Cleaning apparatus, dust collection processing method therefor, terminal apparatus, and program product

Abstract

The application belongs to the technical field of cleaning equipment, and provides a cleaning equipment, a dust collection processing method thereof, a terminal device and a program product. The cleaning equipment comprises: an equipment main body; a sensor assembly connected to the equipment main body, the sensor assembly collecting sensor data of at least one target object in a preset area where the equipment main body is located; and a processor arranged in the equipment main body and connected to the sensor assembly, the processor: obtaining the sensor data of the at least one target object from the sensor assembly; processing the sensor data of the at least one target object to generate characteristic information of the at least one target object; determining a hair shedding amount of the at least one target object according to the characteristic information of the at least one target object; and feeding back at least one of a state and a filling time of a dust bag according to the hair shedding amount. The cleaning equipment can avoid affecting the cleaning performance of the cleaning equipment due to the overfilling of the dust bag, and improve the intelligent level of the cleaning equipment when performing hair cleaning.

Description

Technical Field

[0001] This application belongs to the field of cleaning equipment technology, and more specifically, relates to a cleaning device and its dust collection and treatment method, terminal equipment and program product. Background Technology

[0002] In today's smart home industry, cleaning devices, especially robotic vacuum cleaners, play a vital role in household cleaning. However, current robotic vacuum cleaners on the market cannot correlate hair loss with the frequency of dust bag emptying. Summary of the Invention

[0003] The purpose of this application is to provide a cleaning device and its dust collection and treatment method, terminal device and program product, which aims to solve the problem in the prior art that sweeping robots cannot correlate hair loss with the cleaning frequency of the dust bag.

[0004] To achieve the above objectives, this application provides a cleaning device, which includes a device body, a sensor assembly, and a processor. The sensor assembly is connected to the device body and collects sensor data from at least one target object within the area where the device body is located. The processor is disposed inside the device body and connected to the sensor assembly. The processor acquires the sensor data from the sensor assembly for at least one target object, processes the sensor data for at least one target object and generates feature information for at least one target object, determines the amount of hair loss from at least one target object based on the feature information, and provides feedback on at least one of the status of the dust bag and the filling time based on the amount of hair loss.

[0005] In one possible implementation, the cleaning device further includes a memory disposed within the device body and connected to a processor. The processor also retrieves pre-stored terminal acquisition information associated with at least one target object from the memory, and processes the terminal acquisition information associated with at least one target object to generate feature information of at least one target object.

[0006] In one possible implementation, the cleaning device also includes a dust box connected to the main body of the device, wherein the processor also provides feedback on the frequency of dust collection in the dust box based on the amount of hair shed.

[0007] In one possible implementation, the processor further: acquires the cleaning frequency and target cleaning mode of the main body of the device; and when the main body of the device is operating in the target cleaning mode, determines the dust collection frequency of the dust box based on the amount of hair shed and the cleaning frequency, wherein the target cleaning mode includes at least one of sweeping mode, sweeping and mopping mode, and sweeping and mopping mode.

[0008] In one possible implementation, the processor provides feedback on at least one of the following: the cleaning area of ​​the main body of the device, the total capacity of the dust bag, and the amount of hair shed.

[0009] In one possible implementation, the processor feeds back at least one of the status of the dust bag and the filling time to the cleaning device or a terminal device that is communicatively connected to the cleaning device.

[0010] In one possible implementation, the processor further: obtains the initial hair loss amount of at least one target object; inputs the feature information of at least one target object and the initial hair loss amount into a pre-trained hair loss prediction model to obtain the hair loss amount of at least one target object output by the hair loss prediction model, wherein the hair loss prediction model is obtained by training the initial prediction model based on multiple training samples, and each training sample includes: the feature information and initial hair loss amount of the sample object, and the hair loss amount of the sample object.

[0011] In one possible implementation, the processor also acquires molting season information for at least one target object;

[0012] Obtain the initial hair loss amount of at least one target object; input the feature information, molting season information, and initial hair loss amount of at least one target object into a pre-trained hair loss prediction model to obtain the hair loss amount of at least one target object output by the hair loss prediction model. The hair loss prediction model is obtained by training the initial prediction model based on multiple training samples. Each training sample includes: the feature information, molting season information, and initial hair loss amount of the sample object, as well as the hair loss amount of the sample object.

[0013] In one possible implementation, the sensor data includes image data and radar data, and the sensor components include: an image acquisition component that acquires image data of at least one target object; and a radar component connected to the image acquisition component that acquires radar data of at least one target object when the image acquisition component acquires an image of at least one target object.

[0014] In one possible implementation, the processor further processes image data of at least one target object to generate at least one of the following: number, size, type, and hair length of the target object; and processes radar data of at least one target object to generate at least one of the following: height, volume, and shape of the target object.

[0015] In one possible implementation, the cleaning device further includes a controller and a radar component. The controller is located inside the main body of the device and connected to a processor. The controller responds to control commands received by the cleaning device and controls the cleaning device to activate the dust collection mode. The radar component, during operation after the cleaning device activates the dust collection mode, also acquires radar data of at least one target object when the image acquisition component acquires an image of at least one target object.

[0016] In one possible implementation, the controller is connected to the processor, and the controller controls the main body of the device to perform dust collection operations according to the frequency of dust collection in the dust box.

[0017] To achieve the above objectives, this application also provides a terminal device that is communicatively connected to a cleaning device. The terminal device includes a display screen and a processor. The display screen shows at least one of the dust bag status and filling time reported by the cleaning device. The processor is connected to the display screen, runs a cleaning application, and receives terminal-collected information associated with at least one target object uploaded by a user through the cleaning application, and sends the terminal-collected information to the cleaning device. The cleaning device and the cleaning application are paired and bound together.

[0018] To achieve the above objectives, this application also provides a dust collection and treatment method for a cleaning device, applicable to any cleaning device described in the present invention. The method includes: acquiring sensor data of at least one target object within a preset area where the main body of the cleaning device is located; processing the sensor data of at least one target object to generate feature information of at least one target object; determining the amount of hair loss of at least one target object based on the feature information of at least one target object; and feeding back at least one of the status of the dust bag and the filling time based on the amount of hair loss.

[0019] To achieve the above objectives, this application also provides a cleaning device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it enables the cleaning device to implement a dust collection and treatment method.

[0020] To achieve the above objectives, this application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements a dust collection method for a cleaning device.

[0021] To achieve the above objectives, this application also provides a computer program product that, when run on a cleaning device, causes the cleaning device to perform a dust collection method.

[0022] The beneficial effects of the embodiments in this application compared with the prior art are:

[0023] This application provides a cleaning device and its dust collection and treatment method, terminal device, and program product. The cleaning device includes: a device body; a sensor assembly connected to the device body, the sensor assembly collecting sensor data of at least one target object within a preset area where the device body is located; and a processor disposed inside the device body and connected to the sensor assembly. The processor: acquires sensor data of at least one target object from the sensor assembly; processes the sensor data of at least one target object to generate feature information of at least one target object; determines the amount of hair loss of at least one target object based on the feature information of at least one target object; and, based on the amount of hair loss, provides feedback on at least one of the status of the dust bag and the filling time.

[0024] The cleaning equipment uses sensor components integrated into its main body to collect real-time sensor data from target objects (e.g., indoor objects that shed hair, such as people and pets) within a preset area. A processor, located within the main body and connected to the sensor components, processes the collected sensor data to generate characteristic information about the target object. Based on this information, the processor determines the amount of hair shed from the target object and, based on the amount of hair shed, provides feedback on at least one of the following: the status of the dust bag and its filling time. By establishing a correlation between hair shed data and the frequency of dust bag cleaning, the cleaning equipment allows users to schedule dust bag cleaning based on the dust bag's status and / or filling time, thus avoiding performance issues caused by overfilled dust bags and reducing the frequency of manual dust bag checks. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is a schematic diagram of the structure of a cleaning device provided in an embodiment of this application;

[0027] Figure 2 This is a schematic diagram of the structure of an optional cleaning device provided in an embodiment of this application;

[0028] Figure 3 This is a schematic flowchart of a dust collection and treatment method for an optional cleaning device provided in an embodiment of this application;

[0029] Figure 4 This is a schematic diagram of the home management interface and information management interface of an optional cleaning application provided in an embodiment of this application;

[0030] Figure 5 This is a schematic diagram of the interaction process between an optional cleaning device and a terminal device provided in an embodiment of this application;

[0031] Figure 6 This is a schematic flowchart of another optional dust collection and treatment method for a cleaning device provided in an embodiment of this application;

[0032] Figure 7 This is a schematic diagram of the dust collection mode setting interface of an optional cleaning application provided in an embodiment of this application;

[0033] Figure 8 This is a schematic flowchart of another optional dust collection and treatment method for a cleaning device provided in an embodiment of this application;

[0034] Figure 9 This is a schematic diagram of the user feedback interface of an optional cleaning application provided in an embodiment of this application;

[0035] Figure 10 This is a schematic diagram of the user feedback interface of another optional cleaning application provided in this application embodiment;

[0036] Figure 11 This is a schematic flowchart of a dust collection and treatment method for a cleaning device provided in an embodiment of this application;

[0037] Figure 12 This is a schematic diagram of the structure of a cleaning device provided in an embodiment of this application. Detailed Implementation

[0038] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0039] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0040] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0041] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0042] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0043] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0044] In today's smart home industry, cleaning devices, especially robotic vacuum cleaners, play a vital role in cleaning for pet-owning households. However, current robotic vacuum cleaners on the market have several technical shortcomings when it comes to addressing the critical need for pet hair removal, specifically as follows:

[0045] Firstly, most robotic vacuum cleaners use fixed-capacity dustbins to collect pet hair and dust, but the fullness of the dustbin cannot be accurately determined. A full dustbin is only detected through regular manual checks or when the cleaning performance of the vacuum cleaner noticeably deteriorates. This easily leads to dustbin overloading, severely impacting the cleaning efficiency of the vacuum cleaner and, in serious cases, potentially damaging the device and shortening its lifespan.

[0046] Secondly, different breeds, sizes, and coat types of pets have significantly different amounts of hair shedding and cleaning needs. However, existing robotic vacuum cleaners cannot identify the characteristics of pets in the home, making it difficult to make intelligent predictions and management of pet hair cleaning.

[0047] Thirdly, pets shed significantly more hair during the spring and autumn molting seasons. However, existing robotic vacuum cleaners lack the ability to flexibly adjust dustbin status indicators and cleaning strategies based on seasonal changes and user cleaning frequency. During molting season, the dustbin quickly fills with pet hair, but the robotic vacuum cleaner cannot promptly remind users to empty it or automatically adjust its cleaning mode to cope with the increased hair, resulting in a poor user experience and failing to meet the cleaning needs of pet-owning families during this special period.

[0048] It should be noted that the information collection process (such as the pet image collection process, indoor environment collection process, etc.) / feature extraction process involved in this application is carried out with the user's knowledge and permission. That is, the information collection process / feature extraction process complies with the requirements of laws and regulations and does not constitute an act that harms the public interest.

[0049] This application provides an example of a cleaning device; please refer to... Figure 1 As shown, Figure 1 A schematic structural diagram of a cleaning device provided in this application is shown. The cleaning device includes a main body 100, a sensor assembly 101, and a processor 102.

[0050] A sensor assembly 100 is connected to the main body of the device and collects sensor data of at least one target object within the area where the main body is located. A processor is disposed inside the main body of the device and connected to the sensor assembly. The processor acquires sensor data of at least one target object from the sensor assembly, processes the sensor data of at least one target object and generates feature information of at least one target object, determines the amount of hair loss of at least one target object based on the feature information of at least one target object, and provides feedback on at least one of the status of the dust bag and the filling time based on the amount of hair loss.

[0051] Optional, such as Figure 2 As shown, the cleaning equipment includes a main body 100 and a base station 200. The base station is equipped with a dust bag for collecting the waste dumped from the dust box in the main body. The main body 100 performs cleaning tasks within a preset area. After completion, it can carry the dust box back to the base station 200 and transfer the waste in the dust box to a dust bag located on the side of the base station 200.

[0052] In this application example, the sensor component 101 is disposed at a specific location on the main body of the cleaning device (such as a sweeping robot, a sweeping and mopping robot, a sweeping and mopping robot, or a sweeping and mopping robot) 100, for example, on the top or front end of the main body 100, or as before. Figure 2 As shown, the sensor assembly 101 is installed at the front end of the device body 100 to ensure that sensor data of the target object can be clearly collected during the cleaning process.

[0053] In one embodiment of this application, the sensor component may include an image acquisition component, a radar component, etc. For example, the cleaning device may be equipped with an image acquisition component such as an RGB camera to acquire visual image data of at least one target object (e.g., a human, a pet, etc.) in the room where the cleaning device is located. The distribution of different pixel values ​​in the visual image data reflects the appearance, color, outline, and other features of at least one target object. Alternatively, a radar component such as a lidar may be installed on the cleaning device to acquire radar data of at least one target object when the image acquisition component acquires an image of at least one target object, thereby obtaining data such as the height, volume, and shape of each target object.

