Equipment monitoring method and device, electronic equipment and storage medium

Abstract

The invention provides an equipment monitoring method and device, electronic equipment and a storage medium, and the method comprises the steps: determining a current state of a target scene corresponding to target equipment after the target equipment is controlled, and determining an expected state for the target scene; when the current state is not matched with the expected state, obtaining a target device weight of the target device, and determining a duration in which the current state is not matched with the expected state; determining distance information between the user and the target equipment; and determining deviation degree information of the target equipment according to the weight, the duration and the distance information of the target equipment, and monitoring the target equipment according to the deviation degree information. Through the embodiment of the invention, the accidental abnormal phenomenon and the continuous deviation state can be effectively distinguished on the basis of the multi-dimensional parameters, the running state of which equipment in the current scene control better meets the current actual demand can be intelligently judged, and more intelligent user experience can be brought to a user.

Description

Technical Field

[0001] This application belongs to the field of smart home technology, specifically relating to a device monitoring method, apparatus, electronic device, and storage medium. Background Technology

[0002] In related technologies, the status of devices during scene execution can be determined through static rules and conditions. However, this method can only detect and determine the operating status of related devices after the scene command is executed (e.g., if the indoor temperature is >27 degrees, the indoor air conditioning temperature will be lowered) through pre-set static rules and conditions. It cannot distinguish between occasional abnormal phenomena and continuous deviations, and it cannot intelligently determine which devices in the current scene control have an operating status that better meets the current actual needs, thus failing to bring a more intelligent user experience. Summary of the Invention

[0003] In view of the above problems, a monitoring method, apparatus, electronic device, and storage medium for a device are proposed to overcome or at least partially solve the above problems, including: A method for monitoring a device, the method comprising: After controlling the target device, determine the current state of the target device in the target scenario, and determine the desired state for the target scenario; When the current state does not match the expected state, obtain the target device weight of the target device and determine the duration of the mismatch between the current state and the expected state. Determine the distance information between the user and the target device; Based on the target device weight, the duration, and the distance information, the deviation information of the target device is determined, and the target device is monitored based on the deviation information.

[0004] In some embodiments, determining the desired state for the target scenario includes: Acquire multi-source sensing data, wherein the multi-source sensing data includes at least one of environmental data, user data, and external data; Based on the multi-source perception data, predict the user's target intent; The desired state is determined based on the stated target intent.

[0005] In some embodiments, predicting the user's target intent based on the multi-source perception data includes: Based on the multi-source sensing data, determine the user's explicit instructions or the user's implicit needs; Determine the user's target intent based on the explicit instructions or the user's implicit needs.

[0006] In some embodiments, obtaining the target device weight of the target device includes: Determine the basic weight information of the target device, and the scene weight information of the target device in the target scene; Determine the abnormal status information of the target device; The target device weight is determined based on the basic weight information, the scene weight information, and the abnormal state information.

[0007] In some embodiments, determining the deviation information of the target device based on the target device weight, the duration, and the distance information includes: Determine the time decay factor based on the duration. Based on the distance information, determine the spatial association factor; The deviation information is determined based on the target device weight, the time decay factor, the spatial correlation factor, and the degree of deviation between the current state and the desired state.

[0008] In some embodiments, determining the time decay factor based on the duration includes: Determine the target attenuation coefficient; The time decay factor is determined based on the target decay coefficient and the duration.

[0009] In some embodiments, monitoring the target device based on the deviation information includes: When the deviation information exceeds the deviation threshold, it is determined that the operation of the target device does not meet the user's needs; When the deviation information does not exceed the deviation threshold, it is determined that the operation of the target device meets the user's needs.

[0010] In some embodiments, after determining that the operation of the target device does not meet the user's needs, the method further includes: Obtain a scene association graph, which includes device nodes, physical space nodes, scene nodes, and influence relationship edges between each node; Query the scene association map to determine the global impact information of the adjustment to the target device; Based on the global impact information, control is applied to the target device and other devices.

[0011] In some embodiments, the method further includes: Before executing the critical instructions for the target device, determine whether to execute the critical instructions based on whether the target device exhibits any abnormalities; and / or, Before executing the critical instructions for the target device, a determination is made based on the collected environmental data as to whether to execute the critical instructions; and / or, Before executing the key instructions for the target device, the system determines whether to execute the key instructions based on the user's user data.

[0012] This application embodiment also provides a device for monitoring equipment, the device comprising: The state determination module is used to determine the current state of the target device corresponding to the target scenario after controlling the target device, and to determine the desired state for the target scenario. The weight duration determination module is used to obtain the target device weight of the target device and determine the duration of the mismatch between the current state and the expected state when the current state does not match the expected state. The distance determination module is used to determine the distance information between the user and the target device; The monitoring module is used to determine the deviation information of the target device based on the target device weight, the duration, and the distance information, and to monitor the target device based on the deviation information.

[0013] In some embodiments, the state determination module is configured to acquire multi-source sensing data, the multi-source sensing data including at least one of environmental data, user data, and external data; predict the user's target intent based on the multi-source sensing data; and determine the desired state based on the target intent.

[0014] In some embodiments, the state determination module is configured to determine the user's explicit instructions or the user's implicit needs based on the multi-source sensing data; and to determine the user's target intent based on the explicit instructions or the user's implicit needs.

[0015] In some embodiments, the weight duration determination module is used to determine the basic weight information of the target device and the scene weight information of the target device in the target scenario; determine the abnormal state information of the target device; and determine the weight of the target device based on the basic weight information, the scene weight information, and the abnormal state information.

[0016] In some embodiments, the monitoring module is configured to determine a time decay factor based on the duration; determine a spatial correlation factor based on the distance information; and determine the deviation information based on the target device weight, the time decay factor, the spatial correlation factor, and the degree of deviation between the current state and the desired state.

[0017] In some embodiments, the monitoring module is configured to determine a target attenuation coefficient and, based on the target attenuation coefficient and the duration, determine the time attenuation factor.

