Intelligent household appliance interaction control system based on embedded software

Abstract

This invention discloses an intelligent home appliance interactive control system based on embedded software. The system operation process specifically includes the following steps: acquiring user input signals; parsing the user input signals based on the embedded software architecture to generate standardized control commands; establishing communication connections with multiple intelligent home appliances through an Internet of Things (IoT) protocol and acquiring real-time environmental data and the real-time operating status of the intelligent home appliances; jointly analyzing user historical behavior data and user input signals; adjusting the operating parameters of the intelligent home appliances; and returning interactive response information to the user through at least one of the following methods: voice feedback module, touch interface, gesture recognition interface, or mobile APP. This invention has the following advantages and effects: it can adaptively integrate multimodal commands, dynamically match device capabilities and user intentions, and achieve efficient and accurate control across devices and protocols through standardized command generation, real-time load perception, and dynamic priority calculation.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for home appliances, and in particular to an intelligent home appliance interactive control system based on embedded software. Background Technology

[0002] With the rapid development of IoT technology, the interaction methods of smart home appliances are becoming increasingly diversified. Users can control home appliances through various means such as voice, gestures, touch screens, and mobile apps. However, existing systems have fundamental defects in the deep integration of multimodal commands and dynamic device adaptation, resulting in low efficiency in the generation and execution of control commands, which seriously restricts the user experience.

[0003] The shortcomings of deep integration of multimodal commands are manifested in the lack of a unified command parsing mechanism when the current system receives diverse user inputs; the signal formats generated by different input methods vary greatly, making it difficult to standardize control commands, especially in scenarios where cross-brand and cross-protocol devices coexist, the system cannot efficiently match the unique identifier of the device. For example, when users control devices with fuzzy descriptions, traditional methods rely on precise device name matching. If the device registration information is incomplete or the location identifier conflicts, command failures or misoperations are likely to occur; in addition, the existing parsing logic does not consider the real-time load of the device and the user's historical behavior, making it difficult to dynamically optimize the command execution priority, resulting in response delays or a surge in energy consumption for high-load devices.

[0004] The drawbacks of dynamic device adaptation are that when multiple devices execute instructions concurrently, the fixed priority configuration cannot cope with dynamic load changes, which may cause critical tasks to be blocked by low-priority instructions; in addition, there is a lack of closed-loop verification after parameter adjustment.

[0005] Therefore, there is an urgent need for an interactive control system that can adaptively integrate multimodal commands, dynamically match device capabilities and user intentions, and achieve efficient and accurate control across devices and protocols through standardized command generation, real-time load perception and dynamic priority calculation. Summary of the Invention

[0006] The purpose of this invention is to provide an intelligent home appliance interactive control system based on embedded software to solve the problems mentioned in the background art.

[0007] The above-mentioned technical objective of the present invention is achieved through the following technical solution:

[0008] The intelligent home appliance interactive control system based on embedded software includes the following steps in its operation:

[0009] S100. Acquire user input signals, wherein the user input signals include at least one of voice commands, touch operations, gesture actions, or mobile APP control commands.

[0010] S200: Based on the embedded software architecture, the user input signal is parsed to generate standardized control commands;

[0011] S300 establishes communication connections with multiple smart home appliances through IoT protocols and obtains real-time environmental data and the real-time operating status of smart home appliances;

[0012] S400, performs joint analysis of user historical behavior data and user input signals; including:

[0013] S410. Use a neural network model to perform cluster analysis on historical user operation data to identify user preference patterns and device usage habits;

[0014] S420: Combine real-time environmental data with equipment operating status to generate control strategies;

[0015] S500: Adjust the operating parameters of smart home appliances;

[0016] The S600 returns interactive response information to the user through at least one of the following methods: voice feedback module, touch interface, gesture recognition interface, or mobile APP.

[0017] By adopting the above technical solutions and supporting multiple input methods such as voice commands, touch operations, gestures, and mobile APP control commands, the interaction dimensions between users and smart home appliances are significantly expanded. Furthermore, the fusion design of multimodal inputs can cover the needs of different user groups, enhancing the system's versatility. Through embedded software architecture, the system parses the raw input signals and generates standardized control commands, solving the problem of incompatibility in command formats between smart home appliances of different brands or protocols. The standardized command format, as an intermediate abstraction, shields the heterogeneity of the underlying devices, enabling the system to uniformly manage multiple devices and reducing integration complexity. The system obtains real-time data through IoT protocols. Real-time environmental data and equipment operating status provide a data foundation for the dynamic adjustment of control strategies; this data-driven decision-making mechanism improves the system's response efficiency and reliability; cluster analysis of user historical behavior data can uncover user preference patterns and generate personalized equipment control strategies accordingly, enabling advance adjustment of equipment parameters, reducing the number of manual interventions by users, and improving the level of intelligence; in addition, the system dynamically adjusts equipment parameters based on priority algorithms and energy consumption constraint models to ensure that critical tasks are executed first, while optimizing energy efficiency; furthermore, through the feedback module, users can obtain real-time operation results or abnormal alarms, forming an interactive closed loop and enhancing users' trust in the system.

[0018] A further setting is that S200 specifically includes the following steps:

[0019] S210. Receive the original input signals of the voice command, touch operation, gesture action or mobile APP control command, and store the original input signals in a temporary buffer according to their type.

[0020] S220. Based on the type of the original input signal, call the corresponding parsing module to generate intermediate instructions;

[0021] S230. Map the intermediate instructions to a preset standardized control instruction format to generate standardized control instructions; specifically including the following sub-steps:

[0022] S230.1 Extract the operation type, device description information and expected target parameter value from the intermediate instructions, wherein the operation type includes a switch instruction, parameter adjustment instruction or mode switching instruction, and the device description information includes device name, location identifier or function category;

[0023] S230.2. Match the device in the device registry based on the device description information:

[0024] If the device name is exactly the same as the device identifier in the device registry, then the corresponding unique device code is extracted directly.

[0025] If the device description information is a location identifier, then the identifiers of all devices under that location are retrieved in the device registry, and the target device is filtered according to the function category;

[0026] If a match fails, a fuzzy search algorithm is invoked to rematch based on device functional similarity.

