Household appliance control system, control method, electronic equipment and storage medium

Abstract

The invention discloses a household appliance control system and method, electronic equipment and a storage medium. The household appliance control system comprises: an edge end receives a user instruction, analyzes an intention corresponding to the user instruction, and generates a decision result based on the intention; the cloud end is in communication connection with the edge end, receives first target data sent by the edge end, generates a decision result based on the first target data and sends the decision result to the edge end; wherein the first target data is generated when the edge end cannot analyze the intention corresponding to the user instruction; the edge end is in communication connection with the household appliance, an intelligent agent is deployed on the household appliance, and the edge end is used for sending the decision result to the household appliance so that the household appliance can control the household appliance to work according to the control strategy. Through the mode, the dependence on the cloud is reduced, the response speed of the system is improved, and when the edge end cannot perform reasoning, the cloud is utilized to perform reasoning, so that the normal operation of the system is ensured.

Description

Technical Field

[0001] The present application relates to the technical field of household appliances, and in particular to household appliance control systems, control methods, electronic devices, and storage media. Background Art

[0002] With the rapid development of smart home technology, users are increasingly demanding intelligent control of home appliances. Current smart home systems rely entirely on cloud processing, which can lead to high latency. Summary of the Invention

[0003] The home appliance control system, control method, electronic device, and storage medium provided in this application can reduce dependence on the cloud, improve system response speed, and use the cloud for reasoning when reasoning is impossible on the edge to ensure the normal operation of the system.

[0004] In the first aspect, the present application provides a home appliance control system, which includes: an edge end, which receives user instructions, parses the intention corresponding to the user instructions, and generates a decision result based on the intention; a cloud end, which is communicatively connected to the edge end, receives first target data sent by the edge end, and generates a decision result based on the first target data and sends it to the edge end; wherein the first target data is generated by the edge end being unable to parse the intention corresponding to the user instruction; the edge end is communicatively connected to the home appliance device, and an intelligent agent is deployed on the home appliance device, and the edge end is used to send the decision result to the home appliance device, so that the home appliance device uses the intelligent agent to generate a control strategy and control the operation of the home appliance device according to the control strategy.

[0005] The cloud receives the second target data sent by the edge, generates a decision result based on the second target data, and sends it to the edge; the second target data is generated when the edge detects that the working status feedback from the home appliance is abnormal.

[0006] Among them, the edge end analyzes the decision result, determines the target home appliance corresponding to the decision result, and sends the corresponding decision result to the target home appliance.

[0007] In a second aspect, the present application provides a method for controlling household appliances, which is applied to the edge, and the method includes: receiving user instructions and parsing the intention corresponding to the user instructions; sending first target data to the cloud in response to a first instruction; wherein the first instruction and the first target data are generated by the edge being unable to parse the intention corresponding to the user instruction; receiving the decision result sent by the cloud; the decision result is inferred by the cloud based on the first target data; and sending the decision result to the corresponding household appliance, so that the intelligent agent deployed on the household appliance generates a control strategy based on the decision result and controls the operation of the household appliance according to the control strategy.

[0008] The method also includes: sending second target data to the cloud; receiving the decision result sent by the cloud; the decision result is inferred by the cloud based on the second target data, and the second target data is generated when the edge detects that the working status feedback of the home appliance is abnormal.

[0009] On the third aspect, the present application provides a method for controlling home appliances, which is applied to home appliances, and the method includes: receiving a decision result sent by an edge end; wherein the decision result is generated by parsing the intention corresponding to the user instruction by the edge end or the cloud end; using an intelligent agent to generate a control strategy based on the decision result and control the operation of the home appliance according to the control strategy.

[0010] In a fourth aspect, the present application provides a method for controlling household appliances, which is applied to the cloud. The method includes: receiving first target data sent by the edge end, and generating a decision result based on the first target data; wherein the first target data is generated by the edge end being unable to parse the intention corresponding to the user instruction; sending the decision result to the edge end so that the edge end sends the decision result to the corresponding household appliance; wherein the intelligent agent deployed on the household appliance generates a control strategy based on the decision result and controls the operation of the household appliance according to the control strategy.

[0011] The method further includes: receiving second target data sent by the edge end; generating a decision result based on the second target data and sending it to the edge end; wherein the second target data is generated when the edge end detects that the working status feedback from the home appliance is abnormal.

