An internet-based sweeping robot filter screen abnormality early warning method and system

Abstract

The application relates to the technical field of sweeping robots, and provides an internet-based sweeping robot filter screen abnormality early warning method and system, which comprises the following steps: acquiring a plurality of running information of a sweeping robot; determining an abnormal type of a filter screen and an abnormal confidence of the abnormal type based on the plurality of running information; the abnormal type is dust permeability blockage or filter screen damage; determining an operation strategy of a user according to the abnormal type of the filter screen and the abnormal confidence of the abnormal type; the operation strategy is used for instructing the user to perform a maintenance operation on the filter screen; verifying the maintenance operation after the user performs the maintenance operation on the filter screen, and obtaining a verification result; the verification result is used for indicating that the user succeeds in or fails in the maintenance operation on the filter screen; when the verification result indicates that the user fails in the maintenance operation on the filter screen, sending error correction guidance information corresponding to the operation strategy to the user. Through the scheme, the accuracy of the sweeping robot filter screen abnormality early warning can be improved.

Description

Technical Field

[0001] This application relates to the field of robotic vacuum cleaner technology, and in particular to an internet-based method and system for early warning of filter abnormalities in robotic vacuum cleaners. Background Technology

[0002] Robotic vacuum cleaners play an important role in modern homes. They use internal fans to suck up dust and filter the air to prevent secondary pollution and protect the fan motor. However, with prolonged use, filters may become damaged, excessively clogged, or improperly installed. Currently, many inspection methods still rely on regular manual checks, which are not only slow to react but may also have blind spots in maintenance, leading to undetected problems and increasing the risk of fan motor overload or secondary pollution.

[0003] In existing technologies, robotic vacuum cleaners typically determine the filter's condition by monitoring the operating current of their internal fan. For example, when the filter becomes clogged with everyday debris such as hair and lint, airflow resistance increases, leading to a rise in the fan's operating current. When the current exceeds a preset limit, the system sends a notification to the user, "Filter clogged, please clean it promptly." However, this method of monitoring a single physical quantity has limitations in complex scenarios.

[0004] For example, when a robot vacuum cleaner sucks in fine dust, this dust may evenly penetrate the gaps in the filter fibers, forming a dense gas barrier layer, causing a sharp increase in airflow resistance. At this time, the increase in the fan's operating current is a slow and linear process, and it may not reach the preset alarm limit for a long time, causing users to fail to notice in time that the filter is severely clogged and the cleaning ability is significantly reduced.

[0005] For example, tiny tears or gaps may appear in the filter screen during cleaning or installation. This can cause some airflow to bypass the filter and enter the fan directly, creating an airflow "short circuit." This can actually reduce the fan's operating current, even below the normal value. In this case, the system may mistakenly judge the filter screen to be in "excellent" condition, but in reality, all the dust particles are discharged directly without being filtered, causing serious secondary pollution. Summary of the Invention

[0006] This application provides an internet-based method and system for early warning of filter abnormalities in robotic vacuum cleaners. It aims to solve the problems of existing methods for detecting filter abnormalities in robotic vacuum cleaners being too simplistic and unable to effectively identify complex abnormalities such as dust penetration blockage and filter damage, leading to untimely warnings, reduced cleaning ability, or secondary pollution.

[0007] To achieve the above objectives, this application adopts the following technical solution: Firstly, an internet-based method for early warning of filter anomalies in a robotic vacuum cleaner is provided, comprising the following steps: acquiring multiple operational information of the robotic vacuum cleaner, including the operating current of the robot's fan and the air pressure difference before and after filtration by the robot's filter; determining the anomaly type and confidence level of the anomaly type based on the multiple operational information, wherein the anomaly type is dust penetration blockage or filter damage; determining the user's operation strategy based on the anomaly type and confidence level of the anomaly type, which instructs the user to perform maintenance operations on the filter; verifying the maintenance operation after the user performs the maintenance operation and obtaining a verification result, which indicates whether the user's maintenance of the filter was successful or failed; and sending error correction guidance information corresponding to the operation strategy to the user when the verification result indicates that the user's maintenance of the filter failed.

[0008] Through this technical solution, this application can comprehensively analyze the operating information of the robot vacuum cleaner, identify filter abnormality types that are difficult to detect by traditional methods, and provide targeted operation strategies and error correction guidance. This effectively solves the problems of untimely and inaccurate filter abnormality warnings in the prior art, and improves the cleaning efficiency and user experience of the robot vacuum cleaner.

[0009] Furthermore, when determining the abnormality type and the abnormality confidence level of the filter based on multiple operational information, the method includes: determining whether the current difference of the fan's operating current within a preset time period is always positive, and whether the air pressure difference before and after filtration of the filter is always positive within the preset time period; wherein, the current difference is the difference between the operating current at one moment and the operating current at the previous moment, and the air pressure difference is the difference between the air pressure difference at one moment and the air pressure difference at the previous moment; if the current difference of the fan's operating current within the preset time period is always positive, and the air pressure difference before and after filtration of the filter is always positive within the preset time period, then the abnormality type is determined to be dust penetration blockage; otherwise, the abnormality type is determined not to be dust penetration blockage; when the abnormality type is dust penetration blockage, the abnormality confidence level of the abnormality type is determined based on the operating current of the fan of the filter and the air pressure difference before and after filtration of the filter.

[0010] This technical solution enables the accurate identification of dust-penetrating blockage, a special type of anomaly, by monitoring the changing trends of the fan's operating current and air pressure difference. This compensates for the shortcomings of monitoring a single physical quantity and improves the accuracy of anomaly detection.

[0011] More specifically, if the current difference of the fan's operating current is positive within a preset time period, and the air pressure difference before and after filtration by the filter is positive within the preset time period, and the anomaly type is determined to be dust penetration blockage, the method includes: if the current difference of the fan's operating current is positive within a preset time period, and the air pressure difference before and after filtration by the filter is positive within the preset time period, obtaining the particulate matter concentration in the air detected by the particulate matter sensor installed at the filter outlet within the preset time period; if the absolute value of the particulate matter concentration difference within the preset time period is less than a first preset concentration difference threshold, determining the anomaly type to be dust penetration blockage; otherwise, determining the anomaly type not to be dust penetration blockage; wherein, the particulate matter concentration difference is the difference between the particulate matter concentration at one moment and the particulate matter concentration at the previous moment.

[0012] By introducing particulate matter concentration as an auxiliary criterion through this technical solution, the accuracy of identifying dust-induced blockage is further improved, misjudgment is avoided, and abnormal warnings are made more reliable.

[0013] Based on the above, this application further proposes that when the anomaly type is dust penetration blockage, the anomaly confidence level of the anomaly type is determined based on the operating current of the fan of the filter and the air pressure difference before and after filtration of the filter. The method includes: for each operating current of the fan within a preset time period, obtaining the normal air pressure difference of the operating current, which is the air pressure difference before and after filtration corresponding to the operating current when the filter is in a normal state; determining the similarity between multiple normal air pressure differences and multiple air pressure differences before and after filtration of the filter in the operation information, and taking the reciprocal of the air pressure difference similarity as the anomaly confidence level of the anomaly type.

[0014] This technical solution enables the quantification of the severity of dust penetration blockage, providing a more refined basis for subsequent operational strategies and making maintenance operations more targeted.

[0015] As an optional solution, when determining the abnormality type and the abnormality confidence level of the filter based on multiple operational information, the method includes: determining whether the operating current of the fan is always less than a preset current value within a preset time period, and whether the air pressure difference before and after filtration by the filter is always negative within the preset time period; wherein, the air pressure difference is the difference between the air pressure difference at one moment and the air pressure difference at the previous moment; if the operating current of the fan is always less than the preset current value within the preset time period, and the air pressure difference before and after filtration by the filter is always negative within the preset time period, then the abnormality type is determined to be filter damage; otherwise, the abnormality type is determined not to be filter damage; when the abnormality type is filter damage, the concentration of outlet particulate matter in the air detected by the particulate matter sensor set at the air outlet of the filter and the concentration of inlet particulate matter in the air detected by the particulate matter sensor set at the air inlet of the filter are obtained; the abnormality confidence level of the abnormality type is determined based on the outlet particulate matter concentration and the inlet particulate matter concentration.