[0027] S230.3 Select a standardized control instruction template based on the operation type:

[0028] If it is a switch command, the basic control template is invoked, which includes a device identifier and operation type; if it is a parameter adjustment command, the numerical template is invoked, which includes a device identifier, target parameter value, and adjustment step size; if it is a mode switching command, the mode template is invoked, which includes a device identifier and preset mode code.

[0029] S230.4. Fill in the fields in the basic control template, numerical template, or pattern template, including:

[0030] Fill the device unique code matched from the device registry in S230.2 into the device identifier field of the corresponding standardized control instruction template; if the device description information is a location identifier, select the device code with the highest priority under that location according to the function category;

[0031] If the basic control template is called, fill in the Boolean status parameters according to the on / off state of the intermediate instruction; if the numerical template is called, fill in the target parameter value parsed from the intermediate instruction into the corresponding field; if the mode template is called, match the preset mode code according to the mode name of the intermediate instruction and fill it in; if the numerical template or mode template is called, fill in the expected target parameter value obtained from the parsing of the intermediate instruction into the corresponding field as the reference value for device adjustment.

[0032] Priority values ​​are calculated using a priority algorithm based on the device's real-time load rate and the user's historical operation frequency.

[0033] S230.5 Perform syntax validation on the corresponding standardized control instruction template after filling in the code;

[0034] S230.6 Confirm the verified basic control template, numerical template or pattern template as standardized control instructions.

[0035] By adopting the above technical solution, the system can process instructions from different input sources in parallel by classifying and storing the raw input signals into a temporary buffer, avoiding parsing conflicts caused by multimodal signal mixing. It generates intermediate instructions by calling a dedicated parsing module based on the input type, realizing the conversion from raw signals to intermediate semantics. The intermediate instruction layer provides a unified semantic expression for different input sources, simplifying the standardized instruction generation process. Accurate matching of device description information through the device registry ensures that control instructions are accurately associated with the target device, effectively avoiding misoperations caused by device aliases or ambiguous descriptions. Syntax validation of the generated standardized instructions prevents device malfunctions due to parsing errors or out-of-bounds values, thereby avoiding device damage or security risks.

[0036] A further setting is that the priority algorithm calculates the priority value by including:

[0037] If the real-time load rate of the device is greater than or equal to the first threshold, the priority value will be set to the lowest level.

[0038] If the real-time load rate is less than the second threshold and the user's historical operation frequency exceeds the third threshold, then the priority value is set to the highest level.

[0039] If the real-time load rate is less than the second threshold and the user's historical operation frequency does not exceed the third threshold, then the priority value is calculated.

[0040] If the priority value triggers the lowest level, a load alarm will be sent to the user terminal, and a delayed execution or device switching solution will be recommended in the mobile APP.

[0041] By adopting the above technical solution, the execution order of instructions is dynamically adjusted through a priority algorithm, ensuring that instructions from high-load devices are downgraded, while instructions from frequently used devices are executed first. This mechanism balances the issue of device resource contention and avoids a decline in overall system performance due to local overload. The user's historical operation frequency is incorporated into the priority calculation, giving instructions from frequently used devices a higher execution weight. For example, instructions for humidifiers that users turn on daily at set times will be processed first, while devices used occasionally will be scheduled according to the default priority. This design aligns with user habits and improves perceived response speed.

[0042] A further setting is that S410 specifically includes:

[0043] Data preprocessing includes:

[0044] Extract historical operation data from the user behavior database, including operation timestamps, device identifiers, operation types, and operation parameters;

[0045] Normalization is performed to convert the operation timestamp into a time period label, and the operation parameters are linearly scaled to the [0,1] interval according to the device's allowable range;

[0046] User behavior grouping based on K-means clustering algorithm includes:

[0047] The preprocessed historical operation data is combined into a user operation sequence according to the device identifier and time period label. The user operation sequence consists of operation records within a continuous time window.

[0048] The optimal number of clusters is determined by the elbow rule, and user behavior cluster groups are generated. Each user behavior cluster group is associated with specific device usage habits.

[0049] Construct a user-device association matrix, including:

[0050] Based on the user behavior clusters, the operation frequency and parameter distribution of devices within each cluster are statistically analyzed to generate a user preference pattern matrix. This user preference pattern matrix includes the parameter distribution range P of the target devices. range Store the user preference pattern matrix in the user behavior analysis module.

[0051] By adopting the above technical solution, the impact of differences in the dimensions of parameters of different devices on the clustering results is eliminated by converting operation timestamps into time period labels and scaling parameter values ​​to the [0,1] range, so as to accurately identify user preference combinations. Based on the elbow rule to determine the optimal number of clusters, the system can automatically discover potential behavior patterns without pre-setting user group classifications, which helps the system preload device parameter configurations under different scenarios. The generated user preference pattern matrix provides data support for control strategies. For example, if the matrix shows that users often set the air conditioner temperature to 25-26℃, the system can set this range as the default adjustment range, reducing the number of times users manually input.

[0052] A further setting is that S420 specifically includes:

[0053] Extract real-time environmental data and real-time operating status of smart home appliances obtained from S300, and integrate the parameter distribution range P of the target device in the user preference pattern matrix. range Integrated into a system that includes ambient temperature E temp Ambient humidity E humidity Equipment load rate D load and task queue length Q length Environment state vector S t ;

[0054] Control policies are generated based on a rules engine, including:

[0055] If the operating parameters of the current standardized control command are within the parameter distribution range P range Internally, it directly calls the standardized control instructions generated in S200 and adds them to the execution queue;

[0056] If the operating parameters deviate from the parameter distribution range P range However, if the value is less than the preset threshold, a gradual adjustment command is generated, and a new standardized control command is generated by refilling the basic control template, numerical template, or mode template according to S230.4.

[0057] If the operating parameters deviate from the parameter distribution range P range If the preset threshold is exceeded, a user confirmation process is triggered, and a prompt message is pushed via voice or APP. After the user confirms, a standardized control command is generated according to the confirmation parameters.