[0054] In one optional application scenario, during operation, the cleaning device's sensor components scan a preset area (using 3D scanning technology) where the device itself is located. In one embodiment, the preset area can be the entire room where the cleaning device is located, a user-specified cleaning area, or a cleaning area planned centered on the device itself. For example, when the device is cleaning the living room, it collects data on target objects within the living room area.

[0055] In one example, while the cleaning device moves along a planned path within an indoor space and performs cleaning operations, it simultaneously controls sensor components to collect data at a predetermined frequency (e.g., once every 5 seconds) and angular range (180°, 360°, etc.) to ensure that sensor data of at least one target object in the indoor space can be collected. For example, taking a sensor component that includes an image acquisition component and a radar component as an example, when the device moves within a user's bedroom to perform cleaning operations, the image acquisition component (e.g., an RGB camera) captures images of people and / or pets in the bedroom, obtaining image data of the people and / or pets; while the image acquisition component collects image data of the people and / or pets, the radar component (e.g., a lidar) continuously scans for radar data of the people and / or pets, thereby collecting sensor data of the people and / or pets in the room.

[0056] In this application example, the sensor component is connected to the processor. The sensor component transmits the sensor data (including image data, radar data, etc.) of at least one target object to the processor in real time. The processor inputs the sensor data of at least one target object into the pre-trained feature recognition model and receives the feature recognition model outputting feature information corresponding to the sensor data, such as quantity, size (small, medium, large), age (0.5 years, 5 years), species (human or pet), and hair length (roughly classified based on a predetermined pet hair threshold: long hair, short hair).

[0057] It should be noted that the processor can pre-use deep learning algorithms in the field of computer vision, and use multiple training samples containing sensor data of the target object and corresponding feature information (using labels to annotate the feature information of the target object, such as type, quantity, size, height, volume, hair length, etc.) to train the initial neural network model, thereby constructing a feature recognition model that can accurately identify the feature information of the target object.

[0058] Furthermore, the processor can pre-train an initial prediction model based on multiple training samples to obtain a hair loss prediction model. Each training sample includes: feature information of the sample object and the amount of hair loss of the sample object. Specifically, the hair loss prediction model calculates the amount of hair loss of the target object based on the input feature information of the target object and a prediction mechanism obtained by learning the feature information and hair loss amount of the sample object using a machine learning algorithm. For example, for a long-haired cat, the hair loss prediction model calculates the amount of hair loss of the long-haired cat based on its body size, hair length, and other feature information, combined with the feature information and hair loss amount of similar long-haired cats in the training samples.

[0059] The cleaning equipment provided in this application example is suitable for any location, such as homes, conference rooms, shopping malls, and service halls, and is especially suitable for pet-owning households. It can accurately identify multiple pets or humans in the household simultaneously. Taking pets as an example, based on the unique characteristics of each pet, such as the long-haired cat's hair being long and prone to tangling, and the short-haired dog's hair being short but shedding a lot during molting season, the equipment performs detailed calculations of hair loss and determines and reports at least one of the following: the status of the dust bag and the filling time. This improves the applicability and intelligence of the cleaning equipment in pet-owning households, meeting the complex and diverse cleaning needs of multi-pet households.

[0060] Moreover, as time goes by and the environment changes, the processor can update and adjust the amount of hair loss from the target object in real time based on the newly collected feature information and environmental information, so as to ensure the accuracy and dynamic adaptability of the cleaning equipment.

[0061] After determining the amount of hair shed, the processor provides feedback on at least one of the dust bag's status and filling time based on the hair shed amount. This allows users advance preparation time to replace the dust bag in a timely manner, ensuring the normal operation of the cleaning equipment. For example, the processor can combine the dust bag's capacity information with feedback on at least one of the dust bag's status and filling time based on the hair shed amount.

[0062] In one example, the processor of the cleaning device provides feedback on the dust bag filling time based on the amount of hair shed: assuming the dust bag capacity is a known fixed value, the time required for the dust bag to fill from the current time point is estimated by calculating the amount of hair shed by the main body of the cleaning device per unit time. For example, if the dust bag capacity is 500 grams, the currently calculated hair shed per unit time is 10 grams / hour, and the dust bag currently contains 100 grams of hair, then the feedback indicates that the dust bag will need another 40 hours to fill.

[0063] In one optional example, the cleaning device's built-in voice module can be used to issue voice prompts, such as "Hello, the dust bag is expected to fill within 24 hours, please pay attention." Furthermore, the frequency of the prompts can be gradually increased as the filling time approaches, such as broadcasting again at 12 hours, 6 hours, and 1 hour remaining, and the content of the broadcasts can be more urgency-based, such as "Please note that the dust bag is expected to fill within 1 hour, please handle it as soon as possible."

[0064] In addition to voice prompts, another optional example is to set different sound effects to indicate different dust bag filling states. For example, when the dust bag is expected to fill in 24-48 hours, the cleaning device emits a gentle prompt sound, such as a "ding-dong" sound; when there are 24-12 hours remaining, the prompt sound changes to a slightly faster "beep-beep-beep" sound; if there are less than 12 hours remaining, the prompt sound becomes a rapid, continuous beeping sound to attract the user's attention.

[0065] In another example, the processor of the cleaning equipment provides feedback on the dust bag's status based on the amount of hair shed: assuming the dust bag's capacity is a known fixed value, a pre-determined percentage of the dust bag's load capacity is less than 30% for a lightly filled state, greater than 30% but less than 70% for a moderately filled state, and greater than 70% for a heavily filled state. For example, if the dust bag's maximum hair capacity is 500 grams, and the cleaning equipment calculates the amount of hair shed per unit time, currently calculated to be 10 grams per hour, and the dust bag currently contains 300 grams of hair, the current load capacity is calculated to be 60% (i.e., 300 grams divided by 500 grams, then multiplied by 100%), and the cleaning equipment then determines that the dust bag is currently in a moderately filled state.

[0066] Subsequently, the cleaning device, through its built-in voice broadcast module, clearly and accurately informs the user, "Dear user, the current dust bag filling level is 60%, which is in a medium filling state. Please pay attention to subsequent usage." Simultaneously, the dust bag filling status can be displayed on the cleaning device's LED screen in the form of an intuitive progress bar. The progress indicator line precisely points to the 60% position on the progress bar, and the words "Medium Fill" are prominently marked next to the progress bar for easy user monitoring.

[0067] In another example, the cleaning device's processor provides feedback on the dust bag's status and filling time based on the amount of hair shed: assuming the dust bag capacity is a known fixed value, a pre-defined state is defined as a lightly filled dust bag if the occupied percentage is less than 30%, a moderately filled dust bag if the occupied percentage is greater than 30% but less than 70%, and a heavily filled dust bag if the occupied percentage is greater than 70%. For example, in a pet-owning household, the cleaning device detects that the pet is in molting season, shedding 20 grams of hair per hour. The total dust bag capacity is 600 grams, and the dust bag has currently collected 100 grams of hair. The calculated filling percentage is approximately 16.7%, determining the dust bag to be in a lightly filled state.

[0068] Next, the cleaning equipment calculates the dust bag filling time. First, it calculates the remaining capacity of the dust bag as 600 grams minus 100 grams, which is 500 grams. Then, based on the current hair shedding rate of 20 grams per hour, it calculates that the dust bag filling time is 25 hours by dividing 500 grams by 20 grams per hour.

[0069] To provide users with comprehensive and accurate information about the dust bag, the cleaning device can seamlessly connect with a smart speaker to inform users in a clear and natural voice broadcast: "Hello user, the dust bag is currently lightly filled, with a filling percentage of approximately 16.7%. The dust bag is expected to be full in 25 hours." Simultaneously, the device management interface on the smart home control screen displays the dust bag filling progress bar in an intuitive visual chart format. The progress indicator line precisely points to the 16.7% position on the progress bar, and the text "Lightly Filled" and "Expected to be Full in 25 Hours" are prominently displayed next to the progress bar, allowing users to easily obtain detailed information about the dust bag's status and filling time on the control screen.

[0070] Another alternative is to install LED indicator lights on the cleaning equipment or base station. For example, different colors and flashing frequencies can be used to indicate the status of the dust bag and the time it will take to fill. A solid green light indicates that the dust bag is not full and there is still a long time before it is full; a flashing yellow light indicates that the dust bag is expected to fill within 24-48 hours; and a rapidly flashing red light indicates that the dust bag is expected to fill within 24 hours.

[0071] Another alternative is that if the cleaning device is equipped with a display screen, the screen can show a graphical interface indicating the dust bag filling progress bar, with the estimated filling time indicated next to the bar. For example, the progress bar might show 80% completion, with "estimated to fill in 12 hours" displayed next to it. Simultaneously, the display screen can also show historical dust bag filling data, hair collection volume in different areas, and other relevant information, allowing users to easily understand the cleaning progress.

[0072] As an optional embodiment, in addition to the cleaning device feeding back at least one of the state of the dust bag and the filling time, the feedback can also be performed by a terminal device paired and bound with the cleaning device. For example, a user communicatively connects to the cleaning device through a cleaning application (APP) running on a terminal device such as a smartphone. When at least one of the state of the dust bag and the filling time changes, a push notification message can be sent to the smartphone through the cleaning APP, and the smartphone displays at least one of the state of the dust bag and the filling time on the display screen. It should be understood that in addition to at least one of the state of the dust bag and the filling time, the notification message may also include some suggestions for the user. For example, the notification message may be: "The dust bag is currently in a lightly filled state, and is expected to be full in 36 hours. It is recommended that you prepare a new dust bag in advance for timely replacement". The user can click the notification to enter the detailed page of the APP to view information such as the state of the dust bag, the dynamic change of the filling time, the cleaning history, and the device state.

[0073] There is still another optional embodiment, where the cleaning device can also adopt a feedback mode linked with a smart home system (smart speaker, smart home central control screen, etc.). For example, if the user is equipped with a smart speaker at home, the cleaning device can be paired and bound with the smart speaker to achieve linkage. When it is predicted that at least one of the state of the dust bag and the filling time meets a set condition, the smart speaker automatically broadcasts relevant information. For example, the user sets to be notified when the dust bag is expected to be full within 12 hours, and the smart speaker can give a voice prompt "The dust bag is expected to be full within 12 hours, please deal with it in time". The user can also ask the smart speaker for detailed information about the dust bag through a voice command, such as "Xiao X, what is the current state of the dust bag of my sweeping robot", and the smart speaker obtains real-time data according to the connection with the cleaning device and answers the user: "The dust bag is currently in a lightly filled state".

[0074] It should be noted that the present disclosure is not only applicable to various pet-raising families, which can make accurate predictions according to the hair shedding conditions of different pets, and can timely remind the user of at least one of the state of the dust bag and the filling time during the pet molting season, so as to ensure the cleaning effect. It is also applicable to families or places with high cleaning requirements, such as families with people prone to allergies. By accurately feeding back at least one of the state of the dust bag and the filling time, the dust bag can be cleaned in time to reduce secondary pollution.

[0075] The cleaning device provided in this application example, through the cooperation of the device body, sensor components, processor, and dust bag installed on the base station side, achieves intelligent feedback on the estimated dust bag filling time. Compared to traditional cleaning devices, the cleaning device provided in this application example, through the sensor components installed in the device body, collects sensor data in real time from target objects (e.g., indoor people and pets that shed hair) within a preset area. The processor, located inside the device body and connected to the sensor components, processes the collected sensor data to generate characteristic information of the target object, and determines the amount of hair shed from the target object accordingly. Based on the amount of hair shed, it then provides feedback on at least one of the dust bag status and filling time. By establishing a correlation between hair shed and dust bag cleaning frequency, the cleaning device allows users to rationally schedule dust bag cleaning times based on the dust bag status and / or filling time fed back by the cleaning device. This avoids affecting the cleaning performance of the cleaning device due to overfilled dust bags and reduces the frequency of manual dust bag checks by the user.

[0076] One possible implementation is, such as Figure 3 As shown, the cleaning device also includes a memory 103. Optionally, the memory 103 is located inside the device body 100 and connected to the processor 102, wherein, as... Figure 3 In the flowchart shown, the processor 102 also retrieves terminal acquisition information pre-stored in the memory that is associated with at least one target object, and processes the terminal acquisition information associated with at least one target object to generate feature information of at least one target object.

[0077] The cleaning device stores terminal-collected information associated with at least one target object in its memory (such as an internal storage unit like a hard drive or flash memory). In one specific example, a structured data format, such as a database table structure, can be used for storage, with the target object's identifier (such as a pet's name) as the primary key, facilitating quick retrieval and access later. Furthermore, to ensure data security, the stored data can be encrypted.