[0016] This technical solution effectively identifies filter damage, a serious anomaly that leads to secondary pollution, and solves the problem that traditional methods cannot detect airflow "short circuits," thus ensuring indoor air quality.

[0017] In one embodiment, if the operating current of the fan is less than a preset current value within a preset time period, and the air pressure difference before and after filtration by the filter is negative within the preset time period, the abnormality type is determined to be filter damage. The method includes: if the operating current of the fan is less than a preset current value within a preset time period, and the air pressure difference before and after filtration by the filter is negative within the preset time period, acquiring detection audio data set on the air inlet side of the filter and normal audio data on the air inlet side of the filter when the filter is normal; if the similarity between the detection audio data and the normal audio data is less than a preset first similarity threshold, the abnormality type is determined to be filter damage; otherwise, the abnormality type is determined not to be filter damage.

[0018] By introducing audio data as a basis for judgment through this technical solution, the accuracy of identifying filter damage is further improved, providing users with more reliable early warning.

[0019] In another embodiment, the anomaly confidence level of the anomaly type is determined based on the outlet particulate matter concentration and the inlet particulate matter concentration. The method includes: taking the difference between the inlet particulate matter concentration and the outlet particulate matter concentration as a concentration difference; obtaining a first correspondence, which includes a one-to-one correspondence between multiple concentration difference ranges and multiple anomaly confidence levels, wherein the anomaly confidence level is negatively correlated with the maximum value of the corresponding concentration difference range; and taking the anomaly confidence level corresponding to the concentration difference range in the first correspondence as the anomaly confidence level of the anomaly type.

[0020] This technical solution enables the quantification of the severity of filter damage, providing users with more intuitive abnormal information and facilitating timely maintenance.

[0021] To enhance functionality, the method determines the user's operation strategy based on the anomaly type of the filter and the anomaly confidence level of the anomaly type. The method includes: determining whether the anomaly confidence level of the anomaly type is greater than a preset anomaly confidence level threshold; if so, obtaining a second correspondence, which includes multiple anomaly types and multiple operation strategies; using the operation strategy corresponding to the anomaly type of the filter in the second correspondence as the user's operation strategy; if not, storing the anomaly type of the filter and multiple operation information to a cloud server.

[0022] Through this technical solution, this application can intelligently recommend operation strategies to users based on the severity of anomalies, and store data for less serious anomalies, providing data support for subsequent optimization and analysis, thereby improving the intelligence level of the system.

[0023] To improve the solution, after the user performs maintenance on the filter, the maintenance operation is verified to obtain the verification result. The method includes: controlling the robot vacuum to start a preset self-test program to obtain multiple self-test data of the robot vacuum; when the similarity between the multiple self-test data and multiple preset normal self-test data is greater than a preset second similarity threshold, the verification result is determined to indicate that the user has successfully maintained the filter; otherwise, the verification result is determined to indicate that the user has failed to maintain the filter.

[0024] Through this technical solution, this application can effectively verify the user's maintenance operations, ensure the maintenance effect, avoid problems caused by improper maintenance, and improve the user's trust in the system.

[0025] Secondly, this application also discloses an internet-based early warning system for an abnormal filter screen of a robotic vacuum cleaner, comprising an acquisition device and a processing device; the acquisition device is used to acquire multiple operational information of the robotic vacuum cleaner, including the operating current of the robotic vacuum cleaner's fan and the air pressure difference before and after filtration by the robotic vacuum cleaner's filter screen; the processing device is used to determine the abnormality type of the filter screen and the abnormality confidence level of the abnormality type based on the multiple operational information, wherein the abnormality type is dust penetration blockage or filter screen damage; the processing device is used to determine the user's operation strategy based on the abnormality type of the filter screen and the abnormality confidence level of the abnormality type, the operation strategy being used to instruct the user to perform maintenance operations on the filter screen; the processing device is used to verify the maintenance operation after the user performs the maintenance operation on the filter screen and obtain a verification result, the verification result being used to indicate whether the user's maintenance of the filter screen was successful or failed; the processing device is used to send error correction guidance information corresponding to the operation strategy to the user when the verification result indicates that the user's maintenance of the filter screen failed.

[0026] Beneficial Effects: This application discloses an internet-based early warning method for abnormal filter conditions in robotic vacuum cleaners. It acquires multiple operational information, such as the operating current of the robot's fan and the air pressure difference before and after filtration, and comprehensively judges the type of filter abnormality (e.g., dust penetration clogging or filter damage) and its confidence level based on this information. This allows the system to determine and send targeted operational strategies to the user. After the user performs maintenance, the system also verifies the maintenance effect and provides error correction guidance if maintenance fails.

[0027] Compared to existing methods that solely monitor the fan's operating current, this application overcomes its limitations in complex scenarios. For example, in cases of dust-induced blockage, existing methods fail to provide timely warnings due to slow current growth, leading to reduced cleaning efficiency. This application, however, accurately identifies such anomalies by combining the fan's operating current with the changing trends of air pressure difference. Regarding filter damage, existing methods may misjudge the filter's condition as "excellent" due to a decrease in current caused by airflow "short circuits," resulting in secondary pollution. This application effectively identifies filter damage, avoiding this problem.

[0028] Through the above technical solution, this application effectively solves the technical problems of existing methods for detecting filter anomalies in robotic vacuum cleaners, such as slow response, maintenance blind spots, inability to detect problems in a timely manner, and potential overload of the fan motor or secondary pollution. This application, through multi-dimensional data analysis and intelligent early warning mechanisms, significantly improves the accuracy and timeliness of filter anomaly detection, ensuring the cleaning efficiency and lifespan of the robotic vacuum cleaner, while avoiding secondary pollution, providing users with a more intelligent and reliable robotic vacuum cleaner maintenance experience. Attached Figure Description

[0029] Figure 1 A flowchart illustrating an internet-based early warning method for abnormal filter screens in a robotic vacuum cleaner, as provided in this application; Figure 2 A flowchart illustrating another Internet-based early warning method for abnormal filter screens in a robotic vacuum cleaner provided in this application; Figure 3 This application provides a structural schematic diagram of an Internet-based early warning system for the abnormal filter screen of a robotic vacuum cleaner. Detailed Implementation

[0030] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0031] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0032] Traditional robotic vacuum cleaners often rely on regular manual inspections or monitoring of a single physical quantity when filter problems occur. This can lead to delayed detection of issues, potentially causing overload of the fan motor or secondary pollution. For example, when fine dust evenly penetrates the filter fibers, forming a dense gas barrier layer, the increase in the fan's operating current is a slow and linear process. It may not reach the preset alarm limit for an extended period, preventing users from noticing severe filter blockage and significantly reducing cleaning efficiency. Furthermore, tiny tears or gaps may appear in the filter during cleaning or installation, allowing some airflow to bypass the filter and enter the fan directly, creating an airflow "short circuit." This can cause the fan's operating current to decrease, even falling below normal levels, resulting in serious secondary pollution.

[0033] In this regard, such as Figure 1 As shown, this application proposes an internet-based method for early warning of filter anomalies in robotic vacuum cleaners, including: S101. Obtain multiple operational information from the robotic vacuum cleaner.

[0034] Several operational information items include the operating current of the robot vacuum's fan and the air pressure difference before and after filtration by the robot vacuum's filter.

[0035] S102. Determine the abnormality type of the filter and the abnormality confidence level of the abnormality type based on multiple operating information; the abnormality type is dust penetration blockage or filter damage.

[0036] S103. Determine the user's operation strategy based on the filter's anomaly type and the anomaly confidence level of the anomaly type; the operation strategy is used to instruct the user to perform maintenance operations on the filter.

