Intelligent scene linkage control platform and device
By combining environmental monitoring and user behavior analysis models with a fault detection module, the complex configuration and network dependency issues of existing intelligent scene linkage control platforms have been resolved, achieving efficient and stable intelligent scene control and improving user experience and system stability.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- KUNSHAN ZHONGYIFENG PHOTOELECTRIC TECH CO LTD
- Filing Date
- 2025-01-13
- Publication Date
- 2026-06-05
AI Technical Summary
Existing intelligent scene linkage control platforms have shortcomings in terms of user configuration complexity, network dependence, fault detection accuracy, and adaptability, resulting in poor user experience and system stability issues.
It employs environmental monitoring models and user behavior analysis models for real-time data monitoring and prediction, combined with fault detection modules and buffer technology, to achieve adaptive adjustment and fault prediction, providing personalized recommendations and multi-layered automatic repair strategies.
It simplifies user configuration, improves system usability and real-time response capabilities, enhances fault identification accuracy and automatic repair capabilities, reduces manual intervention, and ensures effective control of scenarios even without a network connection.
Smart Images
Figure CN120029124B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the technical field of scene linkage control, and discloses an intelligent scene linkage control platform and device. Background Technology
[0002] Common shortcomings of intelligent scene linkage control platforms during their development include: while many platforms offer powerful functions, their setup and configuration processes may be too complex for ordinary users, requiring professional knowledge or a significant amount of time for adjustment and optimization; creating and managing scenes (such as automation rules) may involve complex logic and multiple steps, which may be difficult for ordinary users to master; intelligent control platforms typically rely on network connections, and network interruptions may lead to system failure or inability to control devices properly; software malfunctions or untimely updates may cause system instability, affecting user experience; scene triggering conditions may be overly complex, requiring multiple configurations, making them difficult for users to set and manage; sensor data may have latency, affecting real-time performance and accuracy; intelligent recommendations, limited by the quality and quantity of data, are difficult to achieve complete accuracy; self-learning recommendations may require a large amount of data to train the model and are easily affected by abnormal data; monitoring and maintenance, real-time display and monitoring require efficient system support, otherwise there may be delays; fault detection models may have difficulty identifying all types of faults, leading to missed or false alarms; and the limitations of automatic repair functions may not be able to handle complex fault situations, still requiring manual intervention.
[0003] For example, Chinese patent application CN116700035A discloses a smart scene linkage control method based on a smart home platform, which makes user settings for smart scenes more flexible and smart scene execution more efficient. The method includes: A) Creating and storing smart scenes on a cloud platform: the smart scene creation includes creating a trigger scene, creating a linkage execution scene, and configuring the smart scene's effective time; B) The cloud platform initiates device trigger detection tasks and timed trigger detection tasks through parallel threads to perform trigger detection; C) The cloud platform initiates linkage scene execution task generation and linkage scene execution tasks through parallel threads to execute the linkage scene. This invention is applicable to smart home scenarios.
[0004] However, the aforementioned patents have the following drawbacks: they do not provide specific scene recognition logic and methods, fail to monitor and control environmental parameters and target detection within the scene in real time, cannot handle scene control without a network, cannot adaptively adjust to faults within the scene, do not set up a fault detection buffer, which greatly increases the losses caused by all types of faults that are difficult to identify, and the connection to the cloud platform is prone to false alarms and missed alarms. Summary of the Invention
[0005] To address the aforementioned technical problems, the main objective of this invention is to provide a control platform and device for intelligent scene linkage, wherein the intelligent scene linkage control platform includes:
[0006] The analysis module includes an environmental monitoring model and a user behavior analysis model, wherein the environmental monitoring model is used to predict environmental changes, and the user behavior analysis model uses target detection data to identify scenarios and predict user behavior.
[0007] The user interface module includes a user interface, a feedback unit, a scene management unit, an intelligent recommendation unit, a monitoring and maintenance unit, and a security and privacy unit.
[0008] The fault detection module uses a fault model to detect faults in a control platform that enables intelligent scene linkage.