[0078] In one implementation, the information collected by the terminal can be various information about the target object entered by the user through the user interface of a cleaning application (e.g., a cleaning device management application) running on a terminal device (e.g., a smartphone, tablet, or smartwatch), such as the target object's breed, age, gender, health status, fur type, and daily activity area. This cleaning application is associated with, paired with, or bound to the cleaning device used by the user.

[0079] For example, users can... Figure 4Clicking the "Member Information" or "Pet Information" control in the family management interface will redirect you to the corresponding information management interface, where you can enter information.

[0080] It should be noted that users can add or enter the aforementioned terminal-collected information through the cleaning application running on the terminal device when using the cleaning device for the first time or when configuring the cleaning device; alternatively, users can update the configuration information of the cleaning device in real time through the cleaning application running on the terminal device and re-add or enter the aforementioned terminal-collected information during the use of the cleaning device.

[0081] In one example, still as Figure 4 As shown, after the user clicks the "Member Information" control in the home management interface of the cleaning application, the user's terminal device is redirected from the home management interface to... Figure 4 The member information management interface shown on the left allows users to add at least one family member's information through manual input, voice input, or by selecting the matching box. This information includes the member's age (e.g., 25 years old, 30 years old), gender (male, female), special group (elderly, pregnant women, children, etc.), and hair length (roughly classified based on a predetermined human hair threshold: long hair, short hair, medium-long hair).

[0082] In another example, still as Figure 4 As shown, after the user's terminal device (e.g., smartphone) clicks the "Pet Information" control in the home management interface of the cleaning application, it redirects from the home management interface to... Figure 4 The pet information management interface shown on the right allows users to add pet information such as the number of pets they own, the type or breed of each pet (cat, dog, rabbit, etc.), fur length (roughly classified based on a predetermined pet fur threshold: long-haired, short-haired), age (0.5 years), and size (small, medium, large) through manual input, voice input, or selecting the matching box.

[0083] In addition, in this application example, the information collected by the terminal can also be input through the operation interface of the terminal device itself that is paired with or associated with the cleaning device, and stored in the cleaning device in a structured data form, so as to facilitate subsequent reading and use by the processor of the cleaning device.

[0084] It should still be understood that if the user modifies or supplements the previously collected terminal information during the cleaning process, the cleaning application can promptly synchronize the updated terminal information to the storage location. The cleaning device will periodically check for user updates; if so, it will re-acquire the latest terminal information to ensure that the cleaning device uses the most accurate terminal information.

[0085] After reading the terminal-collected information associated with the target object, the processor can specifically provide this information as input to a trained feature recognition model. Based on the patterns and rules learned during training, the feature recognition model outputs the target object's feature information. For example, if the terminal-collected information indicates a pet breed as a "Golden Retriever," the feature recognition model can identify characteristics such as the Golden Retriever's large size, moderate-length and golden fur. If the terminal-collected information indicates a pet as a "puppy," the feature recognition model can determine that the pet is in a growth stage, and its fur length and shedding amount differ from adult pets.

[0086] Through the example in this application, based on the characteristic information of the target object generated from the terminal collecting information of the target object, the processor can more accurately estimate the amount of hair loss of the target object. By inputting the characteristic information of the target object into a pre-established hair loss prediction model, and combining it with other environmental factors (such as season, temperature, etc.), a more accurate amount of hair loss can be obtained. The processor can optimize various operations of the cleaning equipment, such as predicting the filling time of the dust bag, adjusting the dust collection frequency of the dust box, planning the cleaning path (focusing on cleaning areas with more hair), and adjusting the cleaning intensity, thereby improving the working efficiency of the cleaning equipment.

[0087] In another optional embodiment, the processor can further process sensor data and terminal acquisition information of at least one target object to generate feature information of at least one target object. After receiving sensor data of at least one target object sent by the sensor component, the processor obtains pre-stored terminal acquisition information in the memory and associates the sensor data and terminal acquisition information of the same target object obtained by the sensor component to generate feature information of the target object based on the sensor data and terminal acquisition information of the target object.

[0088] For example, sensor data includes images of the target object captured by a camera, and information such as the height and volume of the target object measured by a sensor. The information collected by the terminal may include descriptions of the target object's type, hair length, age, etc. The processor associates this data from different sources with the corresponding target object to enhance the accuracy of identifying the target object's feature information.

[0089] In another optional embodiment, the processor uses terminal-collected information as supplementary information to assist the cleaning device in identifying the target object's characteristic information based on the sensor data collected. For example, if the terminal-collected information indicates that the target object is a short-haired dog, the processor can more specifically identify the parts of the sensor data that match the characteristics of a short-haired dog when analyzing the sensor data, and at the same time perform more detailed feature refinement identification on these matching parts. As another example, the pet's age entered by the user also affects the amount of hair loss; the processor can take the pet's age entered by the user as a factor when calculating the amount of hair loss.

[0090] By comprehensively analyzing sensor data and information collected from the terminal, the processor of the cleaning equipment can more accurately identify the characteristics of the target object, avoiding the errors and uncertainties caused by relying on a single data source. For example, if the terminal data indicates that a pet in the user's home is a long-haired cat, and the image in the sensor data shows the approximate size of the cat, the processor can combine the two information to more accurately identify the long-haired cat's fur length, specific body shape, and other characteristics. At the same time, the processor can also determine whether the long-haired cat is in a special period (such as pregnancy, illness, or other periods that may affect the amount of hair loss) based on the terminal data, thereby more accurately predicting the amount of hair loss of the long-haired cat.

[0091] Furthermore, it should be noted that, considering the individual differences in the information collected by the terminal, the cleaning equipment can provide more personalized services to different users. For example, although different users may have the same breed of pet, they may be at different physiological stages. By collecting information through the terminal, the cleaning equipment can provide targeted services based on each user's specific situation.

[0092] As an optional embodiment, the cleaning device also includes a dust box connected to the main body of the device, wherein the processor also provides feedback on the frequency of dust collection in the dust box based on the amount of hair shed.

[0093] In one example, the dustbin can be located at the bottom or middle of the cleaning device. The frequency of dust collection in the dustbin refers to how often the cleaning device returns to the base station (e.g., a charging station) during the cleaning process to perform a dust collection operation (automatically emptying the dustbin's debris into a dust bag located on the base station side). This can be measured by the number of times the device returns to the base station per unit time, or by different levels.

[0094] For example, when the amount of hair falling out is higher than the predetermined amount, it indicates that the current filling speed of the dust box is greater than the preset normal filling speed. At this time, the processor can increase the frequency of dust collection in the dust box to shorten the time interval between two dust collection cycles of the cleaning equipment. Conversely, when the amount of hair falling out is lower than the predetermined amount, the processor can reduce the frequency of dust collection in the dust box of the cleaning equipment to extend the time interval between two dust collection cycles of the cleaning equipment.

[0095] For example, the dust collection frequency of the dust box can be divided into three levels: high, medium, and low. Different levels correspond to different ranges of hair loss. For instance, the level of the dust collection frequency can be determined based on the range of hair loss.

[0096] In this example, the dust collection frequency of the dust box is not fixed, but can be adjusted in real time according to the dynamic estimate of hair loss. If the amount of hair loss detected by the cleaning equipment suddenly increases during the cleaning process, the cleaning equipment can automatically increase the dust collection frequency of the dust box to avoid the dust box becoming too full and affecting the cleaning effect.

[0097] It should be noted that excessively high re-piling frequency can cause the cleaning equipment to frequently interrupt its cleaning operation, reducing cleaning efficiency; while excessively low dust collection frequency may affect the dust collection effect. Therefore, the cleaning equipment provided in this application example takes into account both cleaning efficiency and dust collection timeliness when determining the dust collection frequency of the dust box, and finds a suitable balance point through continuous optimization to achieve efficient cleaning and dust collection.

[0098] In this application example, the cleaning device accurately predicts the amount of hair loss based on the characteristic information of at least one target object, and then adjusts the dust collection frequency of the cleaning device's dustbin accordingly. This not only improves the cleaning effect of the cleaning device on hair, meeting diverse cleaning needs, but also allows for understanding the dustbin's usage based on the dust collection frequency, avoiding the problem of decreased cleaning efficiency due to a full dustbin, reducing the risk of damage to the cleaning device due to dustbin overload, and thus extending the service life of the cleaning device. Furthermore, this enhances the applicability and intelligence level of the cleaning device, providing users with a more efficient and personalized cleaning service experience.

[0099] In one possible implementation, after determining the dust collection frequency of the cleaning device's dustbin, the processor can send the corresponding control command to the cleaning device's controller (or control chip). The controller controls the cleaning device's movement path and return-to-base behavior according to the control command. When the set return-to-base dust collection time point is reached or the set trigger condition is met (such as the weight or volume of hair in the dustbin reaching a predetermined threshold), the cleaning device automatically stops executing the current cleaning task, returns to the base station according to the preset return-to-base path to perform dust collection, and then continues to execute the cleaning task after completing the dust collection. The user does not need to directly check the dustbin to achieve automatic dustbin management, effectively avoiding the problem of reduced cleaning efficiency or abnormal operation of the cleaning device due to a full dustbin.

[0100] One possible implementation is, such as Figure 5 As shown, processor 102 also:

[0101] The cleaning frequency and target cleaning mode of the main body of the equipment are obtained; and when the main body of the equipment is running in the target cleaning mode, the dust collection frequency of the dust box is determined based on the amount of hair shed and the cleaning frequency.

[0102] In this application example, the cleaning device can record the cleaning frequency of the device body during operation. This cleaning frequency can be determined in various ways, such as recording the number of cleaning operations completed within a certain time period, or statistically analyzing the area cleaned per unit time. It should be understood that the cleaning frequency of the device body can be stored in the cleaning device's memory, thus facilitating processor retrieval.

[0103] In one embodiment of this application, the target cleaning mode includes at least one of a sweeping mode, a sweeping and mopping mode, and a sweeping-then-mopping mode. When the device is started, the current cleaning mode can be preset and recorded, and the processor can obtain the current cleaning mode of the device from the device's operating status information.

[0104] Different cleaning modes of the device have different effects on the dustbin. For example, in sweeping mode, the main focus is on collecting dust and hair, and the dustbin is primarily filled with these items. In sweeping-and-mopping mode, some hair and dust are carried away by the mop or adhered to it, affecting the dustbin's filling speed. In sweeping-then-mopping mode, the dustbin first concentrates on collecting dust and hair. In this mode, the sweeping logic primarily relies on the dustbin to collect dust and hair, while the subsequent mopping has almost no impact on the dustbin.

[0105] Based on the above example, the amount of hair loss determined by the processor reflects the hair loss status of the target object in its current state, and this hair loss status is one of the factors affecting the dustbin's filling speed. Additionally, the cleaning frequency also affects the dustbin's filling speed. A high cleaning frequency means the cleaning device can collect more hair per unit time, making the dustbin fill more easily and reducing the filling time compared to normal conditions. Conversely, a low cleaning frequency means the cleaning device can only collect less hair per unit time, making the dustbin less likely to fill and extending the filling time compared to normal conditions.

[0106] In one example, the processor can use mathematical formulas or experience-based decision tables to determine the dust collection frequency based on hair loss, cleaning frequency, and target cleaning mode. For instance, when the cleaning device is in sweeping mode, the dust collection frequency F1 is related to the hair loss H and cleaning frequency S as F1 = k1 * H * S, where k1 is a coefficient corresponding to the sweeping mode derived from experience or experimentation, and its specific value is based on experience or experimentation. As another example, when the cleaning device is in sweeping-and-mopping mode, the dust collection frequency F2 is F2 = k2 * H * S * m, where k2 is another coefficient corresponding to the sweeping-and-mopping mode derived from experience or experimentation, and its specific value is based on experience or experimentation, and m is... An adjustment factor less than 1 is used to account for the situation where the mop carries away some dust and hair. For example, when the cleaning device is in the sweep-then-mop mode, an initial frequency is first calculated according to the sweeping mode, and then the initial frequency is appropriately adjusted according to the impact of the subsequent mopping operation. This can be obtained as F2 = k3 * H * S * n, where k3 is another coefficient corresponding to the sweep-then-mop mode obtained based on experience or experiments, and n is an adjustment factor that comprehensively considers the impact of mopping. The value of n can be between 0 and 1, and the specific value is determined based on actual testing and experience.

[0107] In this application example, the above-mentioned optional implementation comprehensively considers several important factors related to the cleaning equipment, including hair shedding, cleaning frequency, and cleaning mode. This makes the final determined dust collection frequency more reasonable and accurate, improving the working efficiency of the cleaning equipment and the scientific nature of dust box management. Furthermore, it can be flexibly adjusted according to different cleaning modes, avoiding the limitations of considering only one factor and better adapting to different usage scenarios and user needs.