[0058] If the equipment load rate D load If the load rate is greater than or equal to the device load rate threshold, non-emergency commands will be suspended, and only security commands will be executed.

[0059] By adopting the above technical solutions, and integrating the distribution of ambient temperature, humidity, equipment load rate, task queue length, and user preference parameters, a complete environmental state vector is formed. This allows control decisions to move beyond relying on a single user command, comprehensively assessing the real-time environment and user habits to achieve dynamic adaptive control. This significantly improves the system's intelligence level and adaptability to complex scenarios. Introducing the user preference matrix into the environmental state vector ensures that control commands are intelligently adjusted within parameter ranges that align with user habits. Even without explicit user commands, the system can make appropriate inferences based on preferences, resulting in more natural and user-expected device responses. If operating parameters deviate significantly from the user preference range, the system introduces gradual adjustment or triggers a user confirmation process, effectively preventing misoperation and ensuring device safety and user experience. This flexible response mechanism balances intelligent automation with user control. When equipment load exceeds limits, non-emergency control commands are suspended first, retaining only security tasks to avoid system overload, extend equipment lifespan, and improve overall energy efficiency management. Through standardized environmental state vector processing, the system can coordinate the responses of multiple devices simultaneously, improving the consistency and comfort of the overall home environment, and possessing good scalability and integration.

[0060] A further setting is that S500 specifically includes the following steps:

[0061] S510. Based on device priority, insert the control instructions of high-priority devices at the front of the execution queue;

[0062] S520: Calculate the energy consumption threshold for equipment operation based on the energy consumption constraint model. If the adjusted parameters exceed the threshold, trigger an alarm and recommend an energy-saving alternative.

[0063] S530: After the equipment parameters are adjusted, the system collects feedback data from the equipment in real time through sensors to verify the adjustment effect. If the expected results are not achieved, the control strategy is regenerated.

[0064] By adopting the above technical solutions and employing a priority sorting mechanism, instructions from high-priority devices such as security and environmental control devices are placed at the front of the execution queue, ensuring that critical tasks are completed first and effectively improving the timeliness and reliability of the overall system response. Combined with an energy consumption constraint model, the system calculates the predicted energy consumption of adjusted devices in real time. If the energy consumption exceeds a threshold, it proactively issues an alarm and pushes an energy-saving solution, significantly improving the energy efficiency management level of the smart home appliance system and meeting the needs of green and low-carbon development. By collecting the actual operating parameters of the adjusted devices through sensors and comparing them with target values, the system achieves precise adjustment and anomaly identification, forming a complete self-feedback closed loop, greatly improving the system's control accuracy and reliability. If the adjustment effect is unsatisfactory, the system can adaptively regenerate the control strategy instead of simply reporting an error and interrupting, demonstrating the system's robustness and continuous optimization capabilities, reducing the need for manual intervention, and improving the user experience. Through real-time energy consumption monitoring and prediction mechanisms, the system not only supports total energy consumption control but can also dynamically adjust the operating mode according to time-of-use pricing strategies, saving users energy expenditures and reducing peak grid pressure, resulting in significant benefits.

[0065] A further setting is that S510 specifically includes:

[0066] The device identifier and operation type are extracted from the standardized control instructions and matched with the device priority configuration table stored in the device registry; wherein, the priority configuration table is predefined based on the device function category, with security devices having a higher priority than environmental control devices, and environmental control devices having a higher priority than entertainment devices;

[0067] If multiple devices have the same priority, they are sorted a second time based on the priority value calculated in S230.4, with the larger the value, the earlier the device is sorted.

[0068] Based on the sorting results, control commands for high-priority devices are dynamically inserted into the front position of the execution queue, and the task execution order is updated through the queue management module.

[0069] By adopting the above technical solutions, a preset priority configuration table based on device function categories enables the orderly allocation of execution resources when multiple devices operate collaboratively, avoiding response delays or system lags caused by resource contention. When multiple devices have the same priority, a dynamic priority value calculated based on user frequency and load is further introduced to flexibly adjust the sorting, ensuring that the most urgent and important device control needs are met in different usage scenarios. Dynamic command queuing and queue management avoid the delays in important commands caused by traditional FIFO queues, resulting in smoother and more efficient system response, especially in multi-tasking environments like homes. Priority values ​​can dynamically change according to different users' operating habits, allowing the system to learn and adapt to the different preferences of family members, reflecting a highly personalized and intelligent experience.

[0070] A further setting is that S520 specifically includes:

[0071] Extract the rated power, historical average energy consumption, and current operating mode parameters of the target device from the device registry, and construct an energy consumption prediction function.

[0072] Calculate the real-time energy consumption threshold allowed by the device based on the total energy consumption limit and time-of-use pricing strategy set by the user.

[0073] Substitute the expected target parameter values ​​recorded in the standardized control instructions into the energy consumption prediction function. If the predicted energy consumption value exceeds the real-time energy consumption threshold, an energy consumption alarm signal will be triggered.

[0074] Energy-saving alternatives are retrieved based on device function similarity, including:

[0075] Filter low-power device identifiers with similar functions in the device registry;

[0076] Alternatively, it can generate parameter adjustment suggestions, reduce the expected target parameter value by a preset ratio, and then recalculate the energy consumption value.

[0077] The energy-saving alternative is pushed to the user via a mobile app or voice module, and the user confirmation is awaited.

[0078] By adopting the above technical solutions, the system predicts energy consumption based on parameter adjustments before the device operates. If the energy consumption exceeds the user-set threshold, an early warning is issued, effectively preventing overload operation or unreasonable energy waste and ensuring the safe and stable operation of the system. The system can not only adjust parameters within the device to reduce energy consumption, but also recommend low-power devices with similar functions, providing users with diverse and personalized energy-saving options. Energy-saving suggestions are pushed through the APP or voice, which not only lets users understand the device's energy consumption, but also encourages users to actively participate in energy management, improving the green sustainability of the overall smart energy use in the home.