[0018] In some embodiments, the monitoring module is configured to determine that the operation of the target device does not meet the user's needs when the deviation information exceeds the deviation threshold; and to determine that the operation of the target device meets the user's needs when the deviation information does not exceed the deviation threshold.

[0019] In some embodiments, the monitoring module is further configured to, after determining that the operation of the target device does not meet the user's needs, obtain a scene association graph, the scene association graph including device nodes, physical space nodes, scene nodes, and influence relationship edges between each node; query the scene association graph to determine the global impact information generated by the adjustment to the target device; and control the target device and other devices according to the global impact information.

[0020] In some embodiments, the monitoring module is further configured to determine whether to execute the key instruction based on whether the target device has an anomaly before executing the key instruction for the target device; and / or, determine whether to execute the key instruction based on collected environmental data before executing the key instruction for the target device; and / or, determine whether to execute the key instruction based on the user's user data before executing the key instruction for the target device.

[0021] This application also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and capable of running on the processor. When the computer program is executed by the processor, it implements the device monitoring method described above.

[0022] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the device monitoring method described above.

[0023] The embodiments of this application have the following advantages: In this embodiment, after controlling the target device, the current state of the target device in the target scenario is determined, as well as the desired state for the target scenario. When the current state does not match the desired state, the target device weight is obtained, and the duration of the mismatch is determined. The distance information between the user and the target device is determined. Based on the target device weight, duration, and distance information, the deviation information of the target device is determined, and the target device is monitored based on the deviation information. Through this embodiment, based on multi-dimensional parameters, it is possible to effectively distinguish between occasional abnormal phenomena and continuous deviations, and to intelligently determine which devices in the current scenario control are more in line with the current actual needs, thus providing users with a more intelligent user experience. Attached Figure Description

[0024] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is a flowchart illustrating the steps of a device monitoring method according to an embodiment of this application; Figure 2 This is a flowchart illustrating the steps of another device monitoring method according to an embodiment of this application; Figure 3 This is a schematic diagram of the structure of a smart home cross-scene collaborative decision-making system based on multi-source sensing and dynamic attenuation factor according to an embodiment of this application; Figure 4 This is a schematic diagram of the structure of a monitoring device for an embodiment of this application. Detailed Implementation

[0025] To make the above-mentioned objectives, features, and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only some, not all, of the 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.

[0026] To more intelligently identify and determine whether devices are operating normally after scene execution, this application proposes a device monitoring method. This method uses multi-dimensional parameters to determine the degree of deviation between the triggered scene and the actual execution result, thereby identifying whether the device is operating normally after scene execution. Based on multi-dimensional parameters, this application can effectively distinguish between occasional anomalies and continuous deviations. It can also intelligently determine which devices in the current scene control are operating in a way that better meets the current actual needs, providing users with a more intelligent user experience.

[0027] Reference Figure 1The diagram illustrates a flowchart of a device monitoring method according to an embodiment of this application, which may include the following steps: Step 101: After controlling the target device, determine the current state of the target device in the target scenario, and determine the desired state for the target scenario.

[0028] In some embodiments, when determining to construct a target scene, instructions for constructing the target scene can be generated first; these instructions may include control instructions for different devices, and may also include the conditions of the target scene to be constructed, such as setting the temperature, setting the humidity, etc.

[0029] Upon receiving the instruction, the corresponding target device can be controlled in response. After controlling the target device, it is possible to determine whether the operating state of the target device under the target scene has deviated from the target scene based on the current state of the target device in the target scene and the expected state set for the target scene; that is, whether the target device can still maintain the target scene.

[0030] Specifically, the current state of the target scene constructed by the target device and the expected state set for the target scene can be determined separately. The current state may include the current environmental state, the current device state, the current user state, etc. The expected state may include the expected environmental state, the expected device state, the expected user state, etc. of the target scene. The expected state can be preset or predicted and expected by the user. This application embodiment does not limit this.

[0031] Step 102: When the current state does not match the expected state, obtain the target device weight of the target device and determine the duration of the mismatch between the current state and the expected state.

[0032] After determining the current state and the expected state, it is possible to judge whether the operating state of the target device in the target scenario has deviated based on the current state and the expected state. For example, if the current state and the expected state do not match, it can be determined that the operating state of the target device has deviated and may not be able to maintain the target scenario.

[0033] Conversely, if the current state matches the desired state, it can be determined that the target device's operating state has not deviated and the target scenario can continue to be maintained; this matching process can be carried out by calculating the similarity between the current state and the desired state.

[0034] If the current state matches the expected state, step 101 can be executed; otherwise, if the current state does not match the expected state, further judgment can be made to distinguish between occasional abnormal phenomena and continuous deviations.

[0035] Specifically, you can first obtain the target device weight set for the target device; the target device weight can refer to the value assigned based on the importance or criticality of the device.

[0036] Additionally, the duration of the mismatch between the current state and the desired state can be determined. Specifically, after determining the current state at a preset time interval, the current state at the current moment is matched and calculated with the desired state. If the current state at the current moment does not match the desired state, the current moment can be recorded as the start time of the duration.

[0037] When the next time period arrives, if the current state determined at that moment still does not match the expected state, the duration can be counted, and so on.

[0038] Step 103: Determine the distance information between the user and the target device.

[0039] In some embodiments, the reasonableness of the device execution state offset is also determined by locating the distance between the user and the device; specifically, the distance information between the user and the target device can be obtained; the distance information can refer to the numerical value of the distance between the user and the target device, or it can refer to whether the user and the target device are in the same area in space.

[0040] Step 104: Determine the deviation information of the target device based on the target device weight, duration, and distance information, and monitor the target device based on the deviation information.

[0041] After determining the target device weight, duration, and distance information, the deviation information of the target device can be determined based on the target device weight, duration, and distance information. This deviation information can be represented by a numerical value. The larger the value, the further the target device's operating state deviates from the expected state, and the more difficult it is for the target device to maintain the target scene if it continues to operate. The smaller the value, the less the target device's operating state deviates from the expected state, and the more likely the target device can maintain the target scene if it continues to operate.