[0012] In a fifth aspect, the present application provides an electronic device comprising: a communication interface for communicating with a household appliance; and a processor coupled to the communication interface for implementing the method provided in the second aspect, the third aspect, or the fourth aspect.

[0013] In a sixth aspect, the present application provides a computer-readable storage medium having at least one program, which, when loaded and executed by a processor, is used to implement the method provided in the second aspect, the third aspect or the fourth aspect.

[0014] The beneficial effect of the present application is that, different from the prior art, the home appliance control system, control method, electronic device, and storage medium provided by the present application send corresponding data to the cloud when the edge end cannot parse the intention corresponding to the user instruction, use the cloud end to parse the intention corresponding to the user instruction, generate a decision result and send it to the edge end, and then the edge end sends the decision result to the corresponding home appliance, so that the home appliance uses the intelligent body to generate a control strategy and controls the operation of the home appliance according to the control strategy, and can reasonably allocate computing resources so that the edge end also has reasoning capabilities, thereby reducing dependence on the cloud end, improving the system response speed, and using the cloud end for reasoning when the edge end cannot reason, to ensure the normal operation of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without inventive efforts. Among them:

[0016] Figure 1 This is a structural diagram of an embodiment of a home appliance control system provided by the present application;

[0017] Figure 2 This is a flowchart of an embodiment of a method for controlling a household appliance provided by the present application;

[0018] Figure 3 This is a flow chart of another embodiment of the household appliance control method provided by the present application;

[0019] Figure 4 This is a flow chart of another embodiment of the household appliance control method provided by the present application;

[0020] Figure 5 This is a flow chart of another embodiment of the household appliance control method provided by the present application;

[0021] Figure 6 This is a flow chart of another embodiment of the household appliance control method provided by the present application;

[0022] Figure 7 This is a structural diagram of another embodiment of the home appliance control system provided by the present application;

[0023] Figure 8 This is a schematic diagram of the data processing and encryption process at the edge provided by this application;

[0024] Figure 9 This is a schematic diagram of the cloud-based reasoning process provided by this application;

[0025] Figure 10 This is a schematic diagram of the control flow of the local distributed intelligent group provided by this application;

[0026] Figure 11 This is a schematic diagram of the edge-cloud collaborative workflow of the system provided by this application;

[0027] Figure 12 This is a structural diagram of an embodiment of an electronic device provided by the present application;

[0028] Figure 13 It is a flowchart of an embodiment of a computer-readable storage medium provided by this application. DETAILED DESCRIPTION

[0029] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It will be understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for ease of description, only some, rather than all, structures related to the present application are shown in the drawings. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0030] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0031] With the rapid development of smart home technology, users are increasingly demanding intelligent control of home appliances. Current smart home systems rely entirely on cloud processing, which can lead to high latency.

[0032] Based on this, the present application proposes that when the edge is unable to parse the intent corresponding to the user instruction, the corresponding data is sent to the cloud, the cloud is used to parse the intent corresponding to the user instruction, a decision result is generated and sent to the edge, and then the edge sends the decision result to the corresponding home appliance, so that the home appliance uses the intelligent agent to generate a control strategy and controls the operation of the home appliance according to the control strategy. This can reasonably allocate computing resources, so that the edge also has reasoning capabilities, thereby reducing dependence on the cloud, improving system response speed, and using the cloud for reasoning when the edge is unable to reason, to ensure the normal operation of the system. Please refer to any of the following embodiments for details.

[0033] See Figure 1 , Figure 1 FIG1 is a schematic diagram of a structure of an embodiment of a home appliance control system provided by the present application. The home appliance control system 100 includes: an edge terminal 10 and a cloud terminal 20.

[0034] The edge terminal 10 receives user instructions, analyzes the intention corresponding to the user instructions, and generates a decision result based on the intention.

[0035] The cloud 20 is communicatively connected with the edge 10, receives the first target data sent by the edge 10, generates a decision result based on the first target data and sends it to the edge; wherein, the first target data is generated when the edge 10 is unable to parse the intention corresponding to the user instruction.

[0036] In some embodiments, the reasoning capability of the cloud 20 is greater than the reasoning capability of the edge 10 .

