuring that the trained large language model can recognize cleaning requirements and output cleaning instructions that can be recognized by the cleaning equipment.

[0116] In one embodiment, when the target cleaning device cannot complete the cleaning instruction, the target cleaning device sends an unprocessable message to the terminal device, and the unprocessable message is used to instruct the terminal device to send a cleaning request message to the cleaning staff's device.

[0117] In this embodiment, when the target cleaning device cannot complete the cleaning instruction, the target cleaning device sends an unprocessable information to the terminal device, so that the terminal device sends a cleaning request message to the cleaning personnel's device, so that the cleaning personnel can understand and complete the task, thereby ensuring that the cleaning task can be completed.

[0118] It should be understood that the order of execution of the steps in the above embodiments does not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application. In addition, the data collection in the above embodiments is compliant, and its use or implementation does not involve any infringement on the public interest.

[0119] Corresponding to the method described in the above embodiment, for the sake of convenience of explanation, only the part related to the embodiment of the present application is shown.

[0120] In one embodiment, Figure 6 It is a schematic diagram of the structure of a cleaning device provided in one embodiment of the present application.

[0121] like Figure 6 As shown, the device comprises:

[0122] The communication module 30 is used to receive cleaning request information from the terminal device, where the cleaning request information includes information of the target to be cleaned or information of the target to be cleaned and the target position;

[0123] It is also used to send a cleaning instruction to the target cleaning device according to the information of the target cleaning device, and the cleaning instruction is used to instruct the target cleaning device to perform the cleaning task after moving to the task position;

[0124] A prediction module 31 is used to input the cleaning request information into the trained large language model to obtain the cleaning instructions output by the trained large language model, where the cleaning instructions include information of the target cleaning equipment, the cleaning task and the task location;

[0125] Among them, when the cleaning request information includes information about the target to be cleaned, the trained large language model is used to determine the target cleaning equipment, cleaning task and task location according to the information of the target to be cleaned, and output cleaning instructions; when the cleaning request information includes information about the target to be cleaned and the target location, the trained large language model is used to determine the target cleaning equipment and cleaning task according to the information of the target to be cleaned, determine the task location according to the target position, and output cleaning instructions.

[0126] In one embodiment, the device further comprises:

[0127] The acquisition module is further used to acquire training data and labels, the training data includes first data, second data and third data, the first data includes the user's location and work content, the second data includes the location of the first camera device and the collected images, and the third data includes tasks that are automatically triggered at a preset time;

[0128] A training module, used to fine-tune the large language model using the training data until the value of the loss function of the large language model is less than a preset loss value, where the loss function is used to calculate the difference between the output result of the large language model and the label;

[0129] It is also used to obtain a trained large language model when the value of the loss function is less than a preset loss value.

[0130] In one embodiment, the prediction module is further used to input the cleaning request information into the trained large language model to obtain the unprocessable instructions output by the trained large language model;

[0131] The communication module is further used to send an unprocessable instruction to the terminal device, and the unprocessable instruction is used to instruct the terminal device to send a cleaning request message to the cleaning staff's device;

[0132] The trained large language model is used to determine that the size of the target to be cleaned is larger than a preset size according to the information of the target to be cleaned, or to determine an empty position according to the target position, and to output an instruction that cannot be processed.

[0133] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0134] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0135] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.

[0136] An embodiment of the present application provides a computer program product. When the computer program product runs on an electronic device, the electronic device can implement the steps in the above-mentioned method embodiments when executing the computer program product.

[0137] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the camera / terminal device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, RandomAccess Memory), electric carrier signal, telecommunication signal and software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk. In some cases, the computer-readable medium cannot be an electric carrier signal and a telecommunication signal.

[0138] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0139] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0140] In the embodiments provided in the present application, it should be understood that the disclosed devices / network equipment and methods can be implemented in other ways. For example, the device / network equipment embodiments described above are merely schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0141] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0142] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A cleaning method, characterized in that: include: Receiving cleaning request information from a terminal device, wherein the cleaning request information includes information of a target to be cleaned or information of the target to be cleaned and a target position; Inputting the cleaning request information into a trained large language model to obtain a cleaning instruction output by the trained large language model, wherein the cleaning instruction includes information of a target cleaning device, a cleaning task, and information of a task location; According to the information of the target cleaning device, the cleaning instruction is sent to the target cleaning device, wherein the cleaning instruction is used to instruct the target cleaning device to perform the cleaning task after moving to the task position; Wherein, when the cleaning request information includes information of the target to be cleaned, the trained large language model is used to determine the target cleaning device, the cleaning task and the task location according to the information of the target to be cleaned, and output the cleaning instruction; When the cleaning request information includes the information of the target to be cleaned and the target position, the trained large language model is used to determine the target cleaning equipment and the cleaning task according to the information of the target to be cleaned, determine the task position according to the target position, and output the cleaning instruction.