[0042] After determining the deviation information, the target device can be monitored to identify whether the deviation is sporadic or continuous. Based on the monitoring results, the control module can control or adjust the target device to maintain the continuity of the target scenario.

[0043] In this embodiment, after controlling the target device, the current state of the target device in the target scenario is determined, as well as the desired state for the target scenario. When the current state does not match the desired state, the target device weight is obtained, and the duration of the mismatch is determined. The distance information between the user and the target device is determined. Based on the target device weight, duration, and distance information, the deviation information of the target device is determined, and the target device is monitored based on the deviation information. Through this embodiment, based on multi-dimensional parameters, it is possible to effectively distinguish between occasional abnormal phenomena and continuous deviations, and to intelligently determine which devices in the current scenario control are more in line with the current actual needs, thus providing users with a more intelligent user experience.

[0044] Reference Figure 2 The diagram illustrates a flowchart of another monitoring method for a device according to an embodiment of this application, which may include the following steps: Step 201: After controlling the target device, determine the current state of the target scene corresponding to the target device.

[0045] In some embodiments, when determining to construct a target scene, instructions for constructing the target scene can be generated first; these instructions may include control instructions for different devices, and may also include the conditions of the target scene to be constructed, such as setting the temperature, setting the humidity, etc.

[0046] Upon receiving the instruction, the corresponding target device can be controlled in response. After controlling the target device, it is possible to determine whether the operating state of the target device under the target scene has deviated from the target scene based on the current state of the target device in the target scene and the expected state set for the target scene; that is, whether the target device can still maintain the target scene.

[0047] Specifically, data can be collected through various sensors and actuators to obtain the current state of the target device in the target scenario, such as the current environmental state, the current device state, and the current user state.

[0048] Step 202: Obtain multi-source sensing data, which includes at least one of environmental data, user data, and external data.

[0049] In some embodiments, the desired state of the target scene can also be obtained; specifically, multi-source sensing data can be obtained first; wherein, multi-source sensing data may include at least one of environmental data, user data, and external data.

[0050] Environmental data can be obtained through sensors deployed in the environment, such as temperature sensors for temperature data and humidity sensors for humidity data.

[0051] User data can be obtained by collecting user behavior data, such as obtaining user location and behavior data from an app; it can also be obtained from sensors deployed in the environment, such as obtaining human infrared data from a human infrared sensor.

[0052] External data may include data obtained from external sources, such as outdoor temperature, humidity, and air quality forecasts for the next few hours obtained through a weather API. This application does not limit this.

[0053] Step 203: Based on multi-source perception data, predict the user's target intent.

[0054] After obtaining multi-source perception data, the data from multiple sources can be combined to accurately predict the user's target intent. Specifically, the multi-source perception data can be input into a trained user behavior prediction model to predict the user's target intent in the future. Target intent can include the user's vague needs or potential behavioral goals (e.g., "getting ready to sleep", "leaving home", "the study will be used by someone").

[0055] In some embodiments of this application, the user's target intent can be predicted through the following sub-steps: Sub-step 11: Based on multi-source sensing data, determine the user's explicit instructions or implicit needs.

[0056] In some embodiments, a trained user behavior prediction model can be used to analyze multi-source perception data in order to parse out the user's explicit instructions or implicit needs.

[0057] Explicit instructions can refer to intuitive user behaviors derived from multi-source sensor data analysis; implicit needs can refer to potential user needs derived from explicit instructions analysis. For example: When a user is detected moving toward the study, the user behavior prediction model can analyze the data from multiple sources to determine that the explicit instruction is "moving toward the study" and the implicit need is "the user is preparing to use the study".

[0058] Sub-step 12: Determine the user's target intent based on explicit instructions or the user's implicit needs.

[0059] After identifying explicit instructions or implicit user needs, analysis can be conducted to determine the user's target intent.

[0060] For example, explicit instructions can be analyzed to determine the user's target intent. Specifically, explicit instructions can be analyzed to identify possible intents associated with the behavior, thereby determining the target intent.

[0061] In another example, implicit requirements can also be analyzed directly to determine the target intent corresponding to the requirements.

[0062] Step 204: Determine the desired state based on the target intent.

[0063] Once the target intent is determined, the ideal state of the target scenario that needs to be constructed to satisfy the target intent can be determined, i.e., the expected state.

[0064] Step 205: When the current state does not match the expected state, obtain the target device weight of the target device and determine the duration of the mismatch between the current state and the expected state.

[0065] After determining the current state and the expected state, it is possible to judge whether the operating state of the target device in the target scenario has deviated based on the current state and the expected state. For example, if the current state and the expected state do not match, it can be determined that the operating state of the target device has deviated and may not be able to maintain the target scenario.

[0066] Conversely, if the current state matches the desired state, it can be determined that the target device's operating state has not deviated and the target scenario can continue to be maintained; this matching process can be carried out by calculating the similarity between the current state and the desired state.

[0067] If the current state matches the expected state, step 201 can be executed; otherwise, if the current state does not match the expected state, further judgment can be made to distinguish between occasional abnormal phenomena and continuous deviations.

[0068] Specifically, you can first obtain the target device weight set for the target device; the target device weight can refer to the value assigned based on the importance or criticality of the device.

[0069] Additionally, the duration of the mismatch between the current state and the desired state can be determined. Specifically, after determining the current state at a preset time interval, the current state at the current moment is matched and calculated with the desired state. If the current state at the current moment does not match the desired state, the current moment can be recorded as the start time of the duration.

[0070] When the next time period arrives, if the current state determined at that moment still does not match the expected state, the duration can be counted, and so on.

[0071] In some embodiments of this application, the target device weight can be obtained through the following sub-steps: Sub-step 21: Determine the basic weight information of the target device, and the scene weight information of the target device in the target scenario.

[0072] In some embodiments, different devices may be set with different basic weight information; and the same device may also be set with different scene weight information in different scenarios.