[0037] In some embodiments, cloud 20 and edge 10 can each deploy models of different precision and use these models for reasoning to parse user intent. The cloud 20 model is more precise than the edge 10 model. That is, the higher the model precision, the stronger the reasoning capability.

[0038] In some embodiments, cloud 20 can deploy large models, such as large language models. Edge 10 can deploy lightweight models. Specifically, the reasoning capability of a model is determined by its scale: larger models have stronger reasoning capabilities. In some embodiments, cloud 20 can deploy large-scale models, while edge 10 can deploy lightweight models.

[0039] The edge device 10 is in communication with the home appliance, which is equipped with an agent. The edge device is used to send decision results to the home appliance, so that the home appliance can use the agent to generate a control strategy and control the home appliance according to the control strategy. The agent has the ability to make independent decisions.

[0040] In this embodiment, when the edge terminal 10 is unable to parse the intention corresponding to the user instruction, the corresponding data is sent to the cloud 20. By utilizing the characteristic that the reasoning ability of the cloud 20 is greater than the reasoning ability of the edge terminal 10, the intention corresponding to the user instruction is parsed, and a decision result is generated and sent to the edge terminal 10. Then, the edge terminal 10 sends the decision result to the corresponding home appliance device, so that the home appliance device uses the intelligent agent to generate a control strategy and controls the operation of the home appliance device according to the control strategy. The computing resources can be reasonably allocated, so that the edge terminal 10 also has reasoning ability, thereby reducing dependence on the cloud 20, improving the system response speed, and using the cloud 20 for reasoning when the edge terminal 10 cannot reason, to ensure the normal operation of the system.

[0041] In some embodiments, the cloud 20 receives second target data sent by the edge 10, generates a decision result based on the second target data, and sends it to the edge. The second target data is generated when the edge 10 detects an abnormal operating status reported by a home appliance. The above embodiment is applicable to this scenario. For example, after the edge 10 performs inference, it obtains a preliminary decision result and sends this preliminary decision result to the home appliance. The home appliance operates according to this preliminary decision result, but the reported operating status is abnormal. This indicates that the preliminary decision result inferred by the edge 10 is incorrect. Therefore, the corresponding data needs to be sent to the cloud 20, leveraging the cloud 20's more powerful reasoning capabilities to obtain a more accurate decision result. The edge 10 then sends the accurate decision result to the home appliance. This prevents the home appliance from malfunctioning when the preliminary decision result is abnormal, thereby affecting the user experience. In some embodiments, the second target data can include at least one of the preliminary decision result of the edge 10, user instructions, and information about abnormal operating status reported by the home appliance.

[0042] In some embodiments, the edge device 10 analyzes the decision result, determines the target home appliance corresponding to the decision result, and sends the corresponding decision result to the target home appliance. For example, in the home appliance control system 100, the edge device 10 can be communicatively connected to multiple home appliances. These home appliances have different functions and types. The edge device 10 analyzes the decision result, selects the target home appliance corresponding to the decision result from these home appliances based on the decision result, and sends the corresponding decision result to the target home appliance, so that the intelligent agent in the target home appliance executes the corresponding decision result.

[0043] See Figure 2 , Figure 2 This is a flow chart of an embodiment of a method for controlling a home appliance provided by this application. Applied to the edge, the method includes:

[0044] Step 21: Receive user instructions and parse the intent corresponding to the user instructions.

[0045] Step 22: In response to the first instruction, send the first target data to the cloud.

[0046] Among them, the first instruction and the first target data are generated by the edge end being unable to parse the intention corresponding to the user instruction; the reasoning capability of the cloud end is greater than the reasoning capability of the edge end.

[0047] The first target data may include user instructions and related information of the intention corresponding to the user instructions that the edge end cannot parse.

[0048] Step 23: Receive the decision result sent by the cloud.

[0049] The decision result is inferred by the cloud based on the first target data.

[0050] Step 24: Send the decision result to the corresponding home appliance so that the intelligent agent deployed on the home appliance generates a control strategy based on the decision result and controls the operation of the home appliance according to the control strategy.