[0037] S104. After the user performs maintenance operations on the filter, the maintenance operations are verified, and the verification results are obtained.

[0038] The verification result is used to indicate whether the user's maintenance of the filter was successful or failed.

[0039] S105. When the verification result indicates that the user has failed to maintain the filter, send the user the error correction guidance information corresponding to the operation strategy.

[0040] This application, by comprehensively analyzing multiple operational information such as the working current of the robot vacuum's fan and the air pressure difference before and after filtration, can more accurately identify the abnormal type of the filter (such as dust penetration blockage or filter damage) and its abnormality confidence level. Based on this, personalized operation strategies and error correction guidance information are generated, thereby effectively solving the problem of untimely and inaccurate early warning in complex abnormal scenarios using traditional methods, and significantly improving the maintenance efficiency and user experience of robot vacuums.

[0041] To better understand the technical solution proposed in this application, it is necessary to explain some key terms and implementation environments involved. A robotic vacuum cleaner is an intelligent device capable of autonomously cleaning floors, typically containing a fan and a filter. The fan generates suction, drawing air and dust into the robot; the filter filters dust and particulate matter from the air, preventing them from entering the fan and expelling clean air. Operational information refers to various data generated by the robotic vacuum cleaner during operation, such as the fan's operating current and the air pressure difference before and after filtration. The fan's operating current reflects the fan's load during operation, while the air pressure difference before and after filtration directly reflects the filter's resistance. Anomaly type refers to specific problems that may occur with the filter, such as dust penetration clogging (meaning the filter is uniformly clogged by fine dust) or filter damage (meaning the filter has physical damage). Anomaly confidence indicates the degree of certainty regarding the identified anomaly type; a higher value indicates a greater probability of the anomaly type occurring. Operational strategy refers to maintenance suggestions provided by the system to the user based on the anomaly situation, such as "Please clean the filter" or "Please replace the filter." Error correction guidance information is further guidance provided by the system when user maintenance fails, helping users to correctly resolve the problem. This method is typically implemented in an internet-based system, where the robot vacuum cleaner uploads its operating information to a cloud server, which analyzes and processes the information and sends alerts and guidance messages to the user's smart device.

[0042] The method proposed in this application first requires acquiring multiple operational information data from the robotic vacuum cleaner. This operational information is fundamental for determining the filter's condition. For example, a current sensor can be integrated inside the robotic vacuum cleaner to monitor the fan's operating current in real time, and this data can be uploaded to a cloud server via a wireless communication module. Simultaneously, pressure sensors can be placed before and after the filter to measure the air pressure difference before and after filtration; this pressure difference data is also collected and uploaded. These sensors can periodically collect data, for example, once per second, to ensure the real-time nature and continuity of the data.

[0043] After acquiring this operational information, the system determines the anomaly type of the filter and the anomaly confidence level for that type. Anomaly types primarily include dust penetration clogging or filter damage. For example, the system can preset rules or models; when the changing trends of the fan's operating current and the air pressure difference before and after filtration conform to a certain pattern, it is identified as a specific anomaly type. For instance, a continuous increase in the fan's operating current and a continuous increase in the air pressure difference before and after filtration may indicate filter clogging. Conversely, an abnormal decrease in the fan's operating current and an abnormal decrease in the air pressure difference before and after filtration may indicate filter damage. The anomaly confidence level can be calculated using machine learning models or statistical analysis methods, such as comparing the similarity between current operational information and historical normal and abnormal data.

[0044] Subsequently, based on the type of filter anomaly and its confidence level, the system determines the user's operating strategy. This strategy guides the user in performing filter maintenance. For example, if the anomaly is identified as dust penetration clogging with a high confidence level, the system might generate a "Please clean the filter" strategy. If the anomaly is identified as filter damage with a high confidence level, the system might generate a "Please replace the filter" strategy. These strategies can be sent to the user via the robot vacuum's companion app or a smart speaker.

[0045] After a user performs maintenance on the filter, the system verifies the operation to obtain a result. The verification result indicates whether the filter maintenance was successful or failed. For example, after cleaning or replacing the filter, the system can control the robot vacuum to run for a period of time and then re-acquire its operating information. By analyzing this new operating information, such as whether the fan's operating current and the air pressure difference before and after filtration have returned to normal, the system determines whether the maintenance operation was effective. If the operating information returns to normal, the verification result indicates successful maintenance; otherwise, the verification result indicates maintenance failure.

[0046] Finally, when the verification result indicates that the user's filter maintenance has failed, the system will send the user corrective guidance information corresponding to the operation strategy. This guidance is designed to help users resolve the maintenance failure. For example, if the filter still shows cloggedness after cleaning, the system might send guidance such as "Please check if the filter is completely dry, or try using more professional cleaning tools." If the filter still shows damage after replacement, the system might send guidance such as "Please check if the filter is installed correctly, or confirm that the purchased filter model is correct." This guidance provides more specific and targeted assistance to ensure the problem is ultimately resolved.

[0047] The proposed internet-based early warning method for robot vacuum cleaner filters comprehensively and in real-time monitors the filter's operational status by acquiring multiple operational information, such as the robot vacuum's fan operating current and the air pressure difference before and after filtration. When these operational information become abnormal, the system can accurately identify the type of filter abnormality based on preset logic or models, such as whether it is dust penetration clogging or filter damage, and provide a corresponding anomaly confidence level. This multi-dimensional data analysis avoids the limitations of traditional single-physical-quantity monitoring. For example, when dust penetration clogging causes a slow increase in the fan operating current, traditional methods may not provide timely warnings, while this application, by combining changes in air pressure difference, can detect problems earlier. Similarly, when filter damage causes a decrease in the fan operating current, traditional methods may misjudge it as normal, while this application, through comprehensive analysis, can accurately identify the damage.

[0048] Once the anomaly type and confidence level are determined, the system intelligently generates a user-specific operation strategy to guide maintenance. For example, for dust-induced blockage, the system suggests cleaning the filter; for a damaged filter, it suggests replacing it. This personalized operation strategy ensures users take the correct maintenance measures. More importantly, this application also introduces a maintenance operation verification mechanism. After the user completes maintenance, the system re-tests the robot vacuum to confirm its success. If the verification result indicates maintenance failure, the system does not simply stop but further sends the user corrective guidance information corresponding to the operation strategy. This corrective guidance information provides detailed solutions for specific problems the user may encounter, such as guiding the user to check the filter installation, cleaning method, or replacement part model.

[0049] Through the aforementioned collaborative efforts, the overall technical solution of this application forms a closed-loop intelligent early warning and maintenance system. It can not only detect filter abnormalities promptly and accurately, but also provide users with comprehensive support from initial maintenance to troubleshooting guidance, greatly improving the maintenance efficiency and user experience of the robotic vacuum cleaner, and effectively avoiding risks such as reduced cleaning ability, fan motor overload, or secondary pollution caused by filter problems.

[0050] The internet-based early warning method for abnormal filter screens in robotic vacuum cleaners proposed in this application has significant advantages and innovations compared to existing technologies. Traditional methods mainly rely on regular manual inspections or monitoring of a single physical quantity (such as the fan's operating current). This approach often exhibits limitations in responding to complex and ever-changing filter screen anomalies, resulting in slow response and inaccurate judgment. For example, when the filter screen becomes clogged by dust penetration, the increase in the fan's operating current is a slow and linear process, which may not trigger the preset alarm threshold for a long time, preventing users from detecting the problem in time. Furthermore, when the filter screen has minor damage or installation gaps, the fan's operating current may even decrease, which the system may mistakenly interpret as normal, causing serious secondary pollution.