[0073] Sub-step 22: Determine the abnormal status information of the target device.

[0074] In this embodiment, the target device weight can also be calculated by integrating the abnormal state information of the target device. This abnormal state information can be used to indicate an abnormality occurring in the target device.

[0075] Sub-step 23: Determine the target device weight based on the basic weight information, scene weight information, and abnormal state information.

[0076] After determining the basic weight information, scene weight information, and abnormal state information, the target device weight adapted to the current target device can be calculated by combining the basic weight information, scene weight information, and abnormal state information.

[0077] Step 206: Determine the distance information between the user and the target device.

[0078] In some embodiments, the reasonableness of the device execution state offset is also determined by locating the distance between the user and the device; specifically, the distance information between the user and the target device can be obtained; the distance information can refer to the numerical value of the distance between the user and the target device, or it can refer to whether the user and the target device are in the same area in space.

[0079] Step 207: Determine the deviation information of the target device based on the target device weight, duration, and distance information, and monitor the target device based on the deviation information.

[0080] After determining the target device weight, duration, and distance information, the deviation information of the target device can be determined based on the target device weight, duration, and distance information. This deviation information can be represented by a numerical value. The larger the value, the further the target device's operating state deviates from the expected state, and the more difficult it is for the target device to maintain the target scene if it continues to operate. The smaller the value, the less the target device's operating state deviates from the expected state, and the more likely the target device can maintain the target scene if it continues to operate.

[0081] After determining the deviation information, the target device can be monitored to identify whether the deviation is sporadic or continuous. Based on the monitoring results, the control module can control or adjust the target device to maintain the continuity of the target scenario.

[0082] In some embodiments of this application, the deviation information can be determined through the following sub-steps: Sub-step 31: Determine the time decay factor based on the duration.

[0083] In some embodiments, the corresponding time decay factor can be determined first based on the duration.

[0084] In some embodiments of this application, the time decay factor can be determined in the following manner: Determine the target attenuation coefficient; based on the target attenuation coefficient and duration, determine the time attenuation factor.

[0085] In some embodiments, different target attenuation coefficients can be set for different deviations to calculate the deviation degree based on the scenario; for example, for security-related anomalies that require a rapid response, the target attenuation coefficient should be a small value (such as 0.1) to slow down the attenuation, so that even if the anomaly occurs for several minutes, the deviation degree information remains at a high level, triggering an immediate alarm; for comfort-related anomalies, a larger target attenuation coefficient (such as 0.5) can be set to allow the system to fluctuate briefly, avoid excessive intervention, and improve the smoothness of the experience.

[0086] After determining the target attenuation coefficient, the time attenuation factor can be calculated based on the target attenuation coefficient and the duration.

[0087] Sub-step 32: Determine the spatial correlation factor based on the distance information.

[0088] After determining the distance information, a spatial correlation factor can be determined based on the distance information; this spatial correlation factor can be used to judge the rationality of the device state offset.

[0089] Sub-step 33: Determine the deviation information based on the target device weight, time decay factor, spatial correlation factor, and the degree of deviation between the current state and the desired state.

[0090] After determining the target device weight, time decay factor, and spatial correlation factor, the degree of deviation between the current state and the desired state can also be determined. For example, both the current state and the desired state can be converted into numerical values, and then the numerical difference can be calculated to obtain the degree of deviation.

[0091] After determining the target device weight, time decay factor, spatial correlation factor, and degree of deviation, the deviation information can be calculated based on the target device weight, time decay factor, spatial correlation factor, and degree of deviation.

[0092] In some embodiments of this application, the target device can be monitored through the following sub-steps: Sub-step 41: When the deviation information exceeds the deviation threshold, it is determined that the operation of the target device does not meet the user's needs.

[0093] In some embodiments, a deviation threshold can be set; when the deviation information exceeds the deviation threshold, it can be determined that the operation of the target device cannot meet the user's needs; that is, the deviation of the target device is continuous.

[0094] Sub-step 42: When the deviation information does not exceed the deviation threshold, it is determined that the operation of the target device meets the user's needs.

[0095] Conversely, if the deviation information does not exceed the deviation threshold, it can be determined that the operation of the target device can still meet the user's needs; that is, the deviation of the target device is occasional.

[0096] In some embodiments, deviation information of all target devices controlled in a target scenario can be determined, and then, based on all the deviation information, the total deviation information of the target scenario can be determined. Based on this total deviation information, all target devices can be monitored.

[0097] Specifically, when the total deviation exceeds the deviation threshold, it can be determined that the target device's operation cannot maintain the target scenario. At this time, control measures can be implemented on the target device to adjust the operation of each target device to maintain the target scenario and prevent prolonged deviation of the target device from affecting the maintenance of the target scenario.

[0098] Conversely, if the total deviation information does not exceed the deviation threshold, it can be determined that the operation of all target devices can still maintain the target scenario.

[0099] In some embodiments of this application, after determining that the operation of the target device does not meet the user's needs, the above method may further include the following steps: Obtain the scene association graph, which includes device nodes, physical space nodes, scene nodes, and the influence relationship edges between each node; query the scene association graph to determine the global impact information of the adjustment to the target device; and control the target device and other devices based on the global impact information.

[0100] In some embodiments, after determining that the operation of the target device does not meet the user's needs, the target device needs to be adjusted to correct the target device; specifically, since there may be influence between different decisions, for example, "turning on the study air conditioner" may conflict with "living room energy-saving mode".

[0101] Based on this, this application can first obtain a scene association graph, which can include device nodes, physical space nodes, scene nodes, and the influence relationship edges between each node. The scene association graph is a structured and visualized model that systematically depicts the intricate relationships between different scenes, devices, users, and environmental states within a smart home system. It can be understood as a map of the "brain's neural network" of a smart home, with "nodes" (such as various smart devices, physical spaces, and specific scene modes) scattered across the map. Relationship edges connect these nodes, representing their interactions, influences, or logical sequences. For example, there might be a "reduce" relationship edge between the "living room air conditioner" node and the "living room temperature" node; while the "sleep mode" scene node would be associated with a series of action nodes such as "turn off the main light" and "lower the air conditioner temperature." The scene association graph enables cross-scene collaboration, avoids control and command conflicts, and optimizes decisions based on user habits.