[0051] In this embodiment, when the edge is unable to parse the intention corresponding to the user instruction, the corresponding data is sent to the cloud. By utilizing the characteristic that the cloud's reasoning ability is greater than the edge's reasoning ability, the intention corresponding to the user instruction is parsed, and a decision result is generated and sent to the edge. The edge then sends the decision result to the corresponding home appliance, so that the home appliance uses the intelligent agent to generate a control strategy and controls the operation of the home appliance according to the control strategy. The computing resources can be reasonably allocated so that the edge also has reasoning capabilities, thereby reducing dependence on the cloud and improving the system response speed. When the edge is unable to reason, the cloud is used for reasoning to ensure the normal operation of the system.

[0052] See Figure 3 , Figure 3 This is a flow chart of another embodiment of the home appliance control method provided by this application. It is applied to the edge end. The method includes:

[0053] Step 31: Send the second target data to the cloud.

[0054] Step 32: Receive the decision result sent by the cloud.

[0055] The decision result is inferred by the cloud based on the second target data, and the second target data is generated when the edge detects that the working status feedback from the home appliance is abnormal.

[0056] This embodiment is applicable to this scenario, for example, after the edge performs reasoning, it obtains a preliminary decision result, sends this preliminary decision result to the home appliance, and the home appliance works according to this preliminary decision result, but the feedback working status is abnormal. This indicates that the preliminary decision result inferred by the edge is incorrect, so it is necessary to send the corresponding data to the cloud, and use the more powerful reasoning ability of the cloud to obtain a more accurate decision result, and let the edge send the accurate decision result to the home appliance, so as to avoid the home appliance from not working properly when the preliminary decision result is abnormal, affecting the user experience. In some embodiments, the second target data can include at least one of the preliminary decision result of the edge, user instructions, and information about abnormal working status feedback from the home appliance.

[0057] See Figure 4 , Figure 4 This is a flow chart of another embodiment of the household appliance control method provided by this application. Applied to household appliances, the method includes:

[0058] Step 41: Receive the decision result sent by the edge; wherein the decision result is generated by the edge or cloud side analyzing the intent corresponding to the user instruction.

[0059] In some embodiments, when the edge cannot parse the intention corresponding to the user instruction, the corresponding data is sent to the cloud, and the intention corresponding to the user instruction is parsed by utilizing the characteristic that the reasoning capability of the cloud 20 is greater than that of the edge 10 to generate a decision result.

[0060] Step 42: Use the intelligent agent to generate a control strategy based on the decision result and control the operation of the home appliance according to the control strategy.

[0061] In this embodiment, when the edge is unable to parse the intention corresponding to the user instruction, the corresponding data is sent to the cloud. By utilizing the characteristic that the cloud's reasoning ability is greater than the edge's reasoning ability, the intention corresponding to the user instruction is parsed, and a decision result is generated and sent to the edge. The edge then sends the decision result to the corresponding home appliance, so that the home appliance uses the intelligent agent to generate a control strategy and controls the operation of the home appliance according to the control strategy. The computing resources can be reasonably allocated so that the edge also has reasoning capabilities, thereby reducing dependence on the cloud and improving the system response speed. When the edge is unable to reason, the cloud is used for reasoning to ensure the normal operation of the system.

[0062] In one application scenario, this embodiment is applicable to this scenario. For example, after the edge performs reasoning, it obtains a preliminary decision result and sends this preliminary decision result to the home appliance. The home appliance operates according to this preliminary decision result, but the feedback working status is abnormal. This indicates that the preliminary decision result inferred by the edge is incorrect, so it is necessary to send the corresponding data to the cloud, and use the cloud's more powerful reasoning capabilities to obtain a more accurate decision result. The edge then sends the accurate decision result to the home appliance to avoid the home appliance from not working properly when the preliminary decision result is abnormal, which affects the user experience. In some embodiments, the second target data can include at least one of the preliminary decision result of the edge, user instructions, and information about abnormal working status feedback from the home appliance.

[0063] See Figure 5 , Figure 5 This is a flow chart of another embodiment of the household appliance control method provided by this application. Applied in the cloud, the method includes:

[0064] Step 51: Receive first target data sent by the edge end, and generate a decision result based on the first target data.

[0065] Among them, the first target data is generated by the edge end's inability to parse the intention corresponding to the user instruction; the cloud end's reasoning capability is greater than the edge end's reasoning capability.

[0066] Step 52: Send the decision result to the edge end, so that the edge end sends the decision result to the corresponding home appliance device.

[0067] Among them, the intelligent agent deployed on the home appliance generates a control strategy based on the decision results and controls the operation of the home appliance according to the control strategy.