[0079] A further setting is that S530 specifically includes:

[0080] Real-time operating parameters are obtained from the sensor nodes of the target device via IoT protocols;

[0081] Compare the real-time operating parameters with the expected target parameter values ​​stored in the standardized control instructions:

[0082] If the deviation between the two is less than or equal to the set percentage, the adjustment is confirmed to be effective and recorded in the equipment operation log;

[0083] If the deviation between the two exceeds the set percentage, an anomaly flag is triggered and the deviation type is extracted.

[0084] Invoke the adaptive correction algorithm based on the deviation type:

[0085] If the deviation is due to parameter adjustment, the target parameter value is recalculated using the gradient descent method, and a new standardized control command is generated.

[0086] If the delay is due to device response, the retry interval for the instruction in the execution queue will be extended.

[0087] Add the revised standardized control instructions to the execution queue and perform the following operations:

[0088] Check if the current number of retries exceeds the preset maximum retry threshold; if it does not exceed the threshold, repeat the process of S530; if it exceeds the threshold, trigger the abnormal termination process, which includes sending an adjustment failure alarm message to the user terminal and recommending manual intervention.

[0089] By adopting the above technical solutions, and through real-time comparison of the adjusted equipment status with the target parameters, the actual effect of each control command is ensured, avoiding misadjustment or failure, thus improving the consistency and reliability of system control. Correction algorithms are called according to the type of deviation, and targeted adaptive adjustments are made instead of simple retries or interruptions, greatly enhancing the system's self-healing capabilities and problem-solving efficiency. A maximum retry limit mechanism is designed to avoid resource waste caused by infinite retries while ensuring that control tasks are completed to the greatest extent possible within a reasonable number of attempts, balancing efficiency and stability. When anomalies cannot be automatically corrected, alarms and manual intervention suggestions are promptly pushed to the user, ensuring the user's ultimate control over system operation and improving system transparency and trustworthiness. Finally, based on the adaptive correction mechanism, anomaly handling data can be continuously accumulated to train a more efficient self-learning model.

[0090] In summary, the present invention has the following beneficial effects:

[0091] It can adaptively integrate multimodal commands, dynamically match device capabilities with user intent, and achieve efficient and precise control across devices and protocols through standardized command generation, real-time load perception, and dynamic priority calculation. Attached Figure Description

[0092] Figure 1 This is a schematic diagram of the overall process of an embodiment;

[0093] Figure 2 This is a flowchart illustrating S200 in the embodiment;

[0094] Figure 3 This is a flowchart illustrating S410 in the embodiment;

[0095] Figure 4 This is a flowchart illustrating S500 in the embodiment. Detailed Implementation

[0096] The present invention will be further described in detail below with reference to the accompanying drawings.

[0097] As attached Figure 1 As shown;

[0098] This embodiment discloses an intelligent home appliance interactive control system based on embedded software: the system operation process specifically includes the following steps:

[0099] S100. Acquire user input signals, wherein the user input signals include at least one of voice commands, touch operations, gesture actions, or mobile APP control commands.

[0100] S200: Based on an embedded software architecture, it parses user input signals and generates standardized control commands.

[0101] Acquiring user input signals includes:

[0102] The system receives voice commands through a voice recognition module, performs noise reduction and semantic parsing on the voice commands, and extracts key control keywords.

[0103] The system acquires the coordinates and pressure data of the user's touch operation through a touch sensor, and matches the corresponding control intent according to a preset touch pattern library.

[0104] The system captures user gestures using a camera, extracts and classifies the features of the gesture trajectories using image recognition algorithms, and generates gesture control commands.

[0105] Remote control commands are received via a mobile app, and the commands are encrypted and verified to ensure communication security.

[0106] S300 establishes communication connections with multiple smart home appliances through IoT protocols and obtains real-time environmental data and the real-time operating status of smart home appliances; specifically, it dynamically selects one of Zigbee, Wi-Fi or Bluetooth protocols as the main communication channel according to the device type.

[0107] S400, performs joint analysis of user historical behavior data and user input signals; including:

[0108] S410. Use a neural network model to perform cluster analysis on historical user operation data to identify user preference patterns and device usage habits;

[0109] S420: Combine real-time environmental data with equipment operating status to generate control strategies;

[0110] S500: Adjust the operating parameters of smart home appliances;

[0111] The S600 returns interactive response information to the user through at least one of the following methods: voice feedback module, touch interface, gesture recognition interface, or mobile APP.

[0112] Example 1

[0113] In S100, users send the command "turn on the living room air conditioner" via a mobile app, and the system receives and verifies the validity of the command via HTTPS encryption.

[0114] The voice module receives the voice command "raise the temperature to 25℃", and after noise reduction processing, extracts the keywords "temperature" and "25℃".

[0115] The touchscreen detects the user's swipe gesture (coordinate range X:100-200, Y:50-150) and matches it with the preset "wind speed adjustment" mode;

[0116] The camera captured the user's waving gesture, which was then classified as a "power off device" gesture by image recognition.

[0117] In the S200, the voice analysis module maps "25℃" into a numerical intermediate command, which includes the operation type as parameter adjustment, the device as air conditioner, and the target value as 25.

[0118] The touch parsing module converts the sliding coordinates into an intermediate command of "wind speed +2 levels";

[0119] The protocol parsing module decrypts the APP commands and generates intermediate commands, which include the operation type as switch, the device as air conditioner, and the status as on.

[0120] In the S300, the system selects Wi-Fi as the main communication protocol for the air conditioner to obtain its real-time temperature of 23°C and load rate of 60%; it connects to the temperature and humidity sensor via the Zigbee protocol to collect environmental data of temperature of 24°C and humidity of 50%.

[0121] In S400, neural network clustering discovers that users habitually set their air conditioner to 26°C at night, generating a preference pattern matrix, in which the parameter distribution range P range The temperature range is 24-28℃; the rule engine determines that the current command's temperature of 25℃ falls within the parameter distribution range P. range Internally, control strategies are generated directly.

[0122] In S500 and S600, the system sends standardized control commands to the air conditioner, including device code AC_001 and target temperature 25℃; after the air conditioner is adjusted, it pushes a feedback message "Temperature set to 25℃" through the APP.