[0009] As a preferred embodiment of the intelligent scene linkage control platform of the present invention, wherein:
[0010] The environmental monitoring model uses a built-in data cleaning unit to filter out useless data from the environmental data, and then performs real-time monitoring and prediction of the environmental data within the filtered scene.
[0011] The adaptive reliability weight is dynamically adjusted to enable the environmental monitoring model to focus on specific environmental parameters.
[0012] The environmental data includes temperature data, humidity data, and light intensity data;
[0013] The specific environmental parameters include data on imbalances in environmental parameters within the scene.
[0014] As a preferred embodiment of the intelligent scene linkage control platform of the present invention, wherein:
[0015] The number of labels for the detected targets within the scene is established. A cross-optimization function is used to optimize the probabilities of the left and right positions of the detected target labels, so that the network distribution focuses on the label values. The calculation expression of the cross-optimization function is as follows:
[0016] ;
[0017] in, The network distribution focus value is optimized for the cross-optimization function. Let p be the left-hand position value of the target label count detected by the user behavior analysis model, and p be the correct target label count detected by the user behavior analysis model. The right position value of the target label detected by the user behavior analysis model. The probability to the left of the correct target label quantity is used to focus the network distribution. The probability of focusing the network distribution on the right side of the correct target label quantity;
[0018] Predict environmental data and user behavior using regression models.
[0019] As a preferred embodiment of the intelligent scene linkage control platform of the present invention, wherein:
[0020] The scene management unit includes scene creation and editing, and scene triggering conditions;
[0021] Intelligent recommendations include personalized recommendations and self-learning recommendations;
[0022] Personalized recommendations are used to suggest scenario settings based on users' historical behavior and preferences;
[0023] By using environmental monitoring models and user behavior analysis models to identify user behavior patterns and preferences, customized scenario recommendations can be provided.
[0024] Self-learning recommendations automatically learn user behavior patterns and environmental changes to optimize scenario settings and recommendations.
[0025] As a preferred embodiment of the intelligent scene linkage control platform of the present invention, wherein:
[0026] The fault detection model includes a data buffer, a buffer simulation unit, and a buffer correction unit;
[0027] The data buffer includes a buffer range and a fault buffer;
[0028] The buffer simulation unit includes fault models, fault simulation applications, and fault assessment.
[0029] The buffer correction unit is used for fault assessment.
[0030] As a preferred embodiment of the intelligent scene linkage control platform of the present invention, wherein:
[0031] The buffer scope defines the system components and functions that need to be monitored, and sets the buffer capacity to cope with different types and sizes of failures;
[0032] The fault model uses environmental monitoring models and user behavior analysis models to monitor fault values as data foundation, and establishes a buffer fault monitoring neural convolutional network model to monitor and predict scene environmental parameters and abnormal user behavior parameters in real time, and output scene fault values.
[0033] The scenario fault value input fault simulation application creates a virtual environment to simulate the scenario fault value and records the response of the intelligent scenario linkage control platform.
[0034] As a preferred embodiment of the intelligent scene linkage control platform of the present invention, wherein:
[0035] The fault assessment receives the response from the intelligent scene linkage control platform, predicts the system state by training the fault assessment model, and determines whether a fault will occur. If the scene fault value simulates a fault, the fault buffer parameters, system environment parameters, or user behavior parameters are adjusted.
[0036] A control device for intelligent scene linkage, comprising:
[0037] Environmental monitoring interface, environmental parameter control button, environmental change control button, target recognition switch button, target detection parameter adjustment button, fault warning light, fault alarm light, equipment normal indicator light, equipment fault monitoring interface, target recognition and monitoring interface.
[0038] As a preferred embodiment of the intelligent scene linkage control device of the present invention, wherein:
[0039] The environmental monitoring interface, the equipment fault monitoring interface, and the target identification and monitoring interface are located on the same plane, with the environmental monitoring interface and the equipment fault monitoring interface located on opposite sides of the target identification and monitoring interface, respectively.