2. The method according to claim 1, characterized in that: The cleaning request information is information generated by the terminal device after receiving the work content input by the user and obtaining the position of the user, wherein the work content includes information of the target to be cleaned; The cleaning request information is information generated by the terminal device after receiving an image captured by a first camera device and obtaining the position of the first camera device, wherein the image includes an imaging area of ​​the target to be cleaned, and each of the first camera devices is installed at various locations in the site; The cleaning request information is generated by the terminal device according to a task automatically triggered at a preset time, and the task includes information of the target to be cleaned; The target location includes the location of the user and the location of the first camera device.

3. The method according to claim 1, characterized in that Also includes: Inputting the cleaning request information into a trained large language model to obtain an unprocessable instruction output by the trained large language model; Sending the unprocessable instruction to the terminal device, where the unprocessable instruction is used to instruct the terminal device to send the cleaning request information to a device of a cleaning staff; The trained large language model is used to determine that the size of the target to be cleaned is greater than a preset size according to the information of the target to be cleaned, or to determine an empty position according to the target position, and to output the unprocessable instruction.

4. The method according to any one of claims 1 to 3, characterized in that: The cleaning device comprises a first cleaning robot and a second cleaning robot. The first cleaning robot comprises a mobile base and a clamp, a container and a second camera device arranged on the mobile base. The second cleaning robot is a fully automatic cleaning robot.

5. The method according to claim 4, characterized in that The information of the task location includes information of the target area and identification information representing the target location.

6. The method according to claim 5, characterized in that Before inputting the cleaning request information into the trained large language model, the method further includes: Acquire training data and labels, wherein the training data includes first data, second data, and third data, wherein the first data includes the location and work content of the user, the second data includes the location and captured images of the first camera device, and the third data includes tasks that are automatically triggered at a preset time; Using the training data, fine-tuning the large language model until a value of a loss function of the large language model is less than a preset loss value, the loss function being used to calculate a difference value between an output result of the large language model and the label; When the value of the loss function is less than the preset loss value, the trained large language model is obtained.

7. A cleaning device, characterized in that: include: A communication module, configured to receive cleaning request information from a terminal device, wherein the cleaning request information includes information of a target to be cleaned or information of the target to be cleaned and a target position; It is also used to send the cleaning instruction to the target cleaning device according to the information of the target cleaning device, and the cleaning instruction is used to instruct the target cleaning device to perform the cleaning task after moving to the task position; A prediction module, used for inputting the cleaning request information into a trained large language model to obtain a cleaning instruction output by the trained large language model, wherein the cleaning instruction includes information of a target cleaning device, a cleaning task, and information of a task location; Wherein, when the cleaning request information includes information of the target to be cleaned, the trained large language model is used to determine the target cleaning device, the cleaning task and the task location according to the information of the target to be cleaned, and output the cleaning instruction; When the cleaning request information includes the information of the target to be cleaned and the target position, the trained large language model is used to determine the target cleaning equipment and the cleaning task according to the information of the target to be cleaned, determine the task position according to the target position, and output the cleaning instruction.

8. The device according to claim 7, characterized in that Also includes: an acquisition module, used to acquire training data and labels, wherein the training data includes first data, second data and third data, wherein the first data includes the location and work content of the user, the second data includes the location and captured images of the first camera device, and the third data includes tasks that are automatically triggered at a preset time; A training module, used to fine-tune the large language model using the training data until a value of a loss function of the large language model is less than a preset loss value, wherein the loss function is used to calculate a difference value between an output result of the large language model and the label; It is also used to obtain the trained large language model when the value of the loss function is less than the preset loss value.

9. A cleaning system, characterized in that: including a server, a terminal device and at least one cleaning device; The terminal device is used to send the cleaning request information to the server after obtaining the cleaning request information; The server is configured to receive the cleaning request information from the terminal device, wherein the cleaning request information includes information of a target to be cleaned or information of the target to be cleaned and a target location; Inputting the cleaning request information into a trained large language model to obtain a cleaning instruction output by the trained large language model; According to the information of the target cleaning device, the cleaning instruction is sent to the target cleaning device, wherein the cleaning instruction includes the information of the target cleaning device, the cleaning task and the task location; The target cleaning device is used to receive the cleaning instruction from the server, move to the task position according to the cleaning instruction, and perform the cleaning task; Wherein, when the cleaning request information includes information of the target to be cleaned, the trained large language model is used to determine the target cleaning device, the cleaning task and the task location according to the information of the target to be cleaned, and output the cleaning instruction; When the cleaning request information includes the information of the target to be cleaned and the target position, the trained large language model is used to determine the target cleaning equipment and the cleaning task according to the information of the target to be cleaned, determine the task position according to the target position, and output the cleaning instruction.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images