[0102] Specifically, after obtaining the scene association map, the scene association map can be queried based on the target device and the target scene to determine the potential impact of adjustments made to correct the target device, i.e., global impact information. This global impact information can include the chain reactions that may occur after adjusting the target device.

[0103] After determining the global impact information, the target device and other devices can be controlled based on the global impact information to avoid affecting the execution of other decisions due to adjustments to the target device.

[0104] In some embodiments of this application, the above method may further include the following steps: Before executing critical instructions for the target device, determine whether to execute the critical instructions based on whether there is an anomaly in the target device; and / or, before executing critical instructions for the target device, determine whether to execute the critical instructions based on the collected environmental data; and / or, before executing critical instructions for the target device, determine whether to execute the critical instructions based on the user's user data.

[0105] In some embodiments, a security verification can be performed before executing any critical instructions, such as those related to security, energy, or elderly safety. Specifically, before executing critical instructions targeting a target device, the system can detect whether the target device is malfunctioning. If the target device is malfunctioning, the critical instruction is executed; otherwise, if the target device is malfunctioning, the critical instruction is not executed.

[0106] In other embodiments, before executing the key instruction for the target device, environmental data of the environment in which the target device is located can be collected first, and the key instruction can be determined based on the environmental data. For example, if the key instruction is an instruction to alarm after an elderly person falls, the environmental data can be used to analyze whether the elderly person has fallen. If the analysis confirms that the elderly person has indeed fallen, it can be determined that the key instruction needs to be executed.

[0107] In some embodiments, user data may be acquired before executing a key instruction for the target device. For example, if the key instruction is an alarm command for an elderly person to fall, the user data may include physiological data (e.g., blood pressure, heart rate) generated by the user's wearable smart device (e.g., a smart bracelet). After obtaining this user data, it can be compared with the user's historical user data to determine whether the key instruction needs to be executed. If the user data matches the user's historical user data, it can be determined that the key instruction does not need to be executed; conversely, if the user data does not match the user's historical user data, it can be determined that the key instruction needs to be executed.

[0108] It should be noted that the above three methods for determining whether a critical instruction needs to be executed can be executed individually or in combination. When multiple methods are executed, they can be executed simultaneously, and the final determination of whether the critical instruction needs to be executed can be based on the results of their respective methods.

[0109] When multiple decisions are executed, they can also be executed sequentially according to a set order. The next decision process will only proceed when the previous decision does not require the execution of a key instruction. This application does not impose any restrictions on this.

[0110] In this embodiment, after controlling the target device, the current state of the target scene corresponding to the target device is determined; multi-source sensing data is acquired, including at least one of environmental data, user data, and external data; based on the multi-source sensing data, the user's target intent is predicted; based on the target intent, the desired state is determined; when the current state does not match the desired state, the target device weight is acquired, and the duration of the mismatch is determined; the distance information between the user and the target device is determined; based on the target device weight, duration, and distance information, the deviation information of the target device is determined, and the target device is monitored based on the deviation information. Through this embodiment, based on multi-dimensional parameters, it is possible to effectively distinguish between occasional abnormal phenomena and continuous deviation states, and to intelligently determine which devices' operating states in the current scene control are more in line with the current actual needs, thus providing users with a more intelligent user experience.

[0111] The monitoring methods for the above-mentioned equipment are further explained below with specific examples: Reference Figure 3 This paper illustrates an embodiment of a smart home cross-scenario collaborative decision-making system based on multi-source sensing and dynamic attenuation factor, including cloud, terminal and edge devices.

[0112] The cloud can be used for data training and as a big data center; specifically, it can acquire data from weather APIs, historical user data, and data from third-party services. The cloud can train LSTM user behavior prediction models based on this data and build an environmental adaptation strategy library and scene association graph; the environmental adaptation strategy library can be used to provide decision-making to adjust devices.

[0113] The terminal can serve as the perception and execution layer; specifically, the terminal can be equipped with a sensor network to detect data such as temperature, light, and human body, and can also include an actuator network to control lights, air conditioning, door locks, etc.; it can also include a user interface that can interact with users through apps, voice assistants, etc.

[0114] The edge device can be a home smart gateway, equipped with a real-time behavior inference engine for analyzing user behavior; it can also have a cross-scenario linkage correction engine to correct device behavior based on an environmental adaptability strategy library and scene association graph. Furthermore, it can have a multi-dimensional deviation calculation model to calculate deviation information. Finally, it can have a three-layer distributed evidence chain verification function to verify whether critical instructions need to be executed.

[0115] The cloud can update models, policy libraries, and graphs and then send them to the edge for use. The terminal can acquire data and send it to the edge for recognition and processing; the edge can generate control commands and scene trigger commands and send them to the terminal for execution.

[0116] This system can monitor and manage equipment through the following steps: S01: Multi-source data sensing and acquisition: The system initiates a comprehensive perception of the home environment and user status. Various sensors (such as temperature, humidity, light, and human infrared sensors) and actuators (such as smart air conditioners and lights) continuously collect data.

[0117] At the same time, the system integrates external data sources, such as obtaining outdoor temperature, humidity, and air quality forecasts for the next few hours through a weather API.

[0118] In the cloud, a user behavior prediction model (LSTM network) calls upon historical user behavior data (such as the probability that a user was reading in their study between 8 pm and 10 pm on weekdays in the past few weeks) to form a prediction of user behavior in the near future. All this real-time data, predictive data, and device status data are aggregated at the edge to provide raw materials for subsequent decision-making.