[0108] One possible implementation is still as follows Figure 5 As shown, the processor of the cleaning equipment provides feedback on at least one of the following: the cleaning area of ​​the main body of the equipment, the total capacity of the dust bag, and the amount of hair shed. It also provides feedback on the status of the dust bag and the filling time.

[0109] In one embodiment, the dust bag of the cleaning device can be set at the location of a base station such as a charging pile to hold the hair, dust and other garbage dumped from the dust box in the cleaning device. The base station has more space than the main body of the device, which allows the dust bag to be designed with a large capacity to hold more hair, dust and other garbage transferred from the dust box, greatly reducing the frequency of users cleaning the garbage in the dust box.

[0110] Optionally, the total capacity of the dust bag is a fixed value predetermined based on the design specifications, which can be obtained from the cleaning equipment's product manual, hardware label, or system settings.

[0111] Optionally, the cleaning area of ​​the device is closely related to the amount of debris generated. Generally, the larger the cleaning area, the more debris is collected. During the cleaning process, the device can use components such as LiDAR and vision sensors to create a cleaning map in real time and record the area cleaned. For example, as the robot vacuum moves in a room, its sensors continuously scan the surrounding environment, build a map, and calculate the cleaning area, accurately recording the cleaning area in square meters.

[0112] In one alternative example, the processor first determines the total capacity of the dust bag to hold hair and other debris. Simultaneously, the processor analyzes sensor data and / or terminal-collected information to obtain the real-time amount of hair shed from the target object. It should be understood that real-time acquisition is necessary because the amount of hair shed from the target object changes in real-time with factors such as the environment and the state of the target object's hair. Then, based on at least one of the total amount of debris the dust bag can hold, the cleaning area of ​​the main body of the device, and the currently estimated amount of hair shed, the dust bag filling time can be predicted. It should be understood that because the amount of hair shed from the target object is constantly changing, the dust bag filling time can also be adjusted accordingly in real-time.

[0113] In one example, the cleaning device provides feedback on at least one of the following: the dust bag status and the filling time, based on the cleaning area and the total capacity of the dust bag:

[0114] During operation, the cleaning equipment uses LiDAR and visual sensors to build a real-time map and accurately record the cleaning area of ​​the equipment. For example, after two hours of continuous cleaning, the cleaning area reaches 40 square meters. Assuming, based on past experience, the average amount of hair shed per square meter in this area is 5 grams (ignoring other debris), and given that the total capacity of the dust bag is 300 grams, the total amount of debris generated can be calculated as 40 square meters × 5 grams / square meter = 200 grams based on the cleaned area and the amount of debris generated per unit area. Therefore, the dust bag's filling ratio is approximately 200 grams ÷ 300 grams × 100% ≈ 66.7%. The processor, based on preset values ​​indicating a dust bag's fill ratio of 0-30% (low filling), 30%-70% (medium filling), and 70%-100% (high filling), determines that the dust bag is currently in a high-fill state.

[0115] Assuming the cleaning speed of the sweeping equipment remains stable, cleaning an area of ​​20 square meters per hour, and generating 5 grams of debris per square meter, then the amount of debris generated per hour is 20 square meters × 5 grams / square meter = 100 grams. The remaining capacity of the dust bag is 300 grams - 200 grams = 100 grams. Therefore, the time to fill the dust bag is calculated to be 100 grams ÷ 100 grams / hour = 1 hour. The processor prominently displays the dust bag status as "highly filled" on the sweeping equipment's display screen, and also shows the time to fill the dust bag as "1 hour later." It also provides relevant information to the user through voice prompts so that the user can handle the situation promptly.

[0116] In one example, the cleaning device provides feedback on the status of the dust bag based on at least one of the amount of hair shed and the filling time:

[0117] For example, analysis estimates that a pet sheds 30 grams of hair per hour over a given period. Historical data also indicates that other waste is approximately 0.5 times the amount of hair, meaning that other waste is generated at 30 grams × 0.5 = 15 grams per hour. Therefore, the total waste generated per hour is 30 grams + 15 grams = 45 grams. Assuming the dust bag has a total capacity of 450 grams, after a period of operation (e.g., 5 hours), the total waste generated will be 45 grams / hour × 5 hours = 225 grams. The dust bag's filling percentage is 225 grams ÷ 450 grams × 100% = 50%, at which point the processor determines the dust bag is at a moderate filling level.

[0118] Based on the current waste generation rate of 45 grams per hour, the remaining capacity of the dust bag is 450 grams - 225 grams = 225 grams. Therefore, the time to fill the dust bag is calculated to be 225 grams ÷ 45 grams / hour = 5 hours. The processor sends a notification message to the user via an app connected to their mobile phone, informing them that the dust bag is moderately full and expected to be full in 5 hours. The app also displays the dust bag filling progress and estimated filling time in a chart format.

[0119] In another example, the cleaning device provides feedback on at least one of the following: the total capacity of the dust bag, the amount of hair shed, and the filling time. The device monitors the amount of hair shed by the user's three pet dogs in real time, estimating the hair shed per hour to be 20 grams. Other debris is estimated to be 0.4 times the amount of hair, meaning the hourly amount of other debris is 20 grams × 0.4 = 8 grams, and the hourly total debris is 20 grams + 8 grams = 28 grams. The total capacity of the dust bag is known to be 280 grams. After running for a period of time (e.g., 3 hours), the total debris generated is 28 grams / hour × 3 hours = 84 grams. The dust bag filling percentage is 84 grams ÷ 280 grams × 100% = 30%. The processor then determines that the dust bag is currently in a moderately filled state based on a preset standard.

[0120] The processor announces through the built-in voice module of the cleaning device: "The dust bag is currently in a medium filling state." At the same time, the device's LED display shows the dust bag's 30% filling status in the form of a progress bar, marked with the words "medium filling," allowing users to intuitively understand the status of the dust bag.

[0121] In another alternative example, the cleaning device provides feedback on at least one of the following: the dust bag status and the filling time, based on the cleaning area, the amount of hair shed, and the total capacity of the dust bag.

[0122] First, determine the cleaning area of ​​the sweeping equipment and the expected amount of hair shed during a given period. Dividing the hair shed during this period by the cleaning area gives the average hair shed per unit area. Next, multiplying the hair shed per unit area by the hourly cleaning area yields the total amount of hair and debris generated per hour. Finally, since the total capacity of the dust bag is known and fixed, dividing it by the hourly debris generation gives the dust bag filling time. This method, considering the sweeping area of ​​the equipment, the total dust bag capacity, and the hair shed, allows for a relatively accurate prediction and feedback of at least one of the dust bag's status and filling time. This facilitates timely feedback to the user, prompting them to prepare for dust bag replacement in advance and providing a basis for rationally scheduling dust bag replacement, ensuring the continuous and efficient operation of the sweeping equipment.

[0123] Another possible example is that the filling time of the dust bag can be affected by factors such as the cleaning area and the simple filling time of different types of waste. For example, the dust filling time (pure dust sweeping) / cleaning area represents the average filling time required for dust to accumulate to fill the dust bag per unit cleaning area in a pure dust sweeping scenario. Another example is the short hair filling time / cleaning area, which refers to the average filling time for short hair to accumulate to fill the dust bag per unit cleaning area when sweeping short hair. This reflects the accumulation characteristics of short hair in the cleaning area and provides a basis for formulating cleaning strategies for short hair. Yet another example is the long hair filling time / cleaning area, which refers to the average filling time for long hair to accumulate to fill the dust box or dust bag per unit cleaning area. This reflects the accumulation pattern of long hair and helps optimize the working mode of the cleaning equipment based on the characteristics of long hair. Therefore, the processor can perform data fusion processing based on the average filling time determined separately by the cleaning area and the filling time of different types of waste, and finally determine at least one of the dust bag status and filling time.

[0124] One possible implementation is still as follows Figure 3 As shown, the processor provides feedback to the cleaning equipment or a terminal device that is communicatively connected to the cleaning equipment, including at least one of the status of the dust bag and the filling time.

[0125] It should be noted that the processor feeds back to the cleaning device at least one of the dust bag status and filling time. Please refer to the above-mentioned optional examples, which will not be repeated here.

[0126] Optionally, the terminal device runs a cleaning application, and the cleaning device is paired and bound to the cleaning application. In one example, after the processor anticipates the dust bag's filling time, it can send feedback to the cleaning device or the terminal device connected to the cleaning device at time points such as 12 hours, 6 hours, 2 hours, 1 hour, or 30 minutes before the estimated filling time, to remind the user in advance. This allows the user to know the dust bag's status and at least one of the filling time before the dust bag is full.

[0127] In one embodiment of this application, the terminal device may use various forms to provide feedback on at least one of the dust bag status and filling time, such as text messages, voice messages, etc., to display the dust bag status and filling time on the user interface of the cleaning application on the terminal device, or to display the dust bag status and filling time on the main interface of the terminal device.

[0128] It should be noted that a communication connection is pre-established between the processor and the terminal device. Specifically, this communication connection is implemented through various communication protocols, such as Bluetooth, Wi-Fi, or other wireless communication protocols. Furthermore, the cleaning device can send at least one of the dust bag's status and filling time to the paired terminal device.

[0129] In one optional application scenario, the cleaning device can have a companion cleaning application (e.g., a cleaning device management application), which users can download and install on terminal devices such as smartwatches and tablets. The processor sends at least one of the dust bag's status and filling time to this cleaning application, which then receives and displays it to the user. For example, once the processor determines at least one of the dust bag's status and filling time, it sends this information to the cleaning application on the user's mobile phone via Wi-Fi.

[0130] In another example, the cleaning device can send at least one of the dust bag status and filling time to an application running on a terminal device, or it can send these messages directly to the connected terminal device via near-field communication (NFC) or SMS notification. For instance, if the user can also connect the cleaning device to a smartwatch, the processor can send both the dust bag status and filling time to the user's smartwatch simultaneously with the cleaning application running on the smartphone. This allows the user to conveniently view the information and choose to check the dust bag's status or replace it based on the information provided.

[0131] If a user finds the dust bag is not full after checking, they can provide feedback to the processor via the cleaning application on their terminal device. The processor can then update at least one of the dust bag's status and filling time based on this feedback. If the user has replaced the dust bag for the base station, the processor can record the relevant time information, such as the time of the replacement, and recalculate at least one of the new dust bag's status and filling time.

[0132] Through the above example, by having the processor feed back at least one of the dust bag status and filling time to the cleaning device or a terminal device connected to the cleaning device, a more effective information interaction is established between the cleaning device and the user. This helps to improve the efficiency of the cleaning device and the user experience. It allows the user of the cleaning device to know the status of the dust bag used with the cleaning device in advance, avoiding the impact of the cleaning device's performance due to the dust bag being too full, reducing the frequency of the user manually checking the dust bag, and providing convenience for the user.

[0133] Furthermore, the cleaning device in this application example supports providing feedback on at least one of the dust bag status and filling time to multiple terminal devices, making it convenient for users to check the dust bag status and filling time on different terminal devices, thus meeting different user habits and needs.

[0134] In one possible implementation, the processor also:

[0135] Obtain the initial hair loss amount of at least one target object;

[0136] The feature information of at least one target object and the initial hair loss amount are input into the pre-trained hair loss prediction model to obtain the hair loss amount of at least one target object output by the hair loss prediction model. The hair loss prediction model is obtained by training the initial prediction model based on multiple training samples. Each training sample includes: the feature information of the sample object and the initial hair loss amount, as well as the hair loss amount of the sample object.

[0137] In one embodiment of this application, the processor can obtain the initial hair loss amount of at least one target object from the memory of the cleaning device or an external data source (the manufacturer's server or other cloud storage server). It should be understood that the initial standard value can be pre-configured based on environmental information of a preset area where the cleaning device is located, or the initial standard value can be set based on general conditions or empirical data. For different types of target objects, the corresponding initial standard value for hair loss is determined through previous research and data statistics. For example, for some common pet breeds, a pre-set approximate daily or weekly average hair loss amount is used as the initial standard value.

[0138] It should be noted that the initial standard values ​​can be stored in a database, indexed by the category of the target object (such as different pet breeds), making it easy for the processor to quickly find the initial hair loss amount corresponding to the target object. For example, for a typical short-haired cat, the initial hair loss amount is set to 5 grams of hair per day; for a medium-sized dog, the initial hair loss amount is set to 10 grams of hair per day.

[0139] In this application example, the initial hair loss amount of the target object can be adjusted according to different situations. For example, the initial standard value may vary slightly for target objects of different body types, genders or ages, and the processor can adaptively adjust the initial standard value based on the existing feature information.