[0123] As attached Figure 2 As shown;

[0124] S200 specifically includes the following steps:

[0125] S210: Receives raw input signals from voice commands, touch operations, gestures, or mobile app control commands, and stores the raw input signals in a temporary buffer according to their type.

[0126] S220. Based on the type of the original input signal, call the corresponding parsing module to generate intermediate instructions;

[0127] Specifically: if the input signal is a voice command, the voice parsing module is called to perform semantic segmentation and keyword extraction on the voice command to generate intermediate commands;

[0128] If the input signal is a touch operation, the touch parsing module is called to perform pattern matching on the touch coordinates and pressure data, and then maps them into intermediate instructions according to the preset touch pattern library.

[0129] If the input signal is a gesture, the gesture parsing module is called to track the trajectory of the continuous image frames captured by the camera and classify them into intermediate instructions through a convolutional neural network;

[0130] If the input signal is a control command from a mobile app, the protocol parsing module is called to decrypt the encrypted command and convert it into an intermediate command;

[0131] S230. Map the intermediate instructions to the preset standardized control instruction format to generate standardized control instructions; specifically including the following sub-steps:

[0132] S230.1 Extract the operation type, device description information and expected target parameter value from the intermediate instructions. The operation type includes switch instructions, parameter adjustment instructions or mode switching instructions. The device description information includes device name, location identifier or function category.

[0133] S230.2. Match the device in the device registry based on the device description information:

[0134] If the device name is exactly the same as the device identifier in the device registry, then the corresponding unique device code is extracted directly.

[0135] If the device description information is a location identifier, then the identifiers of all devices under that location are retrieved in the device registry, and the target device is filtered according to the function category;

[0136] If a match fails, a fuzzy search algorithm is invoked to rematch based on device functional similarity.

[0137] S230.3 Select a standardized control instruction template based on the operation type:

[0138] If it is a switch command, the basic control template is called, which contains the device identifier and operation type; if it is a parameter adjustment command, the numerical template is called, which contains the device identifier, target parameter value and adjustment step size; if it is a mode switching command, the mode template is called, which contains the device identifier and preset mode code.

[0139] S230.4, Fill in the fields in the basic control template, numeric template, or schema template, including:

[0140] Fill the device unique code matched from the device registry in S230.2 into the device identifier field of the corresponding standardized control instruction template; if the device description information is a location identifier, select the device code with the highest priority under that location according to the function category;

[0141] If the basic control template is called, fill in the Boolean status parameters according to the on / off state of the intermediate instruction; if the numerical template is called, fill in the target parameter value parsed from the intermediate instruction into the corresponding field; if the mode template is called, match the preset mode code according to the mode name of the intermediate instruction and fill it in; if the numerical template or mode template is called, fill in the expected target parameter value obtained from the parsing of the intermediate instruction into the corresponding field as the reference value for device adjustment.

[0142] Priority values ​​are calculated using a priority algorithm based on the device's real-time load rate and the user's historical operation frequency.

[0143] S230.5, Perform syntax validation on the corresponding standardized control instruction template after filling in the blanks; including:

[0144] Check if the device identifier is a valid code that is already bound in the device registry;

[0145] Verify that the target parameter value is within the threshold range allowed by the device;

[0146] S230.6 Confirm the verified basic control template, numerical template or pattern template as standardized control instructions.

[0147] Example 2

[0148] In S210 and S220, the voice command "turn off the bedroom light" is stored in the cache, and the voice parsing module is called to segment out the keywords "turn off" and "bedroom light";

[0149] In S230, the intermediate instruction "Bedroom Light" matches the code "Light_02" in the device registry.

[0150] Select the basic control template and fill in the device identifier "Light_02" and the status parameter "Off".

[0151] During template verification, it was found that the device code was valid and the status parameters were legal, and standardized control instructions were generated.

[0152] If the user describes "living room light" but this name does not exist in the registry, a fuzzy search is performed to match devices with the location "living room" and the function "lighting" (such as ceiling lights and wall lights), and the ceiling light is selected according to priority.

[0153] Specifically, the process of calculating the priority value using the priority algorithm includes:

[0154] If the real-time load rate of the device is greater than or equal to the first threshold, the priority value will be set to the lowest level.

[0155] If the real-time load rate is less than the second threshold and the user's historical operation frequency exceeds the third threshold, then the priority value is set to the highest level.

[0156] If the real-time load rate is less than the second threshold and the user's historical operation frequency does not exceed the third threshold, the priority value is calculated according to the following formula:

[0157] P=(f user / f max ×W1)+(W2-l real / C)

[0158] Where P is the priority value, f user f represents the user's historical operation frequency. max The preset maximum allowable frequency, l real The real-time load rate of the device is represented by W1, W1, and C, which are configurable weighting coefficients and normalization constants.

[0159] If the priority value triggers the lowest level, a load alarm will be sent to the user terminal, and a delayed execution or device switching solution will be recommended in the mobile APP.

[0160] Example 3

[0161] Scenario A

[0162] The system detected that the load rate of the home theater system with device ID CHANG HT_001 exceeded the first threshold of 80%, and set its priority to the lowest level.

[0163] The robot vacuum cleaner with device ID Cleaner_02 has a load rate of 20%, which is less than the second threshold of 50%. The user's historical operation frequency is 2 times / day, which is less than the third threshold of 5 times / day. According to the formula, the priority value P = 0.44, where W1 = 0.6, W2 = 0.4, and C = 100. Its priority value is lower than the priority value P = 0.07 of the air conditioner.

[0164] The system adds the robot vacuum's command to a low-priority queue and sends a notification to the app: "High load detected in the home theater system. The vacuuming task will be delayed for 30 minutes. Force start immediately?"

[0165] Its technical advantage lies in preventing overload of high-load equipment while providing flexible user intervention options.

[0166] Scene B

[0167] The smoke sensor triggered an alarm; its device ID is Smoke_01. At the same time, the user sent the message "Close all curtains" via the APP.

[0168] The adjustment process is as follows: Security devices are prioritized to the highest level, and environmental control devices are prioritized to level 3; the system immediately interrupts the currently executing curtain closing command and prioritizes the smoke alarm; a security command is forcibly inserted into the execution queue: "Open all windows for ventilation and close the gas valve".