[0068] In this embodiment, when the edge is unable to parse the intention corresponding to the user instruction, the corresponding data is sent to the cloud. By utilizing the characteristic that the cloud's reasoning ability is greater than the edge's reasoning ability, the intention corresponding to the user instruction is parsed, and a decision result is generated and sent to the edge. The edge then sends the decision result to the corresponding home appliance, so that the home appliance uses the intelligent agent to generate a control strategy and controls the operation of the home appliance according to the control strategy. The computing resources can be reasonably allocated so that the edge also has reasoning capabilities, thereby reducing dependence on the cloud and improving the system response speed. When the edge is unable to reason, the cloud is used for reasoning to ensure the normal operation of the system.

[0069] See Figure 6 , Figure 6 This is a flow chart of another embodiment of the household appliance control method provided by this application. The method further includes:

[0070] Step 61: Receive second target data sent by the edge end.

[0071] Step 62: Generate a decision result based on the second target data and send it to the edge end; wherein the second target data is generated when the edge end detects that the working status feedback from the home appliance is abnormal.

[0072] This embodiment is applicable to this scenario, for example, after the edge performs reasoning, it obtains a preliminary decision result, sends this preliminary decision result to the home appliance, and the home appliance works according to this preliminary decision result, but the feedback working status is abnormal. This indicates that the preliminary decision result inferred by the edge is incorrect, so it is necessary to send the corresponding data to the cloud, and use the more powerful reasoning ability of the cloud to obtain a more accurate decision result, and let the edge send the accurate decision result to the home appliance, so as to avoid the home appliance from not working properly when the preliminary decision result is abnormal, affecting the user experience. In some embodiments, the second target data can include at least one of the preliminary decision result of the edge, user instructions, and information about abnormal working status feedback from the home appliance.

[0073] See Figure 7 The home appliance control system of this application mainly includes the following components:

[0074] Edge (such as the smart central control screen): responsible for data collection, preliminary processing, encryption and basic intent understanding.

[0075] Large multimodal models are deployed in the cloud to perform complex reasoning and decision making. Secure communication modules are used to ensure secure data transmission between the edge and the cloud.

[0076] Local distributed agent group: Each smart home appliance is equipped with an agent, which is responsible for direct control and status monitoring of the device.

[0077] Edge-Cloud Collaborative Scheduler: Optimizes task allocation and data flow between edge, cloud, and local agents.

[0078] In some embodiments, as Figure 8 As shown in the figure, the data processing and encryption process at the edge includes:

[0079] Data acquisition: Collecting raw data from various sensors and user input devices.

[0080] Preprocessing: Perform basic processing such as noise reduction and formatting on the original data.

[0081] Preliminary intent understanding: Use lightweight models for basic intent classification.

[0082] Data encryption: Use high-strength encryption algorithms to encrypt processed data.

[0083] Secure transmission to the cloud: Encrypted data is sent to the cloud through a secure channel.

[0084] In some embodiments, as Figure 9 As shown in the figure, the inference process in the cloud is as follows:

[0085] Step 91: Receive and decrypt encrypted data from the edge.

[0086] Step 92: Input the multimodal large model for deep understanding and reasoning.

[0087] A large multimodal model can be GPT-4.

[0088] Step 93: Generate detailed control strategies and decisions.

[0089] Step 94: Encrypt the decision result and send it back to the edge end.

[0090] In some embodiments, as Figure 10 As shown in Figure 2, the control flow of the local distributed agent group is as follows:

[0091] Step 101: Receive a control instruction from an edge terminal.

[0092] Step 102: Refine the control strategy based on the current state of the device and environmental conditions.

[0093] Step 103: Execute specific device control operations.

[0094] Step 104: Monitor the equipment operation status and effect.

[0095] Step 105: Feedback the execution result and status information to the edge end.

[0096] In some embodiments, as Figure 11 As shown in the figure, the edge-cloud collaborative workflow of the system is as follows:

[0097] Step 111: The user interacts with the edge device through voice, gesture, or touch.

[0098] Step 112: The edge performs preliminary processing and encryption, and performs basic intent understanding.

[0099] Step 113: If the intention is simple and clear, the edge directly instructs the local agent to execute it.

[0100] Step 114: If the intent is complex or requires contextual understanding, the edge sends the encrypted data to the cloud.