[0119] S02: Intent Understanding and Scene Recognition The real-time behavior inference engine in the edge gateway begins operation. Based on multi-source sensing data collected by the S01, it uses a user behavior prediction model (LSTM network) to analyze the user's explicit instructions or implicit needs. For example, when the system detects that a user is moving towards the study and, combined with the LSTM model, predicts that the user is about to enter the study, it identifies this scenario as the user's intention to "use the study." Simultaneously, the system refers to an environmental adaptation model to determine the current season (summer) and outdoor weather (hot), injecting environmental context into subsequent decisions.

[0120] The role of "user intent" is to transform a user's vague needs or potential behavioral goals (e.g., "getting ready to sleep", "leaving home", "the study will be used by someone") into explicit task goals that the system can understand and process.

[0121] This "intent" will be directly input into stages S03 and S04. In S03, the intent is used to determine the "ideal state." For example, when the intent is "the user is about to use the study," the system will set the comfort parameters of the study environment (such as temperature 24℃, humidity 50%, and lights on) to the ideal state to compare with the "current state" and calculate the deviation. In S04, the intent is the core basis for cross-scene collaborative decision-making. The system will query the "scene association graph," analyze the potential impact of implementing this intent on other scenes (such as whether turning on the study's air conditioner will affect the quiet environment of the bedroom), and generate a globally optimal collaborative control sequence, rather than controlling individual devices in isolation.

[0122] S03: Multi-dimensional Deviation Calculation and Evaluation: The system invokes a multi-dimensional deviation calculation model to quantitatively evaluate the current state ("current state" refers to the real-time data collected by the system on the home environment and equipment; it is a multi-dimensional vector, specifically including: equipment status (e.g., whether lights are on / off), environmental status (e.g., real-time room temperature), and user status (e.g., user behavior). The current state is one of the benchmarks for calculating "device deviation" information. The system compares it with the "ideal state" defined by the user's "intent," and the difference (i.e., "device status deviation") is the direct cause triggering system intervention). This model calculates a comprehensive "device deviation" information.

[0123] Calculation formula: Deviation = Σ ([target device weight × | current state]) i -Ideal state i |]×Time decay factor×Spatial correlation factor). in: Device status weight: can be dynamically adjusted according to importance, such as security camera weight > lighting (priority of privacy and security), and can be dynamically adjusted through real-time failure rate.

[0124] The essence of device status weight is a fixed value assigned based on the importance or criticality of the device (e.g., smart lock weight 0.9 > lighting weight 0.3), which is initially set during device deployment. For example, although the inherent weight of a lighting fixture is not high, in the specific state of "the user has left home," its abnormal "always-on" state significantly increases its actual importance in this decision due to energy waste and security risks. Therefore, when calculating the deviation, the system comprehensively considers its inherent weight and the severity of the abnormal state, causing the deviation score to accumulate as the duration of the abnormal state increases, ultimately triggering an automatic light-off action after 1 hour.

[0125] "Immediate failure rate" is the probability of a failure occurring near the current moment, predicted based on historical operating data of the equipment (such as continuous operating duration, load cycles, etc.). It serves as an early warning indicator reflecting the health of the equipment. In deviation calculations, this parameter acts as an adjustment factor. If a device has a high "immediate failure rate," the system will increase its weight or decrease its tolerance threshold for state deviation. In this way, the system becomes more sensitive to abnormal states of high-risk equipment, thereby enabling preventative maintenance and priority handling.

[0126] The target device weight W_device can be calculated using the following formula: W_device = W × (1 + α × I + β × C) W (Basic Weight Information): Preset when the device is added to the database, based on the device's type attributes (safety, comfort, energy consumption) and the importance level manually marked by the user.

[0127] I (Abnormal Status Information): Value is 1 when the device is in an abnormal state, and 0 otherwise. Parameter α is the abnormal amplification factor, used to significantly increase the weight of abnormal devices in this calculation.

[0128] C (Scene Weight Information): This coefficient is determined by the current system scenario (e.g., security, sleep, entertainment). Parameter β is the scenario adjustment coefficient. It reflects the difference in importance of the same device in different scenarios.

[0129] Time decay factor: F_time = e (-k·t) (t is the duration, k is the target attenuation coefficient) The longer the deviation of the device's operating state from its actual operating state lasts, the lower its reliability. "e" is a mathematical constant, approximately 2.71828. "k" is an adjustable parameter greater than 0 that determines the attenuation rate. The smaller the value of "k," the slower the attenuation, and historical anomalies will have a more lasting impact on current decisions. The value of "k" is not fixed and needs to be optimized based on the specific business scenario and anomaly type.

[0130] Spatial correlation factor: By determining the distance between the user and the device, the reasonableness of the device's execution state deviation is assessed. For example, if the user is 5 kilometers away from home but the air conditioner is on, the spatial correlation factor = 0.1. The distance between the user and the device is inversely proportional to the value of this factor. That is, the greater the distance, the smaller the factor value, thus reducing the contribution of the device's abnormal state to the overall deviation. For example, when the user and the device are in the same room (d=0), the abnormality is fully considered (because the abnormality may cause a decline in user experience); when the user is in an adjacent room (d=1), its importance is halved. This calculation method does not require precise distance measurement and is easy to implement in a home environment.

[0131] For example, if the system detects that the air conditioner in the study is off, while the ideal comfortable temperature is 24°C, this deviation will be quantified according to a formula. Simultaneously, the system will assess the duration of this anomaly (time decay factor) and the distance between the user and the study (spatial decay factor). If the final calculated total deviation score (i.e., the aforementioned total deviation information) exceeds a preset deviation threshold, the system's collaborative decision-making mechanism will be triggered.

[0132] S04: Cross-scenario collaborative decision generation: When the deviation exceeds the threshold, the cross-scene linkage correction engine will be activated. This engine will query the scene correlation graph and analyze the potential chain reactions that the decision may produce.

[0133] For example, the decision to "turn on the air conditioner in the study" might conflict with "energy-saving mode in the living room." The engine will perform global optimization to generate a coordinated decision sequence, such as: "First, close the study windows to reduce cold air loss, then set the air conditioner to 26℃ (balancing comfort and energy saving), while slightly reducing the power of the study humidifier to balance the potential increase in humidity due to cooling." This process ensures that optimization of a single scenario does not negatively impact other scenarios.