[0169] If the curtain motor is in a half-open state due to a command interruption, the system records the interruption location and automatically resumes operation after the security task is completed.

[0170] As attached Figure 3 As shown;

[0171] S410 specifically refers to:

[0172] Data preprocessing includes:

[0173] Extract historical operation data from the user behavior database, including operation timestamps, device identifiers, operation types, and operation parameters;

[0174] Normalization is performed to convert the operation timestamp into a time period label, and the operation parameters are linearly scaled to the [0,1] interval according to the device's allowable range;

[0175] User behavior grouping based on K-means clustering algorithm includes:

[0176] The preprocessed historical operation data is combined into user operation sequences according to device identifiers and time period labels. The user operation sequence consists of operation records within a continuous time window.

[0177] The optimal number of clusters is determined by the elbow rule, and user behavior clusters are generated. Each user behavior cluster is associated with specific device usage habits.

[0178] Construct a user-device association matrix, including:

[0179] Based on user behavior clustering, the operation frequency and parameter distribution of devices within each cluster are statistically analyzed to generate a user preference pattern matrix. This matrix contains the parameter distribution range P of the target device.range Store the user preference pattern matrix in the user behavior analysis module.

[0180] Example 4

[0181] Historical operation data (such as "18:30, air conditioner, set temperature 26℃") is converted into time period label "evening" and normalization parameter 0.7.

[0182] K-means categorizes user actions into three types: adjusting the environment after get off work on weekdays, high-frequency use of entertainment devices on weekends, and automatic activation of security devices at night.

[0183] Generate a user-device matrix and statistically analyze the parameter distribution range P of the air conditioner at night. range The temperature is 24-28℃.

[0184] Specifically, the S420 is as follows:

[0185] Extract real-time environmental data and real-time operating status of smart home appliances from S300, and integrate the parameter distribution range P of the target device in the user preference pattern matrix. range This is integrated and encoded into an environment state vector S. t =[E temp E humidity D load Q length, P range ]; where E temp For ambient temperature, E humidity For ambient humidity, D load Q represents the equipment load rate. length This is the length of the task queue;

[0186] Control policies are generated based on a rules engine, including:

[0187] If the operating parameters of the current standardized control command are within the parameter distribution range P range Internally, it directly calls the standardized control instructions generated in S200 and adds them to the execution queue;

[0188] If the operating parameters deviate from the parameter distribution range P range However, if the value is less than 10% of the preset threshold, a gradual adjustment command is generated, and a new standardized control command is generated by refilling the basic control template, numerical template, or mode template according to S230.4.

[0189] If the operating parameters deviate from the parameter distribution range P range If the threshold is exceeded by 10%, a user confirmation process is triggered, and a prompt message is pushed via voice or APP. After the user confirms, a standardized control command is generated according to the confirmation parameters.

[0190] If the equipment load rate Dload If the load rate is greater than or equal to the device load rate threshold, non-emergency commands will be suspended, and only security commands will be executed.

[0191] Example 5

[0192] Scenario A

[0193] Environment state vector S t = [24℃, 50%, 60%, 3, 24-28℃];

[0194] The user command "Set the air conditioner to 22℃" deviates from the parameter distribution range P. range If the threshold is exceeded by 10%, an app confirmation prompt will be triggered.

[0195] If the user confirms, a standardized instruction is generated; if they refuse, a gradual adjustment to 24°C is recommended.

[0196] Scene B

[0197] Environment state vector S t = [28℃, 65%, 45%, 2, 26-30℃];

[0198] The user command "Turn on living room devices" was detected.

[0199] Matching "living room equipment" includes: air conditioner with device ID AC_001, dehumidifier with device ID Dehum_01, and fan with device ID Fan_03;

[0200] The dehumidifier's target humidity is 55%, and the parameter distribution range is P. range Automatic correction; the air conditioner is initially set to fan mode, and switches to cooling mode after the humidity drops to 60%, with the temperature set at 28℃.

[0201] If the dehumidifier malfunctions, the system will automatically increase the fan speed to the highest level to assist ventilation and send a maintenance notification.

[0202] As attached Figure 4 As shown;

[0203] The S500 specifically includes the following steps:

[0204] S510. Based on device priority, insert the control instructions of high-priority devices at the front of the execution queue;

[0205] S520: Calculate the energy consumption threshold for equipment operation based on the energy consumption constraint model. If the adjusted parameters exceed the threshold, trigger an alarm and recommend an energy-saving alternative.

[0206] S530: After the equipment parameters are adjusted, the system collects feedback data from the equipment in real time through sensors to verify the adjustment effect. If the expected results are not achieved, the control strategy is regenerated.

[0207] Specifically, the S510 is as follows:

[0208] The device identifier and operation type are extracted from the standardized control instructions and matched with the device priority configuration table stored in the device registry. The priority configuration table is predefined based on the device function category, with security devices having a higher priority than environmental control devices, and environmental control devices having a higher priority than entertainment devices.

[0209] If multiple devices have the same priority, they are sorted a second time based on the priority value calculated in S230.4, with the larger the value, the earlier the device is sorted.

[0210] Based on the sorting results, control commands for high-priority devices are dynamically inserted into the front position of the execution queue, and the task execution order is updated through the queue management module.

[0211] Specifically, the S520 is as follows:

[0212] Extract the rated power, historical average energy consumption, and current operating mode parameters of the target device from the device registry, and construct the energy consumption prediction function E = f(P). rated ,t,M), where P rated Rated power, t is the expected operating time, and M is the operating mode code;

[0213] Calculate the real-time energy consumption threshold allowed by the device based on the total energy consumption limit and time-of-use pricing strategy set by the user.

[0214] Substitute the expected target parameter values ​​recorded in the standardized control instructions into the energy consumption prediction function. If the predicted energy consumption value exceeds the real-time energy consumption threshold, an energy consumption alarm signal will be triggered.