[0101] Step 115: The cloud-based large model performs deep reasoning and generates a detailed control strategy.

[0102] Step 116: The cloud returns the encryption decision result to the edge.

[0103] Step 117: The edge decrypts and analyzes the cloud decision and coordinates the execution of relevant local agents.

[0104] Step 118: The local agent executes the control command and provides real-time feedback on the execution status.

[0105] Step 119: The edge summarizes the feedback information and requests the cloud to assist in decision-making again if necessary.

[0106] In an application scenario, such as intelligent scene control, the workflow is as follows:

[0107] The user said: "I'm going to work from home, please help me adjust the environment."

[0108] The processing flow is as follows:

[0109] The edge collects voice data and performs preliminary intent understanding.

[0110] Because it is a complex scenario, the edge sends encrypted data to the cloud.

[0111] The cloud-based large model understands user needs and generates control strategies for the "home office" scenario.

[0112] The cloud returns encrypted decision results to the edge.

[0113] Edge-side decision-making and coordination of multiple local agents:

[0114] Lighting Intelligence: Adjusts lighting to the right brightness and color temperature for your office.

[0115] Air conditioning agent: Sets the temperature to a comfortable office temperature.

[0116] Curtain intelligence: adjusts the opening and closing of curtains according to outdoor light.

[0117] Audio agent: Play light music as background music.

[0118] Each agent performs control and feeds back status.

[0119] The edge summarizes the feedback and shows that the "home office environment is ready."

[0120] In an application scenario, such as a smart energy management scenario, the system detects changes in electricity prices and the upcoming peak in electricity consumption. The processing flow is as follows:

[0121] The edge receives electricity price information and electricity consumption forecast data.

[0122] A preliminary analysis is performed at the edge to determine the need to optimize the power usage strategy.

[0123] Send encrypted energy data and electricity usage to the cloud.

[0124] A large cloud-based model analyzes data and generates optimized energy usage strategies.

[0125] The cloud returns encrypted decision results to the edge.

[0126] Edge-side decision-making and coordination of multiple local agents:

[0127] Air Conditioning Agent: Adjust to energy-saving mode and slightly increase the temperature setting.

[0128] Water Heater Agent: Preheat water during off-peak hours.

[0129] Washing machine agent: postpone operation to periods when electricity prices are lower.

[0130] Electric vehicle charging intelligent body: arrange charging during the night when electricity prices are low.

[0131] Each agent performs control and continuously monitors energy usage.

[0132] The edge regularly aggregates energy usage data and requests the cloud to re-optimize strategies when necessary.

[0133] In an application scenario, such as smart health monitoring, the system continuously monitors the user's health status through various sensors, adjusts the home environment based on data analysis results, and reminds the user or notifies medical personnel when necessary. The processing flow is as follows:

[0134] The edge continuously collects data such as the user's heart rate, blood pressure, body temperature, and activity level.

[0135] The edge performs preliminary analysis and if the data is within the normal range, it is only stored locally.

[0136] If an anomaly is detected, the edge will send encrypted data to the cloud.

[0137] The cloud-based large model combines user historical data and medical knowledge base for in-depth analysis.

[0138] Generate health assessment reports and environmental adjustment recommendations in the cloud.

[0139] The edge receives the suggestion and coordinates the execution of related agents:

[0140] Air conditioning intelligence: adjusts temperature and humidity to the levels most suitable for user health.

[0141] Lighting Agent: Adjusts lighting based on user conditions.

[0142] Audio Agent: Plays soothing music to improve mood.

[0143] If the situation is serious, the system automatically notifies the preset medical contacts.

[0144] The parameter framework includes: Data sampling frequency: heart rate (every minute), blood pressure (hourly), body temperature (every 10 minutes). Local storage capacity: 7 days of historical data. Abnormality thresholds: heart rate (>100 or <50 bpm), blood pressure (>140 / 90 or <90 / 60 mmHg), body temperature (>37.5°C or <36°C). Cloud analysis cycle: daily routine analysis, real-time analysis of abnormal data. Environmental adjustment response time: <30 seconds.

[0145] In an application scenario, such as smart energy management, the system automatically optimizes household energy usage based on electricity prices, weather, user habits, and device energy consumption, reducing electricity bills and carbon emissions. The process is as follows:

[0146] The edge collects household electricity consumption data, indoor and outdoor temperature, light and other information in real time.