[0134] S05: Distributed chain of evidence security verification: Before executing any critical instructions (such as those related to security, energy, or elderly safety), the system initiates a three-layer distributed evidence chain for security verification. Specifically, as shown in Table 1, the arbitration layers can include device status analysis at the first layer, environmental semantic analysis at the second layer, and user policy matching at the third layer. For the first layer, the input data is device operating data (e.g., device energy consumption, device operating status, device fault codes, etc.), and the output policy can be to mark device hardware anomalies (e.g., device wireless module expiration time, etc.).

[0135] For the second layer, the input data can be multi-sensor fusion data (e.g., data output from temperature and humidity sensors, data output from human infrared sensors, data output from door and window magnetic sensors, etc.); the output strategy can be to identify sudden environmental factors (e.g., failure to close windows in a rainstorm).

[0136] For the third layer, the input data can be user correction records and user activity status recognition, and the output strategy can be to generate personalized suggestions (e.g., turn on all the lights in the house when an elderly person falls).

[0137] Table 1:

[0138] Take "elderly person falling and calling the police" as an example: Equipment level: Check whether the millimeter-wave radar sensor detects a sudden change in posture due to a fall, and whether the equipment itself is working properly with no fault codes.

[0139] Environmental layer: Analyze camera data (while respecting privacy) to confirm the posture of personnel, and analyze sound sensor data to determine if there are cries for help.

[0140] At the user level: Real-time physiological data such as heart rate and blood pressure from wearable devices (e.g., smart bracelets) are referenced and compared with the user's historical behavior to rule out false alarms. Only when evidence from at least two levels supports a "fall" assessment will the system trigger a high-risk alarm and notify family members or emergency contacts; otherwise, it may only be marked as an "abnormal situation" for further observation. This significantly reduces the false alarm rate.

[0141] S06: Instruction Execution and Scene Activation After security verification, the decision instructions are sent to the execution control layer. The home central controller uses a unified protocol conversion engine (supporting different protocols such as Matter and Zigbee) to convert high-level instructions into low-level control commands that can be understood and executed by various brands and types of smart devices. Subsequently, these instructions are sent to the corresponding actuators (such as air conditioner, light, and curtain motors) in an orderly manner. The transaction coordinator ensures that operations involving multiple devices are atomic, meaning that either all operations execute successfully, or all operations are rolled back in the event of a failure, maintaining the consistency of the system state.

[0142] S07: Effectiveness Monitoring and Feedback Learning Once the scene is activated, the system's feedback learning layer begins operation. Sensors continuously monitor changes in environmental parameters (such as whether the temperature has stabilized and dropped to 26°C) and whether the user provides clear feedback (such as a voice command saying "It's too cold" or clicking "Dissatisfied" directly on the app). These execution results and user feedback data are recorded and transmitted back to the cloud. The cloud uses this data to retrain the LSTM prediction model, optimizing the accuracy of user behavior predictions; simultaneously, it updates the parameters in the environmental adaptation strategy library and scene association graph, making the system's future decisions more accurate and personalized. At this point, a complete "perception-decision-execution-learning" closed loop is completed, and the system continues operating, entering the next cycle.

[0143] This application calculates the deviation between the scene triggering conditions and the actual execution results from multiple dimensions, including the device weight of the scene running device, the dynamic decay factor of the scene execution time, and the spatial correlation factor between the user and the device. This allows for a more intelligent identification and judgment of whether the device's operating status after scene execution meets the scene requirements. By calculating the deviation between the scene triggering conditions and the actual execution results using multi-dimensional parameters, it more intelligently identifies and judges whether the device is operating normally after scene execution.

[0144] Furthermore, when the operating status of devices in a user-executed scenario deviates from its intended state, a distributed chain of evidence is generated for verification at the device layer (device energy consumption, device fault codes, etc.), the environment layer (multi-sensor fusion data within the environment), and the user layer (user historical correction records, user status recognition, etc.). Simultaneously, a specially trained LSTM network can be integrated to predict user behavior tendencies in specific home scenarios, thereby achieving multi-source cross-decision evaluation and overcoming the limitations of single-layer decision logic that directly generates instructions based on a preset rule base. After identifying a deviation between the user's execution scenario and the actual operating status of the devices, the system can quickly determine the cause of the problem through multi-source arbitration decision logic and the user behavior prediction model (LSTM network) control mechanism, and generate solutions that better meet user needs.

[0145] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of this application are not limited to the described order of actions, because according to the embodiments of this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of this application.

[0146] Reference Figure 4 The diagram shows a structural schematic of a monitoring device for an embodiment of this application, which may include the following modules: The state determination module 401 is used to determine the current state of the target device corresponding to the target scenario and the desired state for the target scenario after controlling the target device. The weight duration determination module 402 is used to obtain the target device weight of the target device and determine the duration of the mismatch between the current state and the expected state when the current state does not match the expected state. Distance determination module 403 is used to determine the distance information between the user and the target device; The monitoring module 404 is used to determine the deviation information of the target device based on the target device weight, duration, and distance information, and to monitor the target device based on the deviation information.

[0147] In some embodiments, the state determination module 401 is used to acquire multi-source sensing data, which includes at least one of environmental data, user data, and external data; predict the user's target intent based on the multi-source sensing data; and determine the desired state based on the target intent.

[0148] In some embodiments, the state determination module 401 is used to determine the user's explicit instructions or implicit needs based on multi-source sensing data; and to determine the user's target intent based on the explicit instructions or implicit needs.

[0149] In some embodiments, the weight duration determination module 402 is used to determine the basic weight information of the target device and the scene weight information of the target device in the target scenario; determine the abnormal state information of the target device; and determine the weight of the target device based on the basic weight information, the scene weight information, and the abnormal state information.