[0140] In one example, during the model training phase of the cleaning device, developers collect a large number of training samples. Each training sample contains feature information of the sample object, the initial amount of hair loss, and the amount of hair loss of that sample object. These training samples can come from different real-world scenarios, including hair loss data of different pets in different environments, seasons, and physiological states.

[0141] Next, a suitable machine learning or deep learning model is selected as the initial prediction model. For example, it can be a linear regression model, a decision tree model, or a more complex neural network model (such as a multilayer perceptron (MLP), a long short-term memory network (LSTM), or a convolutional neural network (CNN), etc.). Alternatively, a traditional machine learning model, such as a decision tree, a random forest, or a support vector machine, can be selected, depending on the adaptability to the data and the performance requirements. This application does not specifically limit the choice.

[0142] During training, training samples are input into the initial prediction model. Using feature information and the initial hair loss amount as input, and the hair loss amount as output, the model's parameters are adjusted to make the predicted results as close as possible to the actual dynamic estimates in the samples. During training, optimization algorithms, such as stochastic gradient descent (SGD), can be used to continuously optimize the model, achieving good performance on both the training and validation sets. Furthermore, the trained hair loss prediction model can be stored in the cleaning device's processor or on a server accessible to the processor, allowing the processor to call upon the model for prediction when needed.

[0143] Before inputting the target object's feature information and initial hair loss amount into the hair loss prediction model, the data can be preprocessed. For example, feature information can be normalized to scale data from different ranges to a uniform range, improving the performance and stability of the hair loss prediction model. Then, the processor takes the preprocessed target object's feature information and initial hair loss amount as input to the pre-trained hair loss prediction model. Based on learned patterns and relationships, the model outputs the target object's hair loss amount. For example, if the input is a long-haired cat with an initial standard value of 10 grams of hair loss per day, the hair loss prediction model, based on the cat's feature information and considering its specific characteristics (such as large size and being in molting season), will ultimately output a dynamic estimate of 15 grams of hair loss per day.

[0144] In this embodiment, by combining the specific feature information of the target object with initial standard values ​​and utilizing a pre-trained hair loss prediction model, personalized hair loss prediction can be made for different target objects, rather than simply using a uniform standard, thus improving the accuracy of the prediction. Furthermore, the hair loss prediction model trained with a large number of training samples can better capture the complex nonlinear relationship between hair loss and various feature information, making the dynamic estimation results of hair loss for the target object more consistent with reality. This provides more accurate data support for subsequent operations of the cleaning equipment (such as determining the frequency of dust collection in the dust box and the filling time of the dust bag), thereby improving the performance of the cleaning equipment and the user experience.

[0145] In one possible implementation, the processor also:

[0146] Obtain molting season information for at least one target object;

[0147] Obtain the initial hair loss amount of at least one target object;

[0148] The feature information, molting season information, and initial hair loss amount of at least one target object are input into a pre-trained hair loss prediction model to obtain the hair loss amount of at least one target object output by the hair loss prediction model. The hair loss prediction model is obtained by training an initial prediction model based on multiple training samples. Each training sample includes: the feature information, molting season information, and initial hair loss amount of the sample object, as well as the hair loss amount of the sample object.

[0149] In one embodiment of this application, taking at least one target object as a pet as an example, the processor can obtain the molting season information of at least one target object through multiple means. One possible way is to obtain the current time and region information of the cleaning device's geographical location via a network, and then combine this with the built-in molting season information to determine whether it is currently the target object's molting season. For example, for pets, the time range of the molting season varies in different regions. The processor can determine whether it is currently in common molting seasons such as spring and autumn based on the geographical location information of the cleaning device (which can be obtained through a network or input by the user) and the current calendar time.

[0150] It should be noted that the cleaning device, combined with its built-in molting season information, can store it in the device's memory or on a cloud server for easy access by the processor. The processor stores the acquired molting season information of the target objects in its memory and encodes the molting season information for different target objects to retrieve the target object's molting season information and input it into the hair loss prediction model. For example, the molting season information can be represented as binary code, such as 00 indicating no molting season, 01 indicating spring molting, 10 indicating autumn molting, and 11 indicating both spring and autumn molting seasons.

[0151] In one example, during the model training phase of the cleaning device, developers collect a large number of training samples. Each training sample contains detailed information about the sample object, including its features, molting season information, initial hair loss amount, and the amount of hair loss obtained through actual observation or measurement. These training samples can be not only actual observation records of the target object by different users in different environments and seasons, but also simulated tests in a laboratory environment.

[0152] In this example, training samples are pre-input into the initial prediction model. Specifically, the feature information, molting season information, and initial hair loss amount from the training samples are used as input, and the hair loss amount is used as output. The initial prediction model is then trained using any optimization algorithm (such as stochastic gradient descent, Adagrad, Adadelta, or Adam). During training, the model's parameters are continuously adjusted to minimize the prediction error on both the training and validation sets, thereby improving the model's generalization ability and prediction accuracy. Furthermore, the trained hair loss prediction model can be stored in the cleaning device's processor or on a server accessible to the processor, allowing the processor to call upon the model for prediction when needed.

[0153] Before inputting the target object's feature information, molting season information, and initial hair loss amount into the hair loss prediction model, the above data can be preprocessed. For example, continuous variables in the feature information (such as hair length and body size) can be normalized to a range of [0,1]; categorical variables in the feature information (such as species) can be one-hot encoded to transform them into a form suitable for model input.

[0154] The processor inputs preprocessed data into a pre-trained hair loss prediction model, which outputs the amount of hair loss for the target object based on learned patterns and relationships. For example, for a pet, the input feature information indicates that it is a long-haired cat, currently in its autumn molting season, and the initial standard value is 10 grams of hair loss per day. The hair loss prediction model then outputs a higher dynamic estimate, such as 20 grams of hair loss per day, taking into account the impact of the molting season on the amount of hair loss.

[0155] In this embodiment, by using molting season information, target object feature information, and initial hair loss amount as input, various factors affecting hair loss can be considered more comprehensively, improving the accuracy and reliability of prediction. Furthermore, the hair loss prediction model trained with a large number of training samples can automatically learn the complex relationships between different factors, avoiding simple empirical judgments and exhibiting good adaptability to different target objects and environmental conditions. Through this implementation, the processor can more accurately predict the amount of hair loss from the target object, providing more precise data support for dustbin management and other related operations of the cleaning equipment, thereby optimizing the performance of the cleaning equipment and the user experience.

[0156] In one possible implementation, the sensor data includes image data and radar data, such as Figure 6 As shown, the sensor assembly includes:

[0157] Image acquisition component 1011 is used to acquire image data of at least one target object;

[0158] The radar component 1012 is connected to the image acquisition component 1011. When the image acquisition component acquires an image of at least one target object, the radar component acquires radar data of at least one target object.

[0159] In one embodiment of this application, the image acquisition component can be various types of cameras, such as RGB cameras, which can capture visual information such as the appearance, color, and shape of the target object. For example, in a scenario of cleaning pet hair, the RGB camera equipped with the cleaning device can be used to capture photos or video streams of the pet, providing image data for the processor to identify features such as the pet's species, size, and hair length.

[0160] In one embodiment of this application, the radar component is connected to the image acquisition component. When the image acquisition component acquires an image of at least one target object, radar data of the target object is acquired simultaneously. The radar data includes various information, such as the pet's height and size.

[0161] Specifically, the image acquisition component and radar component can be installed at specific locations on the cleaning device, such as on the top or front of a robotic vacuum cleaner, to ensure that sensor data of the target object can be clearly acquired during the cleaning process. Furthermore, in this example, the image acquisition component acquires images of the target object at a predetermined frequency (e.g., several frames per second) to ensure the continuity and timeliness of the acquired image data.

[0162] As an optional embodiment, when the cleaning device starts working and enters an area where a target object may be present, the image acquisition component activates and continuously collects image data of the surrounding environment, constantly scanning the scene within its field of view to search for the appearance of the target object. Furthermore, the image data collected by the image acquisition component is transmitted in real time to the processor (or the image analysis module within the device). The processor uses image recognition algorithms, such as deep learning-based target detection algorithms (e.g., YOLO, Faster R-CNN, etc.), to detect the presence of a target object in the image data. If a target object is detected, the data acquisition operation of the radar component is triggered. After receiving the trigger signal, the radar component simultaneously acquires radar data of the target object. This radar data is closely correlated with the just-acquired image data in time, together constituting the sensor data.

[0163] By collecting image and radar data of the target object, multi-dimensional basic information is provided for the feature analysis of the target object. Not only can feature information of the target object be obtained from image data, but it can also be supplemented more comprehensively by combining radar data, thereby improving the accuracy of predicting the amount of hair loss of the target object.

[0164] One possible implementation is still as follows Figure 6 As shown, processor 120 also includes:

[0165] Process image data of at least one target object to generate at least one of the following: number, size, type and hair length of the target object;

[0166] Process radar data of at least one target object to generate at least one of the target object's height, volume, and shape.

[0167] In one example, for data information, after the image acquisition component obtains the image data of the target object and transmits it to the processor, the processor uses a deep learning-based target detection algorithm (which learns the feature patterns of the target object by pre-training on a large dataset of images containing the target object (such as a pet)), such as the YOLO (You Only Look Once) series of algorithms or the Faster R-CNN algorithm, to search for the contours, shapes and other features of the target object in the image data and determine the number of target objects in the image data.

[0168] Taking pets as an example, the processor divides the image data into multiple grids and performs object detection on each grid. If an object matching the characteristics of a pet is detected in a grid, it is counted once. By traversing the entire image data, the total number of pets in the image data is finally determined.

[0169] In another example, for body size information, the processor can estimate the actual size of the target object by identifying its pixel proportion in the image and known image acquisition component parameters (such as focal length and shooting distance). For example, if a dog occupies a large pixel area in an image, and the shooting distance is known to remain constant, it can be preliminarily determined that the dog is large. Combining pre-established libraries of different body size templates (such as standard outline templates for small, medium, and large dogs), template matching algorithms or deep learning models can be used for more accurate body size determination. Deep learning models can learn the feature differences between target objects of different body sizes, such as the wider body outline and thicker limbs of large dogs, thereby accurately identifying the body size category of the target object.

[0170] In another example, for species information, deep learning models such as Convolutional Neural Networks (CNNs) are used to extract features from image data. CNNs, through multiple convolutional and pooling layers, can automatically learn the local and global features of the target object. For example, for pet species recognition, the model can learn the unique facial features of different pet species (such as the flat face of a Persian cat, the large ears of a Chihuahua), coat color distribution (such as the black and white coat of a Husky), etc. The extracted unique facial features are input into a classifier and compared with a pre-trained species classification model. The classification model stores feature information of various common pet species. By calculating the similarity between the input features and the features of various classes, the species to which the target object belongs is determined.

[0171] In another example, for hair length information, the processor first uses an edge detection algorithm, such as the Canny algorithm, to detect the hair edges of the target object and obtain the approximate length range of the hair. Then, it uses a texture analysis algorithm, such as the Gray-Level Co-occurrence Matrix (GLCM), to analyze the texture features of the hair and determine the density and direction of the texture. For example, long hair usually has more obvious texture and longer edges, while short hair has relatively less obvious texture and shorter edges. Combined with a deep learning model, by learning from a large number of images of target objects with different hair lengths, it can learn the subtle visual differences between long and short hair, thus more accurately determining the hair length of the target object. For example, for long-haired cats and short-haired cats, the deep learning model can accurately determine the corresponding hair length through the visual features of the hair.

[0172] Taking a lidar system as an example, a laser beam is emitted and the time it takes for the beam to travel from emission to reflection is measured. The distance between the target object and the sensor is calculated based on the speed of light. When the radar component acquires radar data of the target object, the lidar measures the distance to the target object from different angles. Then, the processor, based on this distance information, combined with parameters such as the target object's position in the image and the sensor's installation angle, calculates the height of the target object using the principle of triangulation. For example, if the lidar measures the distances to the top and bottom of the target object from different angles, and combines this with the sensor's position and angle information, the height of the target object can be accurately calculated.

[0173] The processor combines distance information from multiple directions acquired from distance sensors with image data of the target object. The image data provides the approximate outline of the target object, while the radar data provides its actual dimensions in three-dimensional space. Using this multi-dimensional data, the volume of the target object is calculated through a volume estimation algorithm (such as a geometry-based approximation method, which approximates the target object as a certain geometric shape, such as a cuboid or cylinder, and calculates the volume based on the measured dimensions). For example, for a pet with an approximate cuboid shape, its volume is calculated by measuring its length, width, and height (obtained from distance sensor data and image data).

[0174] By performing feature recognition processing based on image data and radar data respectively, the processor can acquire rich and comprehensive feature information about the target object. Different types of data reflect the feature information of the target object from different perspectives. Recognition processing based on multiple data sources can reduce the errors and uncertainties caused by single data. These feature information complement and combine with each other, providing a data foundation for subsequent in-depth analysis of the target object, such as hair loss prediction and cleaning strategy formulation.