[0040] The environmental parameter control button and the environmental change control button are located at the bottom of the environmental monitoring interface;
[0041] The target recognition switch button and the target detection parameter adjustment button are located at the bottom of the target recognition and monitoring interface;
[0042] The fault warning light, fault alarm light, and normal equipment indicator light are located at the bottom of the equipment fault monitoring interface.
[0043] As a preferred embodiment of the intelligent scene linkage control device of the present invention, wherein:
[0044] The environment parameter control button is used to manually control the scene environment parameters;
[0045] The environment change control button is used to manually switch scenes;
[0046] The target recognition switch button is used to manually switch the target for scene target detection;
[0047] The target detection parameter adjustment button is used to manually control the target detection parameters;
[0048] The fault warning light is used to respond when the platform reaches a critical fault value;
[0049] The fault alarm light is used to sound an alarm when a platform malfunctions;
[0050] The equipment normal indicator light indicates that the platform and equipment are operating normally.
[0051] The beneficial effects of this invention are:
[0052] This invention reduces the complexity of user configuration and improves system usability through preset templates and dynamic adjustments. Data caching and edge computing reduce latency, enabling the system to respond more quickly and accurately to real-time events. Combining multiple data sources and adaptive algorithms improves the accuracy and stability of intelligent recommendations, meeting personalized needs. Hybrid detection methods improve the accuracy of fault identification, and multi-layer automatic repair strategies ensure timely fault handling, reducing the need for manual intervention. Fault detection buffers greatly reduce the difficulty in identifying all types of faults, which could lead to missed or false alarms. Faults can be adaptively adjusted.
[0053] This invention generates adaptive adjustment parameters for network-free operation by collecting data parameters in real time, thus enabling scene control in the absence of a network. Attached Figure Description
[0054] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:
[0055] Figure 1 This is a system composition diagram of a control platform for intelligent scene linkage according to the present invention;
[0056] Figure 2 This is a topology diagram of a control device for intelligent scene linkage according to the present invention;
[0057] Figure 3 This is a flowchart illustrating the optimization of the network distribution focus value using the cross-optimization function of this invention.
[0058] Figure 4 This is a flowchart illustrating how the present invention achieves correct target location identification.
[0059] Attached reference numerals: 1. Environmental monitoring interface; 2. Environmental parameter control button; 3. Environmental change control button; 4. Target recognition switching button; 5. Target detection parameter adjustment button; 6. Fault warning light; 7. Fault alarm light; 8. Equipment normal indicator light; 9. Equipment fault monitoring interface; 10. Target recognition and monitoring interface. Detailed Implementation
[0060] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
[0061] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.
[0062] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.
[0063] Example 1
[0064] like Figure 1 As shown, a control platform for intelligent scene linkage includes:
[0065] The data acquisition module is used to collect environmental data and user behavior data in real time.
[0066] The data acquisition module includes temperature sensors, humidity sensors, light sensors, motion sensors, etc.
[0067] Temperature sensors collect scene temperature data;
[0068] Humidity sensors collect scene humidity data;
[0069] A light sensor collects scene lighting data;
[0070] Image sensors acquire scene image data;
[0071] Environmental data includes temperature data, humidity data, and light intensity data;
[0072] User behavior data includes target detection data identified from image data;
[0073] The processing module includes a data storage unit and a data preprocessing unit;
[0074] Data storage is used to store collected data and historical records, and is done using a database or data warehouse.
[0075] Data preprocessing is used to clean, standardize, and format environmental and user behavior data for further analysis.
[0076] The analysis module includes an environmental monitoring model and a user behavior analysis model;
[0077] The analysis module is used to analyze and predict data;
[0078] Environmental monitoring models are used to predict environmental changes, such as temperature changes and energy consumption.
[0079] Since environmental changes such as temperature changes and energy consumption are collected as long-term data, environmental monitoring models filter out useless data in the environmental data and adaptively adjust the reliability weights to monitor and predict the environment.