[0147] The edge performs preliminary analysis and predicts short-term electricity demand.

[0148] Send the aggregated data to the cloud periodically (e.g., every hour).

[0149] The large cloud-based model combines electricity price information, weather forecasts, and historical electricity usage patterns for in-depth analysis.

[0150] Generate 24-hour energy usage optimization plan in the cloud.

[0151] The edge receives the solution and coordinates the execution of each agent:

[0152] Air conditioning intelligence: Optimizes operating hours and temperature settings while ensuring comfort.

[0153] Water heater agent: preheat water during periods of low electricity prices.

[0154] Washing machine / dishwasher agent: Schedules operation during times when electricity prices are lower.

[0155] Electric vehicle charging agent: Optimizing charging time and power.

[0156] The system displays energy-saving results and suggestions to users through the App.

[0157] The parameter framework includes:

[0158] Data sampling frequency: power consumption (every 5 minutes), temperature (every 15 minutes), and light intensity (every 30 minutes).

[0159] Local forecast duration: electricity demand within 2 hours.

[0160] Cloud optimization cycle: regular optimization once an hour, and real-time optimization when triggered by high-load equipment (such as electric vehicle charging).

[0161] Energy saving goal: Reduce energy consumption by 10-20% while ensuring user comfort.

[0162] Optimization plan execution delay: <1 minute.

[0163] In an application scenario, such as a smart security system, the system integrates data from multiple sensors and combines it with AI analysis to provide comprehensive home security protection, including intrusion detection, fire warning, and gas leak alarms. The processing flow is as follows:

[0164] The edge continuously collects data such as door and window status, motion sensors, cameras, smoke detectors, and gas sensors.

[0165] The edge performs real-time preliminary analysis and if a potential threat is detected, important data is immediately encrypted and sent to the cloud.

[0166] The cloud-based large model combines multi-source data for in-depth analysis to determine the threat level and type.

[0167] The cloud generates a response strategy and returns it to the edge.

[0168] The edge coordinates the execution of each agent according to the strategy:

[0169] Door lock smart body: Start smart locking.

[0170] Lighting Agent: Adjusts lighting based on the situation (e.g., simulating when someone is home).

[0171] Alarm Agent: issues different alarms based on threat levels.

[0172] Camera Intelligence: Adjusts the shooting angle to improve the video quality.

[0173] The system decides whether to notify the user or call the police directly based on the threat level.

[0174] The parameter framework includes:

[0175] Video analysis frame rate: 10fps (normal) / 30fps (when anomaly is detected).

[0176] Motion sensor sensitivity: Can detect movement within a range of 5 meters.

[0177] Local analysis latency: <100ms.

[0178] Cloud analysis response time: <1 second (critical threats) / <5 seconds (general situations).

[0179] False alarm rate target: <1%.

[0180] System self-check frequency: once a day.

[0181] These application examples demonstrate how the system works in different scenarios, highlighting the advantages of edge computing, cloud-based reasoning, and local agents working together. The parameter framework provides some possible technical indicators to help more specifically illustrate the performance characteristics of the system. In actual applications, these parameters may need to be adjusted based on the specific hardware configuration and user needs.

[0182] See Figure 12 , Figure 12 FIG. 1 is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. The electronic device 120 includes a communication interface 121 and a processor 122 .

[0183] The communication interface 121 is used to communicate with the home appliance. The processor 122 is coupled to the communication interface 121 and is used to implement the method of any of the above embodiments.

[0184] The electronic device 120 can serve as any of the aforementioned cloud, edge, and home appliances.

[0185] See Figure 13 , Figure 13The computer-readable storage medium 130 includes at least one program 131, which, when loaded and executed by a processor, is used to implement the method of any of the above embodiments.

[0186] To summarize, the home appliance control system, control method, electronic device, and storage medium provided by the present application send the corresponding data to the cloud when the edge is unable to parse the intention corresponding to the user instruction, and utilize the characteristic that the cloud's reasoning ability is greater than the edge's reasoning ability to parse the intention corresponding to the user instruction, generate a decision result, and send it to the edge. The edge then sends the decision result to the corresponding home appliance, so that the home appliance uses the intelligent agent to generate a control strategy and controls the operation of the home appliance according to the control strategy. This can reasonably allocate computing resources so that the edge also has reasoning capabilities, thereby reducing dependence on the cloud, improving system response speed, and using the cloud for reasoning when the edge is unable to reason, to ensure the normal operation of the system.