[0175] One possible implementation is still as follows Figure 6 As shown, the cleaning device also includes a controller 104, which is located inside the main body 100 of the device and connected to the processor 102. The controller 104 responds to the control command received by the cleaning device and controls the cleaning device to start the dust collection mode.

[0176] One possible implementation is still as follows Figure 6 As shown, the radar component 1012 also collects radar data of at least one target object when the image acquisition component collects an image of at least one target object during the operation of the cleaning device after the dust collection mode is turned on.

[0177] In one embodiment of the example of the present application, the controller of the cleaning device is connected to the processor, and receives a control instruction issued by the user to the cleaning device. It should be understood that the control instruction can be issued in various ways, for example, through a mobile phone application program matched with the cleaning device, a physical key on the cleaning device itself, or voice control. As Figure 7 shown, after the user clicks the "Smart" control in the "dust collection mode setting interface" of the cleaning application program on the mobile phone and then clicks the "OK" control, the control instruction for the cleaning device is triggered, so as to control the cleaning device to enable the dust collection mode.

[0178] After the cleaning device receives the user's control instruction requesting to enable the dust collection mode, the controller responds immediately and sends corresponding signals to each component of the cleaning device to start the dust collection mode. At this time, all components of the cleaning device, such as the motion system, the image acquisition component, and the radar component, enter corresponding working states to prepare for intelligent obstacle avoidance and target object data acquisition.

[0179] In the dust collection mode, during the operation of the cleaning device, the cleaning device pays more attention to the identification and avoidance of obstacles in the surrounding environment, so as to ensure that the cleaning task is completed safely and efficiently. After the cleaning device enables the dust collection mode and starts to operate, the radar component is in a standby state at any time. Once the image acquisition component captures the image data of at least one target object, it is used as a trigger signal for the radar component to collect radar data, ensuring that the collection of radar data is closely associated with the image acquisition of the target object, so that the acquired sensor data is more targeted and effective.

[0180] The collected image data and radar data are transmitted to the processor, and the processor analyzes the features of the target object based on the image data and the radar data. For example, by combining the appearance features of the target object in the image with the distance and speed information in the radar data, at least one of the type, hair length, body shape and height of the target object can be identified more accurately, so as to optimize the obstacle avoidance strategy and cleaning path planning of the cleaning device. It is ensured that in the intelligent obstacle avoidance mode, combined with the image data and the radar data, the cleaning device can perceive the target objects in the surrounding environment more comprehensively and accurately, which helps to improve the identification accuracy and behavior prediction ability for the target objects.

[0181] In addition, in the example of the present application, the user can flexibly determine whether to enable the dust collection mode of the cleaning device according to the actual scene requirements, so as to ensure that the cleaning device can work normally even in a complex environment (such as a home with pets moving around), reduce equipment damage caused by collision and disturbance to the target object, and provide the user with a more intelligent and convenient use experience.

[0182] In a possible implementation, as Figure 8 shown, the controller 104 further controls the device main body to perform the dust collection operation according to the dust collection frequency of the dust box.

[0183] The processor determines the dust collection frequency based on multiple factors, including the amount of hair shed from the target object, the cleaning frequency, and the cleaning mode, and transmits this calculated frequency to the connected controller. Upon receiving the dust collection frequency, the controller initiates an internal timing or counting mechanism. If the frequency is set based on time, such as returning the device to its dock every two days, the controller starts timing; when the timer reaches the predetermined point, it triggers the dock return command for dust collection. If the frequency is set based on a count, such as returning the device to its dock after cleaning all rooms, the controller issues the dock return command after the main unit has completed cleaning all rooms.

[0184] Upon triggering the return-to-charging-pile command, the controller sends control signals to all relevant components of the device. First, it controls the device's navigation system to plan a path back to the charging pile (i.e., the base station). The navigation system calculates the optimal return path based on the device's current location, the surrounding map, and the charging pile's location. Simultaneously, the controller pauses the ongoing cleaning process, such as turning off the sweeping brushes and stopping the vacuum motor, ensuring that cleaning does not resume during the return process to avoid stirring up dust or causing other unnecessary problems. Then, the device moves along the planned path to the charging pile. Upon arrival, the controller connects the dustbin to the charging pile's dust bag, transferring the collected dust, hair, and other debris to the charging pile's dust bag or other storage container, completing the dust collection process. Afterward, if any uncleaned areas remain, the device restarts the cleaning process according to a preset cleaning program to finish cleaning those areas.

[0185] Through the above implementation, the equipment can automatically and intelligently adjust the frequency and operation of dust collection based on actual waste collection conditions, without requiring manual intervention from the user. This greatly improves the autonomy and ease of use of the cleaning equipment, eliminating the need for users to constantly monitor whether the dustbin is full, thus reducing their operational burden. Furthermore, a reasonable dustbin collection frequency ensures that the dustbin is not overfilled, which would affect cleaning performance, and also avoids wasted cleaning time due to excessively frequent dust collection. For example, when hair loss is low, appropriately reducing the dustbin collection frequency improves the overall cleaning efficiency of the cleaning equipment. By controlling the dustbin collection frequency and preventing overfilling, the equipment's suction system and dustbin structure are protected, extending its service life.

[0186] This application embodiment also provides a terminal device that is communicatively connected to the cleaning equipment, still as... Figure 5 As shown, the terminal device includes: a processor 201 and a display screen 202, wherein:

[0187] The processor 201 runs a cleaning application and receives terminal collection information associated with at least one target object uploaded by the user through the cleaning application, and sends the terminal collection information to the cleaning device, wherein the cleaning device is paired and bound to the cleaning application.

[0188] The display screen 202 is connected to the processor 201 and displays at least one of the status of the dust bag and the filling time, as reported by the cleaning device.

[0189] In one embodiment of this application, the terminal device can be a user device such as a smartphone, tablet, or smartwatch. The terminal device runs a cleaning application (APP) (e.g., a cleaning device management application), which is associated with, paired with, or bound to the cleaning device used by the user.

[0190] In this application example, the terminal device receives at least one of the dust bag status and filling time from the processor of the cleaning device via a communication connection with the cleaning device.

[0191] In one embodiment of this application, the terminal device can provide feedback on at least one of the dust bag status and filling time in various forms, such as text messages, voice messages, pop-up notifications, and persistent reminder windows. This ensures that the user notices the notification message promptly. The dust bag status and filling time can be displayed on the user interface of the cleaning application on the terminal device, or on the main interface of the terminal device. This guides the user to check or replace the dust bag at appropriate times, ensuring the normal operation of the cleaning equipment and preventing the dustbin from becoming too full and affecting the cleaning effect.

[0192] For example, the terminal device can present the received filling time to the user in an intuitive way through the display screen, such as displaying "[X] hours left until the dust bag is full" in the form of a countdown, or displaying the specific filling time in the form of a calendar date and time, such as "The dust bag is estimated to have 10% of its filling space remaining." In this way, the user can clearly understand the usage status of the dust bag and make corresponding preparations in advance.

[0193] For example, the terminal device can use its built-in voice broadcast module to clearly and accurately inform the user, "Dear user, the current dust bag fill level is 60%, which is in a medium-fill state. Please pay attention to subsequent usage." Simultaneously, when the user opens the cleaning application app, the terminal device's display screen can show the dust bag fill status in the form of an intuitive progress bar. The progress indicator line precisely points to the 60% position on the progress bar, and the words "Medium-fill" are prominently marked next to the progress bar for easy user monitoring.

[0194] Furthermore, cleaning apps can also use charts to display predicted trends in dust bag filling time. For example, a line graph could show the changes in dust bag filling time over the past week, helping users understand fluctuations in the filling speed. The chart could also indicate the current estimated filling time and provide comparative data for different time periods, allowing users to better understand the cleaning equipment's operating status and hair shedding.

[0195] The processor of the terminal device is responsible for running the cleaning application bound to the cleaning device. This cleaning application provides a way for users to interact with the cleaning device, and has functions such as viewing device status and remotely controlling the operation of the cleaning device. After the user downloads and installs the cleaning application on the terminal device (such as a mobile phone or tablet), they bind the cleaning application to the specific cleaning device and establish a communication connection through registration, login, and other processes.

[0196] The cleaning application interface allows users to input information associated with at least one target object, resulting in data collected by the terminal. For example, in a pet hair cleaning scenario, users can input information such as the pet's breed, age, gender, and daily activity area, enabling the cleaning device to more accurately analyze the target object's characteristics and behavior, thereby optimizing the cleaning strategy. For instance, users can... Figure 4 Clicking the "Member Information" or "Pet Information" control in the family management interface will redirect you to the corresponding information management interface, where you can enter information.

[0197] It should be noted that users can add or enter the aforementioned terminal-collected information through the cleaning application running on the terminal device when using the cleaning device for the first time or when configuring the cleaning device; alternatively, users can update the configuration information of the cleaning device in real time through the cleaning application running on the terminal device and re-add or enter the aforementioned terminal-collected information during the use of the cleaning device.

[0198] In one example, the user's terminal device responds to the user clicking the "Member Information" control in the home management interface of the cleaning application, and then redirects from the home management interface to a page such as... Figure 4 The member information management interface shown allows users to add at least one family member's information, such as age (e.g., 25 years old, 30 years old), gender (male, female), special group (elderly, pregnant women, children, etc.), and hair length (roughly classified based on a predetermined human hair threshold: long hair, short hair, medium-long hair), through manual input, voice input, or by selecting the adapter box.

[0199] In another example, the user's terminal device (e.g., a smartphone) responds to the user clicking the "Pet Information" control in the home management interface of the cleaning application, and then redirects from the home management interface to a page such as... Figure 4 The pet information management interface shown allows users to add pet information such as the number of pets they own, the type or breed of each pet (cat, dog, rabbit, etc.), fur length (roughly classified based on a predetermined pet fur threshold: long-haired, short-haired), age (0.5 years), and size (small, medium, large) through manual input, voice input, or selecting the adapter box.

[0200] The terminal device's processor monitors user input in the cleaning application in real time. Once the user completes and submits the information, the processor immediately receives the information collected by the terminal. The processor then organizes and encodes this information to conform to the communication protocol format with the cleaning device. Finally, it sends the collected information to the cleaning device via a communication link (e.g., Wi-Fi, Bluetooth, or other wireless communication methods). Upon receiving the terminal information, the cleaning device's processor combines it with feature information acquired from locally collected sensor data. This further improves the accuracy of target object identification and the precision of hair loss prediction, thereby providing users with a more personalized and efficient cleaning service.

[0201] The terminal device provided in this application example intuitively displays at least one of the dust bag's status and filling time on its screen. Users can easily monitor the dust bag's status without physically checking it in real time, significantly improving the convenience and intelligence of the user experience. Furthermore, users can upload detailed information about the target object through a cleaning application running on the terminal device. The terminal device then transmits this information to the cleaning equipment, allowing it to adjust its cleaning strategy precisely according to each user's specific needs, meeting diverse individual requirements. In addition, the cleaning application acts as a bridge between the user and the cleaning equipment, enabling two-way information exchange, enriching the user's control methods and functional experience, enhancing interactivity, and increasing the user's sense of control over the cleaning equipment.

[0202] In addition, an alternative example is the display screen, which can also show the dustbin filling time of the cleaning device. Users can understand the dustbin's near-fullness without directly opening the cleaning device to check it, effectively avoiding decreased cleaning efficiency or abnormal operation due to a full dustbin. Furthermore, based on the dustbin's dust collection frequency and hair shedding volume determined by the cleaning device, when the dustbin is predicted to be near full, the cleaning device can send at least one of the dustbin's status and filling time to the user's terminal device (such as a cleaning application app) through a communication connection, informing the user of the dustbin's status, filling time, and other information. This further reduces the burden on users to frequently check the dustbin's status themselves.

[0203] Furthermore, the terminal device's processor can record and analyze cleaning data, such as cleaning area, cleaning time, and the amount of pet hair collected in different areas, and present this data to the user in the form of visual charts through the cleaning application. For example, a bar chart can be used to show the amount of pet hair collected in different rooms, helping users understand the distribution of pet hair in their homes. Moreover, it can provide personalized cleaning suggestions based on the characteristics of the user's pets and cleaning habits, such as recommending more suitable cleaning modes and dustbin emptying cycles, meeting the personalized needs of different users and improving the user experience.

[0204] It should also be noted that in the actual use of cleaning equipment, the amount of debris generated in the dustbin is affected by various factors, such as fluctuations in the size of the cleaning area, the type of debris, and the amount of hair. This makes the predicted filling time and maintenance cycle inaccurate and potentially inconsistent with the actual situation. Therefore, when the cleaning equipment sends dust bag alerts to the terminal device according to preset logic (such as feedback on the dust bag's status based on hair loss and at least one of the filling time), to compensate for the uncertainty in the cleaning equipment's predictions, such as... Figure 9 The user feedback interface of the robot vacuum shown can also provide users with feedback options through the user feedback interface of the cleaning application, such as "Full, I have replaced it" and "Not full, do not remind me for now". In this way, users can choose to provide feedback on whether the dust bag is full based on what they actually see or check, so that the terminal device and the cleaning device can understand more accurate dust bag status information.