[0051] The core innovation of this application lies in its approach: it goes beyond monitoring a single physical quantity, comprehensively acquiring and analyzing multiple operational information, such as the operating current of the robot vacuum's fan and the air pressure difference before and after filtration. This multi-dimensional data fusion enables the system to perceive the true state of the filter more comprehensively and accurately. Through intelligent analysis of this information, this application can distinguish between different types of anomalies, such as dust penetration blockage and filter damage, and quantify their anomaly confidence levels. This refined anomaly identification capability is unmatched by existing technologies.

[0052] Furthermore, this application introduces an intelligent operation strategy generation and maintenance verification mechanism. Based on the identified anomaly type and confidence level, the system can automatically provide users with personalized maintenance suggestions. Moreover, after the user performs the maintenance operation, the system will verify it to confirm whether the maintenance was successful. If the maintenance fails, the system will provide targeted error correction guidance to help users solve problems that may be encountered in actual operation. This closed-loop management process from early warning and guidance to verification and error correction greatly improves the efficiency and accuracy of user maintenance, effectively avoiding maintenance failures caused by users' lack of professional knowledge in traditional methods.

[0053] In summary, this application significantly improves the accuracy and timeliness of early warning for abnormal filter screens in robotic vacuum cleaners by integrating multi-source information, intelligent anomaly recognition, personalized operation strategies, and closed-loop maintenance verification mechanisms. This provides users with a more intelligent and efficient maintenance experience and effectively solves the problems of untimely and inaccurate early warnings and insufficient maintenance guidance in existing technologies. It has significant technological progress and practical value.

[0054] This application further proposes methods for determining the anomaly type of the filter and the anomaly confidence level of the anomaly type based on multiple operational information, including: Determine whether the current difference of the fan's operating current within a preset time period is always positive, and whether the air pressure difference before and after filtration by the filter is always positive within the preset time period; the current difference is the difference between the operating current at one moment and the operating current at the previous moment, and the air pressure difference is the difference between the air pressure difference at one moment and the air pressure difference at the previous moment; if the current difference of the fan's operating current within the preset time period is always positive, and the air pressure difference before and after filtration by the filter is always positive within the preset time period, determine the anomaly type as dust penetration blockage; otherwise, determine the anomaly type as not dust penetration blockage; when the anomaly type is dust penetration blockage, determine the anomaly confidence level of the anomaly type based on the fan's operating current and the air pressure difference before and after filtration by the filter.

[0055] Specifically, the preset duration refers to a specific time period during which the robot vacuum cleaner continuously runs or samples data, such as several minutes or hours. Within this preset duration, the system periodically collects the operating current of the fan and the air pressure difference before and after filtration by the filter. The current difference can be understood as the difference between the fan operating current collected at the current moment and the fan operating current collected at the previous moment, reflecting the instantaneous trend of the fan operating current. The air pressure difference can be understood as the difference between the air pressure difference before and after filtration by the filter collected at the current moment and the air pressure difference before and after filtration by the filter collected at the previous moment, reflecting the instantaneous trend of the pressure difference across the filter.

[0056] When all calculated current and air pressure differences are positive throughout the preset time period, it indicates that both the fan's operating current and the air pressure difference before and after filtration are showing a continuous upward trend. Under this condition, the filter anomaly can be identified as dust penetration clogging. Conversely, if the above conditions are not met, the anomaly type is not determined to be dust penetration clogging. Once the anomaly type is determined to be dust penetration clogging, the anomaly confidence level will be further determined based on the current fan operating current and the air pressure difference before and after filtration.

[0057] This application's solution effectively identifies dust-penetrating blockage by monitoring the dynamic changes in the fan's operating current and the air pressure difference before and after filtration. When dust-penetrating blockage occurs in the filter, the filter's permeability continuously decreases as dust gradually accumulates, leading to increased airflow resistance. To maintain normal airflow or air pressure inside the robot vacuum, the fan needs to output more power, causing the fan's operating current to show a continuous upward trend. Simultaneously, due to the increased airflow resistance, the air pressure difference across the filter also continuously increases. It is precisely because the continuous increase in fan operating current and the continuous increase in air pressure difference before and after filtration are typical physical manifestations of dust-penetrating blockage that, by judging whether the difference in fan operating current and the difference in air pressure difference before and after filtration are both positive within a preset time period, the gradual blockage state of the filter due to dust accumulation can be accurately reflected, thereby achieving precise identification of this specific abnormal type of dust-penetrating blockage.

[0058] Through the above technical solution, this application provides a more refined and accurate method for identifying filter anomaly types. Compared to general anomaly judgments based solely on operational information, this solution introduces the judgment of changes in fan operating current and air pressure difference, specifically requiring these differences to be positive within a preset time period. This enables the specific identification of dust penetration blockage as a particular anomaly type. This not only improves the accuracy of anomaly warnings but also provides a more specific and targeted basis for determining subsequent user operation strategies, avoiding misjudgments or omissions, thereby improving the efficiency of robot vacuum maintenance and the user experience.

[0059] This application further proposes a scheme in which, if the current difference of the fan's operating current is positive within a preset time period, and the air pressure difference before and after filtration by the filter is positive within the preset time period, the steps to determine the abnormality type as dust penetration blockage include: S201. If the current difference of the working current of the fan is positive within a preset time period, and the air pressure difference before and after filtration of the filter screen is positive within a preset time period, obtain the concentration of particulate matter in the air detected by the particulate matter sensor set at the air outlet of the filter screen within a preset time period.

[0060] S202. When the absolute value of the particulate matter concentration difference within a preset time period is less than the first preset concentration difference threshold, the abnormality type is determined to be dust penetration blockage; otherwise, the abnormality type is determined not to be dust penetration blockage. The particulate matter concentration difference is the difference between the particulate matter concentration at one moment and the particulate matter concentration at the previous moment.

[0061] Specifically, a particulate matter sensor can be understood as a device capable of monitoring the concentration of particulate matter in the air in real time. It is placed at the air outlet of a filter to directly detect the air quality after it has passed through the filter. Its purpose is to provide direct evidence of the filter's actual filtration effectiveness. The particulate matter concentration difference is the difference between the particulate matter concentration at one moment and the particulate matter concentration at the previous moment, used to assess the trend of particulate matter concentration change at the filter outlet over a preset time period. In practical applications, the first preset concentration difference threshold is a pre-set value used to define the allowable range of particulate matter concentration variation. When the absolute value of the particulate matter concentration difference is less than this threshold, it indicates that although the filter may be clogged, its filtration capacity for particulate matter remains relatively stable, and no significant particulate matter leakage has occurred, thus further confirming the anomaly as dust-penetrating blockage.

[0062] This application's solution monitors the particulate matter concentration at the filter outlet using a particulate matter sensor and, combined with changes in the particulate matter concentration difference, provides a secondary confirmation of the initial assessment of dust-induced blockage. When both the fan operating current and air pressure difference show an upward trend, indicating potential filter blockage, if the change in particulate matter concentration at the outlet is small (i.e., the absolute value of the particulate matter concentration difference is less than a first preset concentration difference threshold), it indicates that although the filter is blocked by dust penetration, its structural integrity remains intact and it can still effectively block particulate matter from passing through. This effectively avoids misjudgments that may result from relying solely on indirect parameters (current and pressure difference) and improves the accuracy of anomaly type determination.

[0063] By employing the aforementioned technical solution, based on the initial assessment that the filter may be clogged by dust penetration, further monitoring and analysis of the particulate matter concentration at the air outlet using a particulate matter sensor can more accurately and reliably confirm whether the filter's anomaly is indeed due to dust penetration clogging. This significantly improves the accuracy of anomaly warnings, avoiding unnecessary maintenance or delays in repairing actual faults caused by misjudgments, thereby enhancing the stability of the robot vacuum cleaner's operation and the user experience.

[0064] This application further proposes a method for determining the anomaly confidence level of an anomaly type.