[0215] Energy-saving alternatives are retrieved based on device function similarity, including:

[0216] Filter low-power device identifiers with similar functions in the device registry;

[0217] Alternatively, it can generate parameter adjustment suggestions, reduce the expected target parameter value by a preset ratio, and then recalculate the energy consumption value.

[0218] Energy-saving alternatives are pushed through a mobile app or voice module, and the system waits for user confirmation.

[0219] Specifically, the S530 is as follows:

[0220] Real-time operating parameters are obtained from the sensor nodes of the target device via IoT protocols; real-time operating parameters include power, temperature, speed, or status flags.

[0221] Compare the real-time operating parameters with the expected target parameter values ​​stored in the standardized control instructions:

[0222] If the deviation between the two is less than or equal to the set percentage, the adjustment is confirmed to be effective and recorded in the equipment operation log;

[0223] If the deviation between the two exceeds the set percentage, an anomaly flag is triggered and the deviation type is extracted.

[0224] Invoke the adaptive correction algorithm based on the deviation type:

[0225] If the deviation is due to parameter adjustment, the target parameter value is recalculated using the gradient descent method, and a new standardized control command is generated.

[0226] If the delay is due to device response, the retry interval for the instruction in the execution queue will be extended.

[0227] Add the revised standardized control instructions to the execution queue and perform the following operations:

[0228] Check if the current number of retries exceeds the preset maximum retry threshold; if it does not exceed the threshold, repeat the process of S530; if it exceeds the threshold, trigger the abnormal termination process, which includes sending an adjustment failure alarm message to the user terminal and recommending manual intervention.

[0229] Example 6

[0230] Scenario A

[0231] Users can set the oven with device ID Oven_01 to a temperature of 200℃ and a running time of 1 hour via the app.

[0232] The oven has a rated power of 3kW, an operating mode of M = high-temperature baking, and an energy consumption function of E = 3kW × 1h = 3kWh; the user's daily energy consumption limit is 2.5kWh.

[0233] The system triggered an alarm; recommended alternative:

[0234] Option 1: Switch to "Quick Baking" mode, operating mode M=180℃, energy consumption function E=2.2kWh.

[0235] Option 2: Shorten the duration to 50 minutes, with an energy consumption function E = 2.5 kWh.

[0236] App notification: "Energy consumption exceeded, quick baking mode recommended (saves 0.8kWh)".

[0237] Scene B

[0238] The air conditioner with device ID AC_001 was set to receive a command of 25°C, but the sensor reported a temperature of 23°C. The difference between the two was 8%, which exceeded the set threshold of 5%.

[0239] Initial adjustment: The target value was adjusted to 24℃. After retrying, the sensor detected a temperature of 23.5℃, with an actual deviation of 2.08%, which is less than the set threshold of 5%. The deviation meets the standard, and the adjustment is successful. No further correction is required.

[0240] If the actual temperature after the first adjustment is 22.5℃, the actual deviation is 6.25%, triggering a second adjustment, and the target value is further reduced to 23℃ until the deviation meets the threshold.

[0241] This specific embodiment is merely an explanation of the present invention and is not intended to limit the invention. After reading this specification, those skilled in the art can make modifications to this embodiment without contributing any inventive step, but such modifications are protected by patent law as long as they are within the scope of the claims of the present invention.

Claims

1. An intelligent home appliance interactive control system based on embedded software, characterized in that: The system operation process specifically includes the following steps: S100. Acquire user input signals, wherein the user input signals include at least one of voice commands, touch operations, gesture actions, or mobile APP control commands. S200. Based on the embedded software architecture, the user input signal is parsed to generate standardized control commands; specifically including the following steps: S210. Receive the original input signals of the voice command, touch operation, gesture action or mobile APP control command, and store the original input signals in a temporary buffer according to their type. S220. Based on the type of the original input signal, call the corresponding parsing module to generate intermediate instructions; S230. Map the intermediate instructions to a preset standardized control instruction format to generate standardized control instructions; specifically including the following sub-steps: S230.1 Extract the operation type, device description information and expected target parameter value from the intermediate instructions, wherein the operation type includes a switch instruction, parameter adjustment instruction or mode switching instruction, and the device description information includes the device name, location identifier or function category; S230.

2. Match the device in the device registry based on the device description information: If the device name is exactly the same as the device identifier in the device registry, then the corresponding unique device code is extracted directly. If the device description information is a location identifier, then the identifiers of all devices under that location are retrieved in the device registry, and the target device is filtered according to the function category; If a match fails, a fuzzy search algorithm is invoked to rematch based on device functional similarity. S230.3 Select a standardized control instruction template based on the operation type: If it is a switch command, the basic control template is invoked, which includes a device identifier and operation type; if it is a parameter adjustment command, the numerical template is invoked, which includes a device identifier, target parameter value, and adjustment step size; if it is a mode switching command, the mode template is invoked, which includes a device identifier and preset mode code. S230.