[0080] The environmental monitoring model uses a built-in data cleaning unit to filter out useless data from the environmental data. It then extracts features from the filtered environmental data, such as temperature, humidity, and light intensity. Initial weights are set through convolution operations to obtain feature vectors for temperature, humidity, and light intensity. The initial weights are dynamically adjusted based on the deviation between the environmental data output by the environmental monitoring model and the actual values until the difference between the environmental data output by the environmental monitoring model and the actual data in the scene is minimized. This completes the training of the environmental monitoring model. The trained environmental monitoring model is then used to monitor and predict environmental parameters in the scene in real time.
[0081] After obtaining the predicted values of environmental parameters in the scene, the environment is judged. If the predicted values of environmental parameters show special environmental data, the temperature, humidity and light are compensated through the scene linkage system respectively. If the predicted values of environmental parameters show normal environmental data, the scene linkage system does not output any control commands.
[0082] Dynamically adjust the weights of environmental data such as temperature, humidity, and light intensity to enable the environmental monitoring model to focus on specific environmental parameters;
[0083] Environmental data includes temperature data, humidity data, and light intensity data;
[0084] Special environmental parameters include data on imbalances in environmental parameters within the scene.
[0085] User behavior analysis models use target detection data to identify scenarios and predict user behavior;
[0086] like Figure 3 As shown, the steps of optimizing the network distribution focus value using the cross-optimization function include: marking the approximate rectangular region of the object's position in the image with an initial bounding box; detecting the left and right position values of the label value by the user behavior analysis model; calculating the probability that the left position value is the left position of the target label value and the right position value is the right position of the target label value using the cross-optimization function; adjusting the bounding box position by direction with a pre-set correct target label value as a reference; and finally calculating the error weight between the actual position and the correct position to adjust the initial bounding box.
[0087] The system establishes the label values for detected targets within the scene. Due to the uneven label density and unclear object boundaries, the network distribution cannot focus on the vicinity of the label values. Therefore, a cross-optimization function is used to optimize the probabilities of the left and right positions of the detected labels. A logarithmic function is used to transform the probability values, mapping them to [-∞, 0] to avoid overfitting, ensure numerical stability, and guarantee a smooth, continuous gradient. By estimating the probabilities that the left and right positions of the detected bounding box are correct, the network distribution is focused on the vicinity of the label values. The cross-optimization function is calculated as follows:
[0088] ;
[0089] in, The network distribution focus value is optimized for the cross-optimization function. Let p be the left-hand position value of the target label count detected by the user behavior analysis model, and p be the correct target label count detected by the user behavior analysis model. The right position value of the target label detected by the user behavior analysis model. The probability to the left of the correct target label quantity is used to focus the network distribution. The probability of focusing the network distribution on the right side of the correct target label quantity;
[0090] Furthermore, the left position value of the label is the left position coordinate of the correct target bounding box, and the right position value is the right position coordinate of the correct target bounding box. The bounding box is a rectangular area used to mark the position of an object in the image.
[0091] Furthermore, the location (bounding box) of the object and the category to which the object belongs are determined, and the location of the detected target label is optimized so that the detection result more accurately reflects the actual location of the target. The actual location of the target includes the left position and the right position. The label value difference represents the difference between the detected location and the actual location. The label probability includes the probability that the network distribution focuses on the left side of the correct target label and the probability that the network distribution focuses on the right side of the correct target label. The label probability is used to describe the probability that the detected location is the correct location. The label difference multiplied by the label probability gives the directional error weight.
[0092] Furthermore, the direction is used to describe whether the difference in label values is positive or negative. If it is positive, it is the left position, and if it is negative, it is the right position. The error weight is used to describe the degree of influence of the label probability on the error. By reflecting the degree of influence of the label probability on the error, the product of the uncertainty of the target detection position and the actual error is reflected.
[0093] Predicting environmental data and user behavior using regression models;
[0094] The control module includes equipment control and action execution;
[0095] Equipment control uses interfaces to control various devices, such as lights, air conditioners, and curtains.