[0065] When the above method identifies dust-penetrating blockage as the anomaly type, it determines the anomaly confidence level based on the operating current of the fan and the air pressure difference before and after filtration. The specific steps include: For each operating current of the fan within a preset time period, obtain the normal air pressure difference of the operating current; the normal air pressure difference is the air pressure difference before and after filtration corresponding to the operating current when the filter is in normal condition; determine the similarity between multiple normal air pressure differences and multiple air pressure differences before and after filtration of the filter in the operation information, and use the reciprocal of the air pressure difference similarity as the anomaly confidence of the anomaly type.

[0066] Specifically, normal air pressure difference refers to the air pressure difference generated before and after the filter of a robotic vacuum cleaner when the filter is in an ideal, unblocked, and undamaged normal operating condition and the vacuum cleaner's fan is running at a specific operating current. This normal air pressure difference can be obtained through a series of experimental measurements before the robotic vacuum cleaner leaves the factory or after the filter is replaced with a brand new one, and stored in the device's local memory or cloud server to form a table or curve showing the correspondence between operating current and normal air pressure difference. Its purpose is to provide a benchmark reference for subsequent anomaly detection. Similarity can be understood as a quantitative indicator of the degree of matching between two data sequences or patterns.

[0067] In this application, similarity is used to measure the degree of closeness between the air pressure difference sequence before and after filtration measured during actual operation and the preset normal air pressure difference sequence within a preset time period. The similarity can be calculated using various mathematical or statistical methods, such as cosine similarity, the reciprocal of Euclidean distance, and Pearson correlation coefficient. Its purpose is to objectively reflect the degree of deviation between the current air pressure difference state of the filter and the normal state. In practical applications, the reciprocal of the air pressure difference similarity is used as the anomaly confidence level for the anomaly type, meaning that the lower the similarity (i.e., the greater the deviation between the actual operating state and the normal state), the higher the anomaly confidence level. For example, if the similarity is 1, indicating complete normality, then the reciprocal is 1, which can be set as the lowest anomaly confidence level; if the similarity approaches 0, indicating severe anomaly, then the reciprocal approaches infinity, which can be set as the highest anomaly confidence level. In this way, the relative indicator of similarity can be transformed into a confidence value that intuitively reflects the degree of anomaly, facilitating subsequent decision-making by the system.

[0068] This application's solution quantifies the confidence level of filter anomalies by introducing a normal air pressure difference as a benchmark and calculating the similarity between the actual operating air pressure difference and this benchmark. When the filter becomes clogged by dust penetration, its resistance to airflow increases, resulting in a higher air pressure difference before and after the filter than normal under the same fan operating current. By obtaining the normal air pressure difference corresponding to each operating current of the fan within a preset time period, an ideal air pressure difference curve can be constructed.

[0069] Subsequently, the air pressure difference data measured during actual operation was compared with the normal air pressure difference data for similarity. If the filter is clogged by dust penetration, the actual air pressure difference will deviate from the normal air pressure difference, resulting in a decrease in the similarity between the two. The reciprocal of the air pressure difference similarity was used as the anomaly confidence level, so that the lower the similarity, the higher the anomaly confidence level, thus accurately reflecting the severity of the filter clogging. This quantitative method based on comparing actual data with a normal benchmark effectively solves the problem of fuzzy or inaccurate anomaly confidence level judgment in traditional methods.

[0070] Through the above technical solution, this application provides a more accurate and quantitative method to determine the anomaly confidence level of dust penetration clogging in the filter. By establishing a benchmark of normal air pressure difference and calculating the similarity between actual operating data and this benchmark, the degree of filter anomaly can be objectively assessed. This method avoids the limitations of subjective judgment or simple threshold determination, making the assessment of anomaly confidence level more scientific and reliable. Therefore, the system can more accurately identify the abnormal state of the filter, providing users with more timely and precise maintenance operation strategies, thereby improving the operating efficiency and lifespan of the robotic vacuum cleaner.

[0071] This application further proposes methods for determining the anomaly type of the filter and the anomaly confidence level of the anomaly type based on multiple operational information, including: Determine whether the fan's operating current is consistently lower than the preset current value within a preset time period, and whether the air pressure difference before and after filtration by the filter is consistently negative within the preset time period; the air pressure difference is the difference between the air pressure difference at one moment and the air pressure difference at the previous moment. If the fan's operating current is consistently lower than the preset current value within the preset time period, and the air pressure difference before and after filtration by the filter is consistently negative within the preset time period, the anomaly type is determined to be filter damage; otherwise, the anomaly type is determined not to be filter damage. When the anomaly type is filter damage, obtain the outlet particulate matter concentration detected by the particulate matter sensor located at the filter outlet and the inlet particulate matter concentration detected by the particulate matter sensor located at the filter inlet; determine the anomaly confidence level of the anomaly type based on the outlet particulate matter concentration and the inlet particulate matter concentration.

[0072] Specifically, the preset duration refers to a period of time used to observe changes in the robot vacuum's operating status. This could be several seconds, minutes, or hours, and its length can be set according to the actual application scenario and the timeliness requirements for responding to anomalies. The fan's operating current refers to the actual current consumed by the robot vacuum's internal fan during operation; its magnitude reflects the fan's workload and air resistance. The preset current value is a threshold set based on the fan's operating current range under normal operating conditions. When the fan's operating current consistently falls below this threshold, it may indicate an abnormal decrease in air resistance.

[0073] The air pressure difference before and after the filter refers to the change in air pressure difference before and after the filter measured at different times. When the air pressure difference at one time is negative compared to the air pressure difference at the previous time, it means that the air pressure difference before and after the filter is continuously decreasing. This is usually due to damage to the filter, which reduces the airflow resistance.

[0074] A particulate matter sensor can be understood as a device that detects the number or concentration of particulate matter in the air; for example, it could be a laser scattering particulate matter sensor or an infrared particulate matter sensor. The filter outlet refers to the location where air flows out after being processed by the filter, while the filter inlet refers to the location where the air to be filtered enters the filter. The outlet particulate matter concentration and inlet particulate matter concentration refer to the content of particulate matter in the air detected at the filter outlet and inlet, respectively. By comparing the particulate matter concentrations at these two locations, the filtration efficiency of the filter and whether it is damaged can be directly assessed.

[0075] This application's solution effectively identifies abnormal filter damage by monitoring changes in the fan's operating current and the air pressure difference before and after the filter. When the filter is damaged, its resistance to airflow decreases significantly, causing the fan to operate at the same or lower speeds with a reduced workload, resulting in a continuously lower operating current than the normal preset value. Simultaneously, due to the reduced air resistance, the air pressure difference before and after the filter also decreases, even showing a trend towards a negative pressure difference. It is precisely because of these abnormal changes in physical quantities that this application can preliminarily determine that the filter may be damaged.

[0076] Furthermore, by installing particulate matter sensors at both the inlet and outlet of the filter, the inlet and outlet particulate matter concentrations can be acquired in real time. In the event of filter damage, particles that should have been trapped by the filter will pass directly through the damaged area, causing an abnormally high outlet particulate matter concentration, even approaching or equaling the inlet particulate matter concentration. Analysis of this concentration data allows for the quantification of the degree of filter damage, thereby accurately determining the anomaly confidence level for the anomaly type.

[0077] Through the above technical solution, this application provides a more accurate and reliable identification and early warning mechanism for the specific anomaly type of filter damage in robotic vacuum cleaners. By comprehensively analyzing the fan operating current, changes in filter air pressure difference, and inlet / outlet particulate matter concentration, false positives and false negatives can be effectively avoided, improving the accuracy of anomaly warnings. Furthermore, by quantifying the anomaly confidence level, more instructive maintenance operation suggestions can be provided to users, thereby extending the filter's lifespan, ensuring the cleaning effect of the robotic vacuum cleaner, and enhancing the user experience.