[0150] In some embodiments, the monitoring module 404 is used to determine a time decay factor based on the duration; determine a spatial correlation factor based on distance information; and determine deviation information based on the target device weight, the time decay factor, the spatial correlation factor, and the degree of deviation between the current state and the desired state.

[0151] In some embodiments, the monitoring module 404 is used to determine the target attenuation coefficient and determine the time attenuation factor based on the target attenuation coefficient and the duration.

[0152] In some embodiments, the monitoring module 404 is configured to determine that the operation of the target device does not meet the user's needs when the deviation information exceeds the deviation threshold, and to determine that the operation of the target device meets the user's needs when the deviation information does not exceed the deviation threshold.

[0153] In some embodiments, the monitoring module 404 is further configured to, after determining that the operation of the target device does not meet the user's needs, obtain a scene association graph, the scene association graph including device nodes, physical space nodes, scene nodes, and influence relationship edges between each node; query the scene association graph to determine the global impact information generated by the adjustment to the target device; and control the target device and other devices according to the global impact information.

[0154] In some embodiments, the monitoring module 404 is further configured to determine whether to execute a key instruction based on whether there is an anomaly in the target device before executing the key instruction for the target device; and / or, determine whether to execute a key instruction based on collected environmental data before executing the key instruction for the target device; and / or, determine whether to execute a key instruction based on user data before executing the key instruction for the target device.

[0155] In this embodiment, after controlling the target device, the current state of the target device in the target scenario is determined, as well as the desired state for the target scenario. When the current state does not match the desired state, the target device weight is obtained, and the duration of the mismatch is determined. The distance information between the user and the target device is determined. Based on the target device weight, duration, and distance information, the deviation information of the target device is determined, and the target device is monitored based on the deviation information. Through this embodiment, based on multi-dimensional parameters, it is possible to effectively distinguish between occasional abnormal phenomena and continuous deviations, and to intelligently determine which devices in the current scenario control are more in line with the current actual needs, thus providing users with a more intelligent user experience.

[0156] This application also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and capable of running on the processor. When the computer program is executed by the processor, it implements the device monitoring method described above.

[0157] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the device monitoring method described above.

[0158] As the device embodiment is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.

[0159] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0160] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, or computer program products. Therefore, embodiments of this application can take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Furthermore, embodiments of this application can take the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0161] This application describes embodiments with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0162] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0163] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal equipment, causing a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0164] Although preferred embodiments of the present application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present application.

[0165] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0166] The monitoring method, apparatus, electronic device, and storage medium of the provided device have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for monitoring equipment, characterized in that, The method includes: After controlling the target device, determine the current state of the target device in the target scenario, and determine the desired state for the target scenario; When the current state does not match the expected state, obtain the target device weight of the target device and determine the duration of the mismatch between the current state and the expected state. Determine the distance information between the user and the target device; Based on the target device weight, the duration, and the distance information, the deviation information of the target device is determined, and the target device is monitored based on the deviation information.

2. The method according to claim 1, characterized in that, Determining the desired state for the target scenario includes: Acquire multi-source sensing data, wherein the multi-source sensing data includes at least one of environmental data, user data, and external data; Based on the multi-source perception data, predict the user's target intent; The desired state is determined based on the stated target intent.

3. The method according to claim 2, characterized in that, The step of predicting the user's target intent based on the multi-source perception data includes: Based on the multi-source sensing data, determine the user's explicit instructions or the user's implicit needs; Determine the user's target intent based on the explicit instructions or the user's implicit needs.

4. The method according to claim 1, characterized in that, The step of obtaining the target device weight of the target device includes: Determine the basic weight information of the target device, and the scene weight information of the target device in the target scene; Determine the abnormal status information of the target device; The target device weight is determined based on the basic weight information, the scene weight information, and the abnormal state information.

5. The method according to claim 1, characterized in that, Determining the deviation information of the target device based on the target device weight, the duration, and the distance information includes: Determine the time decay factor based on the duration. Based on the distance information, determine the spatial association factor; The deviation information is determined based on the target device weight, the time decay factor, the spatial correlation factor, and the degree of deviation between the current state and the desired state.

6. The method according to claim 5, characterized in that, The step of determining the time decay factor based on the duration includes: Determine the target attenuation coefficient; The time decay factor is determined based on the target decay coefficient and the duration.

7. The method according to claim 1, characterized in that, The step of monitoring the target device based on the deviation information includes: When the deviation information exceeds the deviation threshold, it is determined that the operation of the target device does not meet the user's needs; When the deviation information does not exceed the deviation threshold, it is determined that the operation of the target device meets the user's needs.

8. The method according to claim 7, characterized in that, After determining that the operation of the target device does not meet the user's needs, the method further includes: Obtain a scene association graph, which includes device nodes, physical space nodes, scene nodes, and influence relationship edges between each node; Query the scene association map to determine the global impact information of the adjustment to the target device; Based on the global impact information, control is applied to the target device and other devices.

9. The method according to claim 1, characterized in that, The method further includes: Before executing the critical instructions for the target device, determine whether to execute the critical instructions based on whether the target device exhibits any abnormalities; and / or, Before executing the critical instructions for the target device, a determination is made based on the collected environmental data as to whether to execute the critical instructions; and / or, Before executing the key instructions for the target device, the system determines whether to execute the key instructions based on the user's user data.

10. A monitoring device for equipment, characterized in that, The device includes: The state determination module is used to determine the current state of the target device corresponding to the target scenario after controlling the target device, and to determine the desired state for the target scenario. The weight duration determination module is used to obtain the target device weight of the target device and determine the duration of the mismatch between the current state and the expected state when the current state does not match the expected state. The distance determination module is used to determine the distance information between the user and the target device; The monitoring module is used to determine the deviation information of the target device based on the target device weight, the duration, and the distance information, and to monitor the target device based on the deviation information.

11. An electronic device, characterized in that, The device includes a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program, when executed by the processor, implements the monitoring method of the device as claimed in any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, which, when executed by a processor, implements the monitoring method of the device as described in any one of claims 1 to 9.

Images