[0096] The system controls the equipment based on the environmental monitoring model and executes corresponding actions or scene settings based on the user behavior analysis model.
[0097] The user interface module includes a user interface, a feedback unit, a scene management unit, an intelligent recommendation unit, a monitoring and maintenance unit, and a security and privacy unit.
[0098] The user interface includes mobile applications, web interfaces, voice control, etc., allowing users to view and manage smart scenarios;
[0099] The feedback unit is used to provide real-time feedback and notifications, allowing users to understand the current system status and events;
[0100] The scene management unit includes scene creation and editing, and scene triggering conditions;
[0101] User Scene Creation and Editing: Create, edit, and delete various scenes, such as "Away from Home Mode" and "Home Mode".
[0102] Set scene trigger conditions by setting trigger conditions, such as time, sensor data thresholds, user behavior, etc.
[0103] For example, set specific times or periodic times (such as daily or weekly) to trigger scenarios, or trigger scenarios based on calendar events or appointments, such as starting "Morning Mode" at 7 a.m. on weekdays.
[0104] Utilize GPS or Wi-Fi location to trigger scenarios. For example, automatically activate "away mode" when a user is more than 5 kilometers away from home, or trigger scenarios when a user behavior analysis model monitors a user entering or leaving a specific geographic area.
[0105] The system monitors scene data based on user behavior analysis models to trigger scenarios. For example, it automatically activates "Home Mode" when a door is detected to be open.
[0106] The environmental monitoring model monitors changes in battery levels, such as triggering specific scenarios when a phone's battery is low.
[0107] Intelligent recommendations include personalized recommendations and self-learning recommendations;
[0108] Personalized recommendations are used to suggest suitable scenario settings based on users' historical behavior and preferences;
[0109] By using environmental monitoring models and user behavior analysis models to identify user behavior patterns and preferences, the system can provide customized scenario recommendations. For example, if a user often dims the lights at night, the system will recommend similar night mode settings;
[0110] Users select or set preferences, and the system recommends scenarios that meet their needs. For example, if a user marks themselves as "preferring a warm environment," the system will recommend scenarios with corresponding temperature settings.
[0111] Self-learning recommendations automatically learn user behavior patterns and environmental changes to optimize scenario settings and recommendations.
[0112] The platform analyzes user behavior in real time using a user behavior analysis model and monitors environmental changes using an environmental monitoring model to automatically adjust recommendations. For example, if the system detects that a user frequently plays music within a specific time period, it will automatically recommend music-related scene settings.
[0113] The platform uses machine learning algorithms and adaptive coverage of new user behaviors and habits to optimize recommendations by adapting to new user behavior patterns. For example, if a user starts working from home, the system might recommend "work mode" scenarios.
[0114] The monitoring and maintenance unit is used to display the status of each device and environmental data in real time, detect equipment faults or abnormalities, and provide maintenance suggestions or automatic repairs.
[0115] The fault detection module uses a fault model to detect faults in a control platform for intelligent scene linkage.
[0116] The fault detection model includes a data buffer, a buffer simulation unit, and a buffer correction unit;
[0117] The data buffer includes a buffer range and a fault buffer;
[0118] The buffer range defines the system components and functions that need to be monitored, such as the real-time status of devices (including the startup, operation, and shutdown status of devices, as well as the operating parameters of key components), network connectivity (such as network bandwidth, packet loss rate, connection stability, etc.), data processing flow (such as the efficiency and accuracy of data acquisition, transmission, storage, and processing, etc.), and many other aspects.
[0119] The fault buffer focuses on setting the size of the buffer to deal with different types and scales of faults. Its capacity setting needs to take into account factors such as the complexity of the system, the frequency of data generation, and the severity of the impact of potential faults on the system.
[0120] The buffer simulation unit includes fault models, fault simulation applications, and fault assessment.
[0121] The fault model uses the fault values monitored by the environmental monitoring model and the user behavior analysis model as the basic