[0078] This application further proposes that if the operating current of the fan is less than the preset current value within a preset time period, and the air pressure difference before and after filtration by the filter is negative within the preset time period, the abnormality type is determined to be filter damage, including: If the fan's operating current is less than the preset current value within a preset time period, and the air pressure difference before and after filtration by the filter is negative within the preset time period, acquire the detection audio data set on the air inlet side of the filter and the normal audio data on the air inlet side of the filter when the filter is normal; if the similarity between the detection audio data and the normal audio data is less than the preset first similarity threshold, determine the anomaly type as filter damage; otherwise, determine the anomaly type as not filter damage.

[0079] Specifically, detection audio data refers to the environmental sound data collected in real time by a microphone or other acoustic sensors located on the air intake side of the filter during the operation of the robotic vacuum cleaner. Normal audio data refers to baseline audio data collected under the same or similar operating conditions when the filter is in a brand new or known normal state. This normal audio data can be pre-stored on the robotic vacuum cleaner's local machine or a cloud server as a comparison reference.

[0080] Similarity can be understood as the degree of matching between audio data and normal audio data in terms of spectral features, volume variations, and specific frequency components. For example, it can be achieved by calculating the cross-correlation coefficient, Euclidean distance, dynamic time warping (DTW) distance between two audio segments, or by using machine learning models for feature extraction and similarity evaluation. Its purpose is to quantify the difference between the sound characteristics of the current filter in operation and the sound characteristics under normal conditions.

[0081] The preset first similarity threshold is a pre-defined value used to define the acceptable range of similarity between the detected audio data and normal audio data. When the similarity is below this threshold, it indicates that the sound characteristics of the current filter differ significantly from the normal state, thus suggesting that the filter may be damaged. This threshold can be set and optimized through extensive experimental data and expert experience to balance detection sensitivity and false alarm rate.

[0082] This application's solution, by introducing the analysis of audio data from the air inlet side of the filter, effectively overcomes the shortcomings of relying solely on current and pressure difference data for filter damage assessment. When a filter is damaged, such as by cracks or holes, the airflow passing through the damaged area will generate abnormal eddies or whistling sounds. These sound characteristics are significantly different from the smooth airflow sound produced by a normal filter during operation. By acquiring the detected audio data and comparing its similarity with normal audio data, these subtle sound changes can be captured. When the similarity between the detected audio data and normal audio data is less than a preset first similarity threshold, it indicates the presence of abnormal sound characteristics, thus providing a more sensitive and accurate indication that the filter may be damaged. This acoustic feature-based detection method provides a new dimension for early warning of filter damage, and is particularly suitable for detecting initial or minor damage that has not yet significantly affected the fan's operating current and air pressure difference.

[0083] Through the above technical solution, this application provides a more sensitive and accurate filter damage detection mechanism. Compared to traditional methods that rely solely on fan operating current and air pressure difference, introducing audio data analysis can capture the unique acoustic characteristics produced by filter damage, thus issuing an early warning when the damage is in its early stages or minor. This significantly improves the timeliness and accuracy of filter anomaly warnings, helping users to detect and address filter problems earlier, avoiding decreased cleaning efficiency and secondary pollution caused by filter damage, thereby extending the lifespan of the robot vacuum and improving the user experience.

[0084] This application further proposes a method for determining the anomaly confidence level of anomaly types based on export particulate matter concentration and import particulate matter concentration, including: The difference between the concentration of imported particulate matter and the concentration of exported particulate matter is taken as the concentration difference; the first correspondence is obtained; the first correspondence includes a one-to-one correspondence between multiple concentration difference ranges and multiple anomaly confidence levels; the anomaly confidence level is negatively correlated with the maximum value of the corresponding concentration difference range; the anomaly confidence level corresponding to the concentration difference range in the first correspondence is taken as the anomaly confidence level of the anomaly type.

[0085] Specifically, the inlet particulate matter concentration refers to the concentration of particulate matter in the air detected by the particulate matter sensor installed at the filter inlet, while the outlet particulate matter concentration refers to the concentration of particulate matter in the air detected by the particulate matter sensor installed at the filter outlet. The difference between the inlet and outlet particulate matter concentrations is used as the concentration difference value to quantify the filter's ability to block particulate matter. When the filter is damaged, its blocking ability decreases, resulting in a smaller decrease in the relative decrease in outlet particulate matter concentration compared to inlet particulate matter concentration, i.e., a smaller concentration difference value.

[0086] The first correspondence can be understood as a pre-established data structure or rule set used to map concentration differences to anomaly confidence levels. This correspondence includes multiple concentration difference ranges, each corresponding one-to-one with a specific anomaly confidence level. The anomaly confidence level is negatively correlated with the maximum value of the corresponding concentration difference range, meaning that the smaller the concentration difference (i.e., the more severe the filter damage), the higher the anomaly confidence level. For example, when the concentration difference is in a small range, the corresponding anomaly confidence level will be higher, and conversely, when the concentration difference is in a large range, the corresponding anomaly confidence level will be lower.

[0087] In practical applications, the anomaly confidence level corresponding to the concentration difference range in the first correspondence is taken as the anomaly confidence level of the anomaly type. The purpose is to directly obtain the quantified anomaly confidence level based on the actual measured concentration difference by looking up a table or calculation, thereby providing a clear basis for subsequent operational strategies.

[0088] This application's solution addresses the potential quantification deficiencies and ambiguities in determining anomaly confidence levels in the aforementioned basic solution by introducing concentration differences and a first correspondence. Specifically, when the filter is damaged, its filtration efficiency for particulate matter decreases significantly, leading to a reduction in the concentration difference between the particulate matter entering the filter (inlet particulate matter concentration) and the particulate matter leaving the filter (outlet particulate matter concentration).

[0089] By calculating this concentration difference, the degree of filter damage can be intuitively reflected. Furthermore, through a pre-established first correspondence, different concentration difference ranges are associated with corresponding anomaly confidence levels, and anomaly confidence levels are set to be negatively correlated with the maximum value of the concentration difference range, so that the smaller the concentration difference, the higher the anomaly confidence level. Thus, this method can transform qualitative damage phenomena into quantitative anomaly confidence levels, thereby providing a more accurate and reliable anomaly assessment.

[0090] Through the above technical solution, this application provides a quantitative and structured method for determining the anomaly confidence level of filter damage anomalies. Compared to simply obtaining particulate matter concentration, this solution calculates the difference between the inlet and outlet particulate matter concentrations and combines this with a preset first correspondence to more accurately assess the degree of filter damage and convert it into a specific anomaly confidence level value. This not only improves the accuracy and reliability of anomaly warnings but also provides a clearer and more operable basis for subsequent maintenance operation strategies, avoiding misjudgments or delayed maintenance caused by subjective judgment or vague assessments, thereby enhancing the intelligence level and user experience of the robot vacuum cleaner filter anomaly warning system.

[0091] This application further proposes a method for determining a user's operation strategy based on the anomaly type of the filter and the anomaly confidence level of the anomaly type, which specifically includes: Determine whether the anomaly confidence level of the anomaly type is greater than the preset anomaly confidence level threshold; if so, obtain the second correspondence; the second correspondence includes multiple anomaly types and multiple operation strategies; use the operation strategy corresponding to the anomaly type of the filter in the second correspondence as the user's operation strategy; if not, store the anomaly type of the filter and multiple operation information to the cloud server.

[0092] Specifically, anomaly confidence refers to the degree of certainty or reliability with which the system identifies a filter anomaly type (such as dust penetration clogging or filter damage), usually expressed as a percentage or decimal. The preset anomaly confidence threshold is a pre-defined value used to differentiate the reliability of anomaly judgments; for example, it can be set to 70% or 0.7. When the anomaly confidence is higher than this threshold, it indicates that the system's judgment of the anomaly type has high reliability; conversely, it indicates that the reliability of the judgment is relatively low.

[0093] The second correspondence can be understood as a pre-established mapping table or rule set, the purpose of which is to associate specific filter anomaly types with corresponding user operation strategies. For example, when the anomaly type is "dust penetration clogging," the corresponding operation strategy could be "clean the filter"; when the anomaly type is "filter damage," the corresponding operation strategy could be "replace the filter." This correspondence can be stored on a local device or a cloud server and can be updated and adjusted as needed.