4. Fill in the fields in the basic control template, numerical template, or pattern template, including: Fill the device unique code matched from the device registry in S230.2 into the device identifier field of the corresponding standardized control instruction template; if the device description information is a location identifier, select the device code with the highest priority under that location according to the function category; If the basic control template is called, fill in the Boolean status parameters according to the on / off state of the intermediate instruction; if the numerical template is called, fill in the target parameter value parsed from the intermediate instruction into the corresponding field; if the mode template is called, match the preset mode code according to the mode name of the intermediate instruction and fill it in; if the numerical template or mode template is called, fill in the expected target parameter value obtained from the parsing of the intermediate instruction into the corresponding field as the reference value for device adjustment. Priority values ​​are calculated using a priority algorithm based on the device's real-time load rate and the user's historical operation frequency. S230.5 Perform syntax validation on the corresponding standardized control instruction template after filling in the code; S230.6 Confirm the verified basic control template, numerical template or pattern template as standardized control instructions; S300 establishes communication connections with multiple smart home appliances through IoT protocols and obtains real-time environmental data and the real-time operating status of smart home appliances; S400, performs joint analysis of user historical behavior data and user input signals; including: S410. Cluster analysis of user historical operation data is performed using a neural network model to identify user preference patterns and device usage habits; specifically: Data preprocessing includes: Extract historical operation data from the user behavior database, including operation timestamps, device identifiers, operation types, and operation parameters; Normalization is performed to convert the operation timestamp into a time period label, and the operation parameters are linearly scaled to the [0,1] interval according to the device's allowable range; User behavior grouping based on K-means clustering algorithm includes: The preprocessed historical operation data is combined into a user operation sequence according to the device identifier and time period label. The user operation sequence consists of operation records within a continuous time window. The optimal number of clusters is determined by the elbow rule, and user behavior cluster groups are generated. Each user behavior cluster group is associated with specific device usage habits. Construct a user-device association matrix, including: Based on the user behavior clusters, the operation frequency and parameter distribution of devices within each cluster are statistically analyzed to generate a user preference pattern matrix. This user preference pattern matrix includes the parameter distribution range P of the target devices. range Store the user preference pattern matrix in the user behavior analysis module; S420: Combines real-time environmental data with equipment operating status to generate control strategies; specifically: Extract real-time environmental data and real-time operating status of smart home appliances obtained from S300, and integrate the parameter distribution range P of the target device in the user preference pattern matrix. range Integrated into a system that includes ambient temperature E temp Ambient humidity E humidity Equipment load rate D load and task queue length Q length Environment state vector S t ; Control policies are generated based on a rules engine, including: If the operating parameters of the current standardized control command are within the parameter distribution range P range Internally, it directly calls the standardized control instructions generated in S200 and adds them to the execution queue; If the operating parameters deviate from the parameter distribution range P range However, if the value is less than the preset threshold, a gradual adjustment command is generated, and a new standardized control command is generated by refilling the basic control template, numerical template, or mode template according to S230.

4. If the operating parameters deviate from the parameter distribution range P range If the preset threshold is exceeded, a user confirmation process is triggered, and a prompt message is pushed via voice or APP. After the user confirms, a standardized control command is generated according to the confirmation parameters. If the equipment load rate D load If the load rate is greater than or equal to the device load rate threshold, non-emergency commands will be suspended, and only security commands will be executed. S500: Adjust the operating parameters of smart home appliances; The S600 returns interactive response information to the user through at least one of the following methods: voice feedback module, touch interface, gesture recognition interface, or mobile APP.

2. The intelligent home appliance interactive control system based on embedded software according to claim 1, characterized in that: The process of calculating the priority value using the priority algorithm includes: If the real-time load rate of the device is greater than or equal to the first threshold, the priority value will be set to the lowest level. If the real-time load rate is less than the second threshold and the user's historical operation frequency exceeds the third threshold, then the priority value is set to the highest level. If the real-time load rate is less than the second threshold and the user's historical operation frequency does not exceed the third threshold, then the priority value is calculated. If the priority value triggers the lowest level, a load alarm will be sent to the user terminal, and a delayed execution or device switching solution will be recommended in the mobile APP.

3. The intelligent home appliance interactive control system based on embedded software according to claim 1, characterized in that: The S500 specifically includes the following steps: S510. Based on device priority, insert the control instructions of high-priority devices at the front of the execution queue; S520: Calculate the energy consumption threshold for equipment operation based on the energy consumption constraint model. If the adjusted parameters exceed the threshold, trigger an alarm and recommend an energy-saving alternative. S530: After the equipment parameters are adjusted, the system collects feedback data from the equipment in real time through sensors to verify the adjustment effect. If the expected results are not achieved, the control strategy is regenerated.

4. The intelligent home appliance interactive control system based on embedded software according to claim 3, characterized in that: Specifically, S510 is as follows: The device identifier and operation type are extracted from the standardized control instructions and matched with the device priority configuration table stored in the device registry; wherein, the priority configuration table is predefined based on the device function category, with security devices having a higher priority than environmental control devices, and environmental control devices having a higher priority than entertainment devices; If multiple devices have the same priority, they are sorted a second time based on the priority value calculated in S230.4, with the larger the value, the earlier the device is sorted. Based on the sorting results, control commands for high-priority devices are dynamically inserted into the front position of the execution queue, and the task execution order is updated through the queue management module.

5. The intelligent home appliance interactive control system based on embedded software according to claim 3, characterized in that: Specifically, S520 is: Extract the rated power, historical average energy consumption, and current operating mode parameters of the target device from the device registry, and construct an energy consumption prediction function. Calculate the real-time energy consumption threshold allowed by the device based on the total energy consumption limit and time-of-use pricing strategy set by the user. Substitute the expected target parameter values ​​recorded in the standardized control instructions into the energy consumption prediction function. If the predicted energy consumption value exceeds the real-time energy consumption threshold, an energy consumption alarm signal will be triggered. Energy-saving alternatives are retrieved based on device function similarity, including: Filter low-power device identifiers with similar functions in the device registry; Alternatively, it can generate parameter adjustment suggestions, reduce the expected target parameter value by a preset ratio, and then recalculate the energy consumption value. The energy-saving alternative is pushed to the user via a mobile app or voice module, and the user confirmation is awaited.

6. The intelligent home appliance interactive control system based on embedded software according to claim 3, characterized in that: Specifically, S530 is: Real-time operating parameters are obtained from the sensor nodes of the target device via IoT protocols; Compare the real-time operating parameters with the expected target parameter values ​​stored in the standardized control instructions: If the deviation between the two is less than or equal to the set percentage, the adjustment is confirmed to be effective and recorded in the equipment operation log; If the deviation between the two exceeds the set percentage, an anomaly flag is triggered and the deviation type is extracted. Invoke the adaptive correction algorithm based on the deviation type: If the deviation is due to parameter adjustment, the target parameter value is recalculated using the gradient descent method, and a new standardized control command is generated. If the delay is due to device response, the retry interval for the instruction in the execution queue will be extended. Add the revised standardized control instructions to the execution queue and perform the following operations: Check if the current number of retries exceeds the preset maximum retry threshold; if it does not exceed the threshold, repeat the process of S530; if it exceeds the threshold, trigger the abnormal termination process, which includes sending an adjustment failure alarm message to the user terminal and recommending manual intervention.

Images