[0205] In one example, if the user selects the "Full, I've Replaced" option in the feedback settings, it means the dust bag has been replaced, and the cleaning device restarts its calculation of the next reminder time, allowing it to recalculate the reminder time based on the usage of the new dust bag. In another example, if the user selects the "Not Full, Don't Remind" option in the feedback settings, a pop-up message may appear... Figure 10Another user feedback interface for a robotic vacuum cleaner is shown. In this interface, users can provide more specific options regarding the remaining space in the dust bag, such as "nearly half," "a small portion," or "soon to be notified." The cleaning device then optimizes the notification frequency based on this feedback, adjusting previously predicted notification and filling times. By analyzing the discrepancies between actual conditions and predictions, it determines to extend the next notification time and adjust the currently calculated filling time accordingly. This ensures that subsequent feedback from the cleaning device regarding the dust bag status and / or filling time more accurately reflects actual usage, avoiding unnecessary notifications and improving the user experience.

[0206] This application provides an example of a dust collection and treatment method for a cleaning device. Please refer to... Figure 11 As shown, Figure 11 A schematic flowchart illustrating a dust collection and treatment method for a cleaning device provided in this application is shown. This is an example and not a limitation; the method can be applied to or operated in a cleaning device. The method includes:

[0207] S1101, acquire sensor data of at least one target object within the preset area where the main body of the device is located;

[0208] S1102, Process sensor data of at least one target object to generate feature information of at least one target object;

[0209] S1103, determining the amount of hair loss of at least one target object based on the feature information of at least one target object; and

[0210] S1104 provides feedback on at least one of the following: the status of the dust bag and the filling time, based on the current amount of hair loss.

[0211] In this application example, the sensor data of the target object includes image data and radar data. For example, the cleaning device can be equipped with an image acquisition component such as an RGB camera to acquire image data of at least one target object (such as a human, pet, etc.) in the room where the cleaning device is located. The distribution of different pixel values ​​in the image data reflects the appearance, color, outline, and other features of at least one target object. As another example, a radar component such as a lidar can also be set on the cleaning device to acquire radar data of at least one target object when the image acquisition component acquires an image of at least one target object, thereby obtaining data such as the height, volume, and shape of each target object.

[0212] In one optional application scenario, during operation, the cleaning device's sensor components scan a preset area (using 3D scanning technology) where the device itself is located. In one embodiment, the preset area can be the entire room where the cleaning device is located, a user-specified cleaning area, or a cleaning area planned centered on the device itself. For example, when the device is cleaning the living room, it collects data on target objects within the living room area.

[0213] In one example, while the cleaning device moves along a planned path within an indoor space and performs cleaning operations, it simultaneously controls sensor components to collect data at a predetermined frequency (e.g., once every 5 seconds) and angular range (180°, 360°, etc.) to ensure that sensor data of at least one target object in the indoor space can be collected. For example, taking a sensor component that includes an image acquisition component and a radar component as an example, when the device moves within a user's bedroom to perform cleaning operations, the image acquisition component (e.g., an RGB camera) captures images of people and / or pets in the bedroom, obtaining image data of the people and / or pets; while the image acquisition component collects image data of the people and / or pets, the radar component (e.g., a lidar) continuously scans for radar data of the people and / or pets, thereby collecting sensor data of the people and / or pets in the room.

[0214] In this application example, the sensor component transmits the sensor data (including image data, radar data, etc.) of at least one target object to the processor of the cleaning device in real time. The processor of the cleaning device inputs the sensor data of at least one target object into the pre-trained feature recognition model and receives the feature recognition model outputting feature information corresponding to the sensor data, such as quantity, size (small, medium, large), age (0.5 years, 5 years), species (human or pet), and hair length (roughly classified based on a predetermined pet hair threshold: long hair, short hair).

[0215] It should be noted that the processor can pre-use deep learning algorithms in the field of computer vision, and use multiple training samples containing sensor data of the target object and corresponding feature information (using labels to annotate the feature information of the target object, such as type, quantity, size, height, volume, hair length, etc.) to train the initial neural network model, thereby constructing a feature recognition model that can accurately identify the feature information of the target object.

[0216] Furthermore, the cleaning equipment can pre-train an initial prediction model based on multiple training samples to obtain a hair loss prediction model. Each training sample includes: the feature information of the sample object and the amount of hair loss of the sample object. Specifically, the hair loss prediction model calculates the amount of hair loss of the target object based on the input feature information of the target object and a prediction mechanism obtained by learning the feature information and hair loss amount of the sample object using a machine learning algorithm. For example, for a long-haired cat, the hair loss prediction model calculates the amount of hair loss of the long-haired cat based on its body size, hair length, and other feature information, combined with the feature information and hair loss amount of similar long-haired cats in the training samples.

[0217] The cleaning device provided in this application example can accurately identify multiple pets or humans in a household simultaneously. Taking pets as an example, it performs detailed hair loss calculations based on the unique characteristics of each pet, such as long-haired cats having long, easily tangled fur and short-haired dogs having short fur but shedding a lot during molting season. Based on this, it determines and reports at least one of the following: the status of the dust bag and the filling time. This improves the applicability and intelligence of the cleaning device in pet-owning households, meeting the complex and diverse cleaning needs of multi-pet families.

[0218] Moreover, as time goes by and the environment changes, the cleaning equipment can update and adjust the amount of hair loss from the target object in real time based on the newly collected characteristic information of the target object and environmental information, so as to ensure the accuracy and dynamic adaptability of the cleaning equipment.

[0219] After determining the amount of hair shed, the processor provides feedback on at least one of the dust bag's status and filling time based on the hair shed amount. This allows users advance preparation time to replace the dust bag in a timely manner, ensuring the normal operation of the cleaning equipment. For example, the processor can combine the dust bag's capacity information with feedback on at least one of the dust bag's status and filling time based on the hair shed amount.

[0220] In one example, the processor of the cleaning device provides feedback on the dust bag filling time based on the amount of hair shed: assuming the dust bag capacity is a known fixed value, the time required for the dust bag to fill from the current time point is estimated by calculating the amount of hair shed by the main body of the cleaning device per unit time. For example, if the total capacity of the dust bag (maximum hair carrying capacity) is 500 grams, the currently calculated hair shed per unit time is 10 grams / hour, and the dust bag currently contains 100 grams of hair, then the feedback indicates that the dust bag will need another 40 hours to fill.

[0221] In one optional example, the cleaning device's built-in voice module can be used to issue voice prompts, such as "Hello, the dust bag is expected to fill within 24 hours, please pay attention." Furthermore, the frequency of the prompts can be gradually increased as the filling time approaches, such as broadcasting again at 12 hours, 6 hours, and 1 hour remaining, and the content of the broadcasts can be more urgency-based, such as "Please note that the dust bag is expected to fill within 1 hour, please handle it as soon as possible."

[0222] In addition to voice prompts, another optional example is to set different sound effects to indicate different dust bag filling states. For example, when the dust bag is expected to fill in 24-48 hours, the cleaning device emits a gentle prompt sound, such as a "ding-dong" sound; when there are 24-12 hours remaining, the prompt sound changes to a slightly faster "beep-beep-beep" sound; if there are less than 12 hours remaining, the prompt sound becomes a rapid, continuous beeping sound to attract the user's attention.

[0223] In another example, the processor of the cleaning equipment provides feedback on the dust bag's status based on the amount of hair shed: assuming the dust bag's capacity is a known fixed value, a pre-determined percentage of the dust bag's load capacity is less than 30% for a lightly filled state, greater than 30% but less than 70% for a moderately filled state, and greater than 70% for a heavily filled state. For example, if the dust bag's maximum hair capacity is 500 grams, and the cleaning equipment calculates the amount of hair shed per unit time, currently calculated to be 10 grams per hour, and the dust bag currently contains 300 grams of hair, the current load capacity is calculated to be 60% (i.e., 300 grams divided by 500 grams, then multiplied by 100%), and the cleaning equipment then determines that the dust bag is currently in a moderately filled state.

[0224] Subsequently, the cleaning device, through its built-in voice broadcast module, clearly and accurately informs the user, "Dear user, the current dust bag filling level is 60%, which is in a medium filling state. Please pay attention to subsequent usage." Simultaneously, the dust bag filling status can be displayed on the cleaning device's LED screen in the form of an intuitive progress bar. The progress indicator line precisely points to the 60% position on the progress bar, and the words "Medium Fill" are prominently marked next to the progress bar for easy user monitoring.

[0225] In another example, the cleaning device's processor provides feedback on the dust bag's status and filling time based on the amount of hair shed: assuming the dust bag capacity is a known fixed value, a pre-defined state is defined as a lightly filled dust bag if the occupied percentage is less than 30%, a moderately filled dust bag if the occupied percentage is greater than 30% but less than 70%, and a heavily filled dust bag if the occupied percentage is greater than 70%. For example, in a pet-owning household, the cleaning device detects that the pet is in molting season, shedding 20 grams of hair per hour (ignoring other debris), the total dust bag capacity is 600 grams, and the dust bag has currently collected 100 grams of hair. The calculated filling percentage is approximately 16.7%, determining the dust bag to be in a lightly filled state.

[0226] Next, the cleaning equipment calculates the dust bag filling time. First, it calculates the remaining capacity of the dust bag as 600 grams minus 100 grams, which is 500 grams. Then, based on 500 grams divided by the current hair shedding rate of 20 grams per hour, it calculates that the dust bag filling time is 25 hours.

[0227] To provide users with comprehensive and accurate information about the dust bag, the cleaning device can seamlessly connect with a smart speaker to inform users in a clear and natural voice broadcast: "Hello user, the dust bag is currently lightly filled, with a filling percentage of approximately 16.7%. The dust bag is expected to be full in 25 hours." Simultaneously, the device management interface on the smart home control screen displays the dust bag filling progress bar in an intuitive visual chart format. The progress indicator line precisely points to the 16.7% position on the progress bar, and the text "Lightly Filled" and "Expected to be Full in 25 Hours" are prominently displayed next to the progress bar, allowing users to easily obtain detailed information about the dust bag's status and filling time on the control screen.

[0228] Another alternative is to install LED indicator lights on the cleaning device itself. For example, different colors and flashing frequencies could be used to indicate the status of the dust bag and when it will be full. A solid green light indicates that the dust bag is not full and there is still a long time before it is full; a flashing yellow light indicates that the dust bag is expected to be full within 24-48 hours; and a rapidly flashing red light indicates that the dust bag is expected to be full within 24 hours.

[0229] Another alternative is that if the cleaning device is equipped with a display screen, the screen can show a graphical interface indicating the dust bag filling progress bar, with the estimated filling time indicated next to the bar. For example, the progress bar might show 80% completion, with "estimated to fill in 12 hours" displayed next to it. Simultaneously, the display screen can also show historical dust bag filling data, hair collection volume in different areas, and other relevant information, allowing users to easily understand the cleaning progress.

[0230] As an optional implementation, in addition to the cleaning device feeding back at least one of the state of the dust bag and the filling time, the feedback can also be performed through a terminal device paired and bound with the cleaning device. For example, a user communicates and connects with the cleaning device through a cleaning application APP running on a terminal device such as a smartphone. When at least one of the state of the dust bag and the filling time changes, a push prompt message can be sent to the smartphone through the cleaning application APP, and the smartphone displays at least one of the state of the dust bag and the filling time on its display screen. It should be understood that in addition to at least one of the state of the dust bag and the filling time, the prompt message may also include some suggestions for the user. For example, the prompt message may be: "The dust bag is currently in a lightly filled state, and it is expected to be full in 36 hours. It is recommended that you prepare a new dust bag in advance for timely replacement". The user can click the notification to enter the detailed page of the APP to view information such as the state of the dust bag, the dynamic change of the filling time, the cleaning history record, and the device state.

[0231] There is another optional implementation. The cleaning device may also adopt a feedback mode that is linked with a smart home system (a smart speaker, a smart home central control screen, etc.). For example, if a smart speaker is provided in the user's home, the cleaning device can be paired and bound with the smart speaker to realize linkage. When it is predicted that at least one of the state of the dust bag and the filling time meets a set condition, the smart speaker automatically broadcasts relevant information. For example, the user sets that a notification is to be sent when the dust bag is expected to be full within 12 hours, and the smart speaker can give a voice prompt "The dust bag is expected to be full within 12 hours, please deal with it in time". The user can also ask the smart speaker for detailed information about the dust bag through a voice instruction, such as "Little X, what is the current state of the dust bag of my sweeping robot", and the smart speaker obtains real-time data according to the connection with the cleaning device and answers the user: "The dust bag is currently in a lightly filled state".