[0094] In practical applications, if the anomaly confidence level of an anomaly type does not reach the preset anomaly confidence threshold, it indicates that the reliability of the current anomaly judgment is insufficient to directly provide maintenance suggestions to the user. In this case, to avoid misleading the user, the system will store the currently identified filter anomaly type and multiple operational information pieces (including the operating current of the robot vacuum's fan, the air pressure difference before and after filtration by the robot vacuum's filter, etc.) on a cloud server. The purpose of this is to accumulate data for subsequent in-depth analysis, model optimization, or manual review by professionals, thereby improving the accuracy and intelligence level of the early warning system.

[0095] This application's solution achieves intelligent hierarchical processing of user operation strategies by introducing anomaly confidence level judgment. When the system has a high confidence level in judging the filter anomaly type, i.e., the anomaly confidence level is greater than the preset anomaly confidence level threshold, the system can directly and accurately provide the user with the corresponding maintenance operation strategy based on the preset second correspondence. This ensures that when the anomaly situation is clear, the user can receive effective guidance in a timely manner, thereby quickly resolving the problem. Conversely, when the anomaly confidence level is low, i.e., it has not reached the preset threshold, the system will not immediately send the operation strategy to the user, but will instead choose to store the anomaly type and operation information to the cloud server.

[0096] This approach avoids misleading users when there is uncertainty in the judgment, while providing the system with valuable data accumulation. This data can be used for subsequent machine learning model training, algorithm optimization, or manual analysis, thereby gradually improving the accuracy and robustness of the early warning system. It is precisely because of this confidence-based decision-making mechanism that the early warning system is more prudent and reliable in providing maintenance advice.

[0097] By employing the aforementioned technical solution, this application effectively avoids providing users with uncertain or erroneous maintenance suggestions when the confidence level of anomaly judgment is low, thereby significantly improving the accuracy of the early warning system and the user experience. This solution not only ensures that users receive timely and accurate maintenance guidance when anomalies are clear, but also provides a foundation for continuous learning and optimization of the system through a data storage mechanism when there is uncertainty in the judgment, contributing to the construction of a more intelligent, reliable, and user-friendly early warning system for robot vacuum cleaner filter anomalies.

[0098] This application further proposes a step for verifying the maintenance operation after the user performs maintenance on the filter and obtaining the verification result, including: The robot vacuum cleaner is controlled to start a preset self-test program to obtain multiple self-test data. When the similarity between the multiple self-test data and multiple preset normal self-test data is greater than a preset second similarity threshold, the verification result is determined to indicate that the user has successfully maintained the filter. Otherwise, the verification result is determined to indicate that the user has failed to maintain the filter.

[0099] Specifically, controlling a robotic vacuum cleaner to initiate a preset self-test program means that after the user completes maintenance operations on the filter (such as cleaning or replacing it), the system can send instructions to the robot to enter a specific operating mode. In this mode, the robot will perform a series of preset test tasks designed to simulate daily working conditions or specific test conditions to comprehensively evaluate the filter's performance. For example, the self-test program may include operating the fan at a specific wind speed, working in different suction modes, or performing short-term cleaning in specific environments.

[0100] As a result, the robotic vacuum cleaner generates multiple self-test data points. These data points consist of various sensor data and operating parameters collected by the robot during its self-test process, such as the fan's operating current, the air pressure difference before and after filtration, the particulate matter concentration at the air outlet, and the noise level in the air duct. This data reflects the actual working condition and performance of the filter after maintenance.

[0101] The multiple preset normal self-test data can be understood as baseline data collected and stored by executing the same preset self-test procedure when the filter is in a brand new or completely normal condition. These data represent the performance characteristics of the filter under ideal operating conditions and can serve as the gold standard for measuring maintenance effectiveness. In practical applications, when the similarity between multiple self-test data and multiple preset normal self-test data exceeds a preset second similarity threshold, the verification result indicates that the user has successfully maintained the filter.

[0102] Similarity can be calculated using various mathematical or statistical methods, such as Euclidean distance, cosine similarity, and Pearson correlation coefficient, to quantify the degree of matching between self-test data and normal baseline data. A preset second similarity threshold is a pre-defined value used to determine whether the maintenance operation achieved the expected results. If the similarity exceeds this threshold, it indicates that the filter's performance has recovered to near-normal levels, and the maintenance operation is considered successful. Conversely, if the similarity does not reach the threshold, the verification result indicates that the user's filter maintenance failed, meaning the maintenance operation did not effectively resolve the filter's abnormality.

[0103] This application's solution addresses the potential objectivity deficiencies and inefficiencies in the aforementioned maintenance operation verification by introducing an automated, pre-set self-test program and a data similarity comparison mechanism. Specifically, after the user completes filter maintenance, the robotic vacuum cleaner is controlled to initiate a pre-set self-test program. This program simulates the actual working state of the filter and generates a series of self-test data reflecting its current performance. This self-test data is then compared with multiple pre-stored pre-set normal self-test data representing the filter's normal state. By calculating the similarity between the two and comparing it to a pre-set second similarity threshold, the system can objectively and quantitatively evaluate the effectiveness of the maintenance operation. This data analysis-based verification method avoids the introduction of subjective judgment, ensuring the accuracy and reliability of the verification results.

[0104] Through the above technical solution, this application enables automated and objective verification of the maintenance operation of the robot vacuum cleaner's filter. This significantly improves the accuracy and reliability of the verification results, avoiding misjudgments caused by subjective judgment or insufficient inspection. Furthermore, this solution reduces the time and effort required by users to verify the maintenance effect, enhancing the user experience. When maintenance fails, the system can promptly identify and provide further error correction guidance, forming a closed-loop anomaly warning and maintenance support system, thereby effectively extending the filter's lifespan and ensuring the cleaning efficiency of the robot vacuum cleaner.

[0105] This application proposes an internet-based early warning system for an abnormal filter screen of a robotic vacuum cleaner, comprising: an acquisition device and a processing device; the acquisition device is used to acquire multiple operational information of the robotic vacuum cleaner; the multiple operational information includes the operating current of the robotic vacuum cleaner's fan and the air pressure difference before and after filtration by the robotic vacuum cleaner's filter screen; the processing device is used to determine the abnormality type of the filter screen and the abnormality confidence level of the abnormality type based on the multiple operational information; the abnormality type is dust penetration blockage or filter screen damage; the processing device is used to determine the user's operation strategy based on the abnormality type of the filter screen and the abnormality confidence level of the abnormality type; the operation strategy is used to instruct the user to perform maintenance operations on the filter screen; the processing device is used to verify the maintenance operation after the user performs the maintenance operation on the filter screen and obtain a verification result; the verification result is used to indicate whether the user's maintenance of the filter screen was successful or failed; the processing device is used to send error correction guidance information corresponding to the operation strategy to the user when the verification result indicates that the user's maintenance of the filter screen failed.

[0106] Specifically, the data acquisition device may include, but is not limited to: 1. Sensor module: For example, a current sensor integrated inside the robot vacuum cleaner to monitor the operating current of the fan in real time, and pressure sensors placed before and after the filter to measure the air pressure difference before and after filtration. These sensors can periodically collect data, for example, once per second, to ensure the real-time nature and continuity of the data. 2. Communication module: Used to upload the data collected by the sensor module to a cloud server or directly transmit it to the processing device via wireless communication (e.g., Wi-Fi, Bluetooth, or cellular network). 3. Data interface: Can be a physical interface or a software interface, used to receive various operational data from inside the robot vacuum cleaner.

[0107] As a preferred implementation, the acquisition device can be configured to be integrated inside the robot vacuum cleaner and directly connected to the fan and filter components to achieve local data collection and preliminary processing.

[0108] Specifically, the processing device can be an independent server cluster (cloud server) that receives data from multiple robotic vacuum cleaners for centralized processing; or, some processing functions can be deployed in the microcontroller or embedded system inside the robotic vacuum cleaner to achieve edge computing.