[0187] Specifically, it improves data security: sensitive data is encrypted at the edge, reducing the risk of leakage. It also optimizes system response speed: by rationally allocating tasks, unnecessary data transmission is reduced. It also enhances intelligent decision-making: by leveraging the powerful computing power of the cloud to run complex AI models. It also enables more refined personalized control: by combining the local environment and user habits, customized control strategies are provided. And it enhances device collaboration: by enabling efficient device collaboration through local distributed intelligent agents.

[0188] In the several embodiments provided in this application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules or units is merely a logical functional division. In actual implementation, other division methods may be used, such as combining or integrating multiple units or components into another system, or ignoring or not implementing certain features.

[0189] If the integrated units in the above other embodiments are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) or a processing circuit component (processor) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0190] The above description is only an implementation method of the present application and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the description and drawings of this application, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A household appliance control system, characterized in that: The household appliance control system comprises: The edge receives the user instruction, analyzes the intent corresponding to the user instruction, and generates a decision result based on the intent; A cloud, in communication connection with the edge end, receives first target data sent by the edge end, generates a decision result based on the first target data and sends the decision result to the edge end; wherein the first target data is generated when the edge end cannot parse the intention corresponding to the user instruction; The edge end is communicatively connected with the home appliance, an intelligent agent is deployed on the home appliance, and the edge end is used to send the decision result to the home appliance, so that the home appliance generates a control strategy using the intelligent agent and controls the operation of the home appliance according to the control strategy.

2. The household appliance control system according to claim 1, characterized in that: The cloud receives the second target data sent by the edge, generates a decision result based on the second target data and sends it to the edge; wherein the second target data is generated when the edge detects that the working status feedback from the home appliance is abnormal.

3. The household appliance control system according to claim 1, characterized in that: The edge end analyzes the decision result, determines the target home appliance corresponding to the decision result, and sends the corresponding decision result to the target home appliance.

4. A method for controlling a household appliance, characterized in that: Applied to the edge end, the method includes: Receive user instructions and analyze the intent corresponding to the user instructions; In response to a first instruction, sending first target data to the cloud; wherein the first instruction and the first target data are generated by the edge end being unable to parse the intention corresponding to the user instruction; receiving a decision result sent by the cloud; the decision result is inferred by the cloud end based on the first target data; The decision result is sent to the corresponding household appliance, so that the intelligent agent deployed on the household appliance generates a control strategy according to the decision result and controls the operation of the household appliance according to the control strategy.

5. The method according to claim 4, characterized in that The method further comprises: second target data sent to the cloud; Receive a decision result sent by the cloud; the decision result is inferred by the cloud based on the second target data, and the second target data is generated when the edge end detects that the working status feedback from the home appliance is abnormal.

6. A method for controlling a household appliance, characterized in that: Applied to household electrical appliances, the method comprises: Receiving a decision result sent by the edge end; wherein the decision result is generated by the edge end or the cloud end parsing the intent corresponding to the user instruction; An intelligent agent is used to generate a control strategy according to the decision result and control the operation of the household appliance according to the control strategy.

7. A method for controlling household electrical appliances, characterized in that: Applied to the cloud, the method includes: Receiving first target data sent by the edge end, and generating a decision result based on the first target data; wherein the first target data is generated by the edge end being unable to parse the intention corresponding to the user instruction; The decision result is sent to the edge end so that the edge end sends the decision result to the corresponding home appliance; wherein the intelligent agent deployed on the home appliance generates a control strategy according to the decision result and controls the operation of the home appliance according to the control strategy.

8. The method according to claim 7, characterized in that The method further comprises: Receiving second target data sent by the edge end; A decision result is generated based on the second target data and sent to the edge end; wherein the second target data is generated when the edge end detects that the working status feedback from the home appliance is abnormal.

9. An electronic device, characterized in that: The electronic device comprises: A communication interface, used to communicate with home appliances; A processor, coupled to the communication interface, configured to implement the method according to any one of claims 4 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium has at least one program, and when the at least one program is loaded and executed by the processor, it is used to implement the method according to any one of claims 4 to 8.

Images