[0232] It should be noted that the present application is not only applicable to various pet-raising families, and can perform accurate prediction according to the hair shedding situation of different pets, and can timely remind the user of at least one of the state of the dust bag and the filling time during the pet molting season, so as to ensure the cleaning effect. It is also applicable to families or places with high cleaning requirements, for example, there are people susceptible to allergies in the family, by accurately feeding back at least one of the state of the dust bag and the filling time, the dust bag is cleaned in time to reduce secondary pollution.

[0233] The dust collection control method of the cleaning device provided in this application uses a sensor component installed in the main body of the device to collect sensor data in real time from target objects (e.g., indoor objects that shed hair, such as people and pets). A processor, located within the main body of the device and connected to the sensor component, processes the collected sensor data to generate characteristic information of the target object and determines the amount of hair shed. Based on the amount of hair shed, the processor provides feedback on at least one of the dust bag's status and filling time. By establishing a correlation between hair shed and dust bag cleaning frequency, the cleaning device allows users to rationally schedule dust bag cleaning based on the dust bag's status and / or filling time, thereby avoiding the impact of overfilled dust bags on the cleaning device's performance and reducing the frequency of manual dust bag checks.

[0234] One possible implementation method also includes:

[0235] Acquire terminal acquisition information pre-stored in the memory and associated with at least one target object;

[0236] Process terminal-collected information associated with at least one target object to generate feature information of at least one target object.

[0237] In one possible implementation, the cleaning device also includes a dust box connected to the main body of the device, and the method further includes: feeding back the frequency of dust collection in the dust box based on the amount of hair shed.

[0238] In one possible implementation, the frequency of dust collection in the dust box is fed back based on the amount of hair shed, including: obtaining the cleaning frequency of the main body of the device and the target cleaning mode; and when the main body of the device is operating in the target cleaning mode, determining the frequency of dust collection in the dust box based on the amount of hair shed and the cleaning frequency, wherein the target cleaning mode includes at least one of a sweeping mode, a sweeping and mopping mode, and a sweeping and mopping mode.

[0239] In one possible implementation, feedback is provided on at least one of the status of the dust bag and the filling time based on the amount of hair shed, including: feedback on at least one of the cleaning area of ​​the main body of the device, the total capacity of the dust bag, and the amount of hair shed.

[0240] One possible implementation method also includes:

[0241] Based on the dust bag's filling time, a notification message is output to the terminal device paired with the cleaning equipment. The terminal device runs a cleaning application, and the cleaning equipment is paired with the cleaning application. The notification message indicates the status of the dust bag before the filling time is reached.

[0242] In one possible implementation, determining the amount of hair loss of at least one target object based on the feature information of at least one target object includes:

[0243] Obtain the initial hair loss amount of at least one target object;

[0244] The feature information of at least one target object and the initial hair loss amount are input into the pre-trained hair loss prediction model to obtain the hair loss amount of at least one target object output by the hair loss prediction model. The hair loss prediction model is obtained by training the initial prediction model based on multiple training samples. Each training sample includes: the feature information of the sample object and the initial hair loss amount, as well as the hair loss amount of the sample object.

[0245] In one possible implementation, determining the amount of hair loss of at least one target object based on the feature information of at least one target object further includes:

[0246] Obtain molting season information for at least one target object;

[0247] Obtain the initial hair loss amount of at least one target object;

[0248] The feature information, molting season information, and initial hair loss amount of at least one target object are input into a pre-trained hair loss prediction model to obtain the hair loss amount of at least one target object output by the hair loss prediction model. The hair loss prediction model is obtained by training an initial prediction model based on multiple training samples. Each training sample includes: the feature information, molting season information, and initial hair loss amount of the sample object, as well as the hair loss amount of the sample object.

[0249] In one possible implementation, acquiring sensor data of at least one target object within a preset area where the main body of the cleaning device is located further includes:

[0250] An image acquisition component that acquires image data of at least one target object;

[0251] The radar component is connected to the image acquisition component. When the image acquisition component acquires an image of at least one target object, the radar component acquires radar data of at least one target object.

[0252] In one possible implementation, processing sensor data of at least one target object to generate feature information of at least one target object includes:

[0253] Processing image data of at least one target object to generate at least one of the following: number, size, type, and hair length of the target object; and

[0254] Process radar data of at least one target object to generate at least one of the target object's height, volume, and shape.

[0255] The dust collection and treatment methods of the cleaning equipment provided in the above embodiments of this application can all be applied to or operated on the cleaning equipment provided above. Therefore, the beneficial effects that can be achieved can be referred to the beneficial effects of the cleaning equipment provided above, and will not be repeated here.

[0256] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0257] This application embodiment also provides a cleaning device, which includes one or more processors and a memory;

[0258] The memory is coupled to one or more processors. The memory is used to store computer program code, which includes computer instructions. One or more processors call the computer instructions to cause the cleaning device to perform the dust collection method of the cleaning device described above.

[0259] Figure 12 This is a schematic diagram of a cleaning device provided in an embodiment of this application. The cleaning device 1200 can be a mobile phone, smart screen, tablet computer, wearable cleaning device, vehicle-mounted cleaning device, augmented reality (AR) device, virtual reality (VR) device, laptop computer, ultra-mobile personal computer (UMPC), netbook, personal digital assistant (PDA), projector, or a communication device such as a server, storage device, or base station, or a smart car, etc. This application embodiment does not limit the specific type of cleaning device.

[0260] The memory 1201 can be used to store computer software programs 1202 and modules. The processor 1203 executes various functional applications and data processing of the cleaning device by running the software programs and modules stored in the memory 1201. The memory 1201 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created based on the use of the cleaning device (such as audio data, telephone book, etc.). In addition, the memory 1201 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0261] The processor 1203 may include one or more processors such as a central processing unit (CPU), an application processor (AP), and a baseband processor. The processor can serve as the nerve center and command center of the wireless router. The processor 1203 can generate operation control signals based on instruction opcodes and timing signals to control instruction fetching and execution. The memory 1201 can be used to store executable program code, including instructions. The processor 1203 executes various functional applications and data processing of the network device by running the instructions stored in the memory. The memory 1201 may include a program storage area and a data storage area, such as storing data for audio signals to be played. For example, the memory may be Double Data Rate Synchronous Dynamic Random Access Memory (DDR) or Flash memory.

[0262] This application also provides a computer-readable storage medium storing computer instructions; when the computer-readable storage medium is used in a cleaning device, it causes the cleaning device to perform the dust collection method of the cleaning device described above.

[0263] The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or can include one or more data storage devices such as servers or data centers that can be integrated with media. The available medium can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media, or semiconductor media (e.g., solid-state disks (SSDs)).

[0264] This application also provides a computer program product containing computer instructions, which, when run on a cleaning device, enables the cleaning device to perform the dust collection and processing method described above.

[0265] The computer storage medium and computer program products provided in the embodiments of this application are used to execute the methods provided above. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects corresponding to the cleaning equipment provided above, and will not be repeated here.

[0266] In the above embodiments, implementation can also be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line, DSL) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer, or a data storage device such as a server or data center that integrates one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., Digital Versatile Discs (DVDs)), or semiconductor media (e.g., Solid State Disks (SSDs)).

[0267] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0268] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments claimed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0269] In the embodiments provided in this application, it should be understood that the disclosed apparatus / network devices and methods can be implemented in other ways. For example, the apparatus / network device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0270] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0271] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A cleaning device, characterized in that, include: Equipment body; A sensor assembly is connected to the main body of the device, and the sensor assembly collects sensor data of at least one target object within a preset area where the main body of the device is located; as well as A processor, disposed within the main body of the device and connected to the sensor assembly, wherein the processor: Acquire sensor data of the at least one target object from the sensor assembly; Process the sensor data of the at least one target object to generate feature information of the at least one target object; The amount of hair loss of the at least one target object is determined based on the feature information of the at least one target object; as well as Based on the amount of hair loss, feedback is provided on at least one of the following: the status of the dust bag and the filling time.

2. The cleaning device according to claim 1, characterized in that, The cleaning device further includes a memory, which is disposed inside the main body of the device and connected to the processor, wherein the processor further includes: Retrieve from the memory terminal acquisition information pre-stored in the memory and associated with the at least one target object, and The terminal acquisition information associated with the at least one target object is processed to generate feature information of the at least one target object.

3. The cleaning device according to claim 1, characterized in that, The cleaning device also includes a dust box connected to the main body of the device, wherein the processor also provides feedback on the dust collection frequency of the dust box based on the amount of hair shed.

4. The cleaning device according to claim 3, characterized in that, The processor also includes: Obtain the cleaning frequency and target cleaning mode of the main body of the device; and When the main body of the device is operating in the target cleaning mode, the dust collection frequency of the dust box is determined based on the amount of hair shed and the cleaning frequency. The target cleaning mode includes at least one of the following: sweeping mode, sweeping and mopping mode, and sweeping and mopping mode.

5. The cleaning device according to claim 1, characterized in that, The processor is also configured to provide feedback on at least one of the status of the dust bag and the filling time in any of the following ways: Based on the amount of hair loss and the total capacity of the dust bag, feedback is provided on at least one of the status of the dust bag and the filling time. Based on the cleaning area of ​​the main body of the device, the total capacity of the dust bag, and the amount of hair shed, feedback is provided on at least one of the status of the dust bag and the filling time.

6. The cleaning device according to claim 1, characterized in that, The processor provides feedback to the cleaning device or a terminal device communicatively connected to the cleaning device, at least one of the status of the dust bag and the filling time.

7. The cleaning device according to any one of claims 1 to 6, characterized in that, The processor also includes: Obtain the initial hair loss amount of the at least one target object; The feature information of the at least one target object and the initial hair loss amount are input into a pre-trained hair loss prediction model to obtain the current hair loss amount of the at least one target object, output by the hair loss prediction model. The hair loss prediction model is obtained by training an initial prediction model based on training samples. The training samples include: feature information of the sample object and the initial amount of hair loss, as well as the current amount of hair loss of the sample object.

8. The cleaning device according to any one of claims 1 to 6, characterized in that, The processor also includes: Obtain molting season information for at least one target object; Obtain the initial hair loss amount of the at least one target object; The feature information of the at least one target object, the molting season information, and the initial hair loss amount are input into a pre-trained hair loss prediction model to obtain the current hair loss amount of the at least one target object output by the hair loss prediction model. The hair loss prediction model is obtained by training an initial prediction model based on multiple training samples. The training samples include: feature information of the sample object, molting season information and initial hair loss amount, as well as the current hair loss amount of the sample object.

9. The cleaning device according to any one of claims 1 to 6, characterized in that, The sensor data includes image data and radar data, and the sensor components include: An image acquisition component, wherein the image acquisition component acquires image data of the at least one target object; A radar component is connected to the image acquisition component, and the radar component acquires radar data of the at least one target object when the image acquisition component acquires an image of the at least one target object.

10. The cleaning device according to claim 9, characterized in that, The processor also includes: Process the image data of the at least one target object to generate at least one of the number, body size, type and hair length of the at least one target object; as well as Process the radar data of the at least one target object to generate at least one of the height, volume, and shape of the at least one target object.

11. The cleaning device according to claim 9, characterized in that, The cleaning device further includes: a controller, which is disposed inside the main body of the device and connected to the processor. The controller responds to the control command received by the cleaning device and controls the cleaning device to start the dust collection mode. The radar component also collects radar data of the at least one target object when the image acquisition component acquires an image of the at least one target object during the operation of the cleaning device after the dust collection mode is activated.

12. A terminal device, characterized in that, The terminal device is communicatively connected to the cleaning device according to any one of claims 1 to 11, and includes: The display screen shows at least one of the status of the dust bag and the filling time, as reported by the cleaning device. A processor, connected to the display screen, runs a cleaning application and receives terminal collection information associated with the at least one target object uploaded by the cleaning application, and sends the terminal collection information to the cleaning device, wherein the cleaning device corresponds to the cleaning application.

13. A dust collection and treatment method for a cleaning device, characterized in that, The cleaning apparatus as described in any one of claims 1 to 12 includes: Acquire sensor data of at least one target object within a preset area where the main body of the cleaning equipment is located; Process the sensor data of the at least one target object to generate feature information of the at least one target object; Determine the amount of hair loss of the at least one target object based on the feature information of the at least one target object; and Based on the amount of hair loss, feedback is provided on at least one of the status of the dust bag and the filling time.

14. A cleaning device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it causes the cleaning device to perform the method as described in claim 13.

15. A computer program product, characterized in that, Includes a computer program, which, when run, causes the method of claim 13 to be performed.

16. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in claim 13.

Images