[0109] The above are merely embodiments of this application and are not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for early warning of filter abnormalities in a robotic vacuum cleaner based on the Internet, characterized in that, include: Obtain multiple operational information from the robotic vacuum cleaner; Several operational information items include the operating current of the robot vacuum's fan and the air pressure difference before and after filtration by the robot vacuum's filter. The anomaly type and confidence level of the filter were determined based on multiple operational information; the anomaly type was either dust penetration clogging or filter damage. The user's operation strategy is determined based on the anomaly type of the filter and the anomaly confidence level of the anomaly type; the operation strategy is used to instruct the user to perform maintenance operations on the filter. After the user performs maintenance operations on the filter, the maintenance operation is verified, and a verification result is obtained. The verification result is used to indicate whether the user's maintenance of the filter was successful or failed. When the verification result indicates that the filter maintenance has failed, the system sends the user error correction guidance information corresponding to the operation strategy.

2. The method for early warning of filter abnormalities in a robotic vacuum cleaner based on the Internet, as described in claim 1, is characterized in that... The anomaly types of the filter and the anomaly confidence levels of these types are determined based on multiple operational information, including: Determine whether the current difference of the fan's operating current within a preset time period is always positive, and whether the air pressure difference before and after filtration by the filter screen within a preset time period is always positive; the current difference is the difference between the operating current at one moment and the operating current at the previous moment, and the air pressure difference is the difference between the air pressure difference at one moment and the air pressure difference at the previous moment. If the difference in the operating current of the fan is positive within the preset time period, and the difference in the air pressure difference before and after filtration by the filter is positive within the preset time period, the abnormality type is determined to be dust penetration blockage; otherwise, the abnormality type is determined not to be dust penetration blockage. When the anomaly type is dust penetration blockage, the anomaly confidence level is determined based on the operating current of the fan of the filter and the air pressure difference before and after filtration.

3. The method for early warning of filter abnormalities in a robotic vacuum cleaner based on the Internet, as described in claim 2, is characterized in that... If the difference in the fan's operating current is positive throughout the preset time period, and the difference in air pressure before and after filtration is positive throughout the preset time period, the anomaly type is determined to be dust penetration blockage, including: If the current difference of the working current of the fan is positive within a preset time period, and the air pressure difference before and after filtration of the filter is positive within a preset time period, the concentration of particulate matter in the air detected by the particulate matter sensor set at the air outlet of the filter is obtained within a preset time period. If the absolute value of the particulate matter concentration difference within a preset time period is less than the first preset concentration difference threshold, the anomaly type is determined to be dust penetration blockage; otherwise, the anomaly type is determined not to be dust penetration blockage. The particulate matter concentration difference is the difference between the particulate matter concentration at one moment and the particulate matter concentration at the previous moment.

4. The method for early warning of filter abnormalities in a robotic vacuum cleaner based on the Internet, as described in claim 2, is characterized in that... When the anomaly type is dust penetration clogging, the anomaly confidence level is determined based on the operating current of the fan connected to the filter and the air pressure difference before and after filtration, including: For each operating current of the fan within a preset time period, obtain the normal air pressure difference of the operating current; the normal air pressure difference is the air pressure difference before and after filtration when the filter is in normal condition; Determine the similarity between multiple normal air pressure differences and multiple air pressure differences before and after filtration in the operation information, and use the reciprocal of the air pressure difference similarity as the anomaly confidence level of the anomaly type.

5. A method for early warning of filter abnormalities in a robotic vacuum cleaner based on the Internet, as described in claim 1, characterized in that, The anomaly types of the filter and the anomaly confidence levels of these types are determined based on multiple operational information, including: Determine whether the operating current of the fan is always less than the preset current value within a preset time period, and whether the air pressure difference before and after filtration by the filter is always negative within a preset time period; the air pressure difference is the difference between the air pressure difference at one moment and the air pressure difference at the previous moment. If the operating current of the fan is less than the preset current value within the preset time period, and the air pressure difference before and after filtration by the filter is negative within the preset time period, the abnormality type is determined to be filter damage; otherwise, the abnormality type is determined not to be filter damage. When the anomaly type is filter damage, obtain the concentration of particulate matter in the air at the outlet of the filter and the concentration of particulate matter in the air at the inlet of the filter. The anomaly confidence level of the anomaly type is determined based on the concentration of particulate matter at export and import.

6. A method for early warning of filter abnormalities in a robotic vacuum cleaner based on the Internet, as described in claim 5, is characterized in that... If the fan's operating current is consistently lower than the preset current value within a preset time period, and the air pressure difference before and after filtration is consistently negative within the preset time period, the anomaly is determined to be filter damage, including: If the operating current of the fan is less than the preset current value within the preset time period, and the air pressure difference before and after filtration of the filter is negative within the preset time period, acquire the detection audio data set on the air inlet side of the filter and the normal audio data on the air inlet side of the filter when the filter is normal. If the similarity between the detected audio data and normal audio data is less than a preset first similarity threshold, the anomaly type is determined to be a broken filter; otherwise, the anomaly type is determined not to be a broken filter.

7. A method for early warning of filter abnormalities in a robotic vacuum cleaner based on the Internet, as described in claim 5, is characterized in that... Anomaly confidence levels are determined based on the concentrations of exported and imported particulate matter, including: The difference between the concentration of inbound particulate matter and the concentration of outbound particulate matter is taken as the concentration difference. Obtain the first correspondence; the first correspondence includes a one-to-one correspondence between multiple concentration difference ranges and multiple anomaly confidence levels; the anomaly confidence level is negatively correlated with the maximum value of the corresponding concentration difference range; The anomaly confidence level corresponding to the concentration difference range in the first correspondence is taken as the anomaly confidence level of the anomaly type.

8. A method for early warning of filter abnormalities in a robotic vacuum cleaner based on the Internet, as described in claim 1, characterized in that, The user's operational strategy is determined based on the anomaly type of the filter and the anomaly confidence level of that type, including: Determine whether the anomaly confidence level of the anomaly type is greater than the preset anomaly confidence threshold; If so, obtain the second correspondence; the second correspondence includes multiple exception types and multiple operation strategies; The operation strategy corresponding to the anomaly type of the filter in the second correspondence is used as the user's operation strategy; If not, store the filter's anomaly type and multiple operational information to the cloud server.

9. A method for early warning of filter abnormalities in a robotic vacuum cleaner based on the Internet, as described in claim 1, characterized in that, After the user performs maintenance operations on the filter, the maintenance operations are verified, and the verification results are obtained, including: Control the robot vacuum cleaner to start a preset self-test program and obtain multiple self-test data of the robot vacuum cleaner; If the similarity between multiple self-test data and multiple preset normal self-test data is greater than a preset second similarity threshold, the verification result indicates that the user has successfully maintained the filter; otherwise, the verification result indicates that the user has failed to maintain the filter.

10. An internet-based early warning system for abnormal filter screens in a robotic vacuum cleaner, characterized in that: include: Acquisition device and processing device; Acquisition device, used to acquire multiple operational information of the sweeping robot; Several operational information items include the operating current of the robot vacuum's fan and the air pressure difference before and after filtration by the robot vacuum's filter. The processing device is used to determine the type of abnormality of the filter and the confidence level of the abnormality based on multiple operating information; the abnormality type is dust penetration blockage or filter damage. The processing device is used to determine the user's operation strategy based on the type of abnormality of the filter and the confidence level of the abnormality type; the operation strategy is used to instruct the user to perform maintenance operations on the filter. The processing device is used to verify the maintenance operation after the user performs maintenance on the filter screen and obtain the verification result; the verification result is used to indicate whether the user has successfully maintained the filter screen or failed to maintain the filter screen. The processing device is used to send error correction guidance information corresponding to the operation strategy to the user when the verification result indicates that the filter maintenance has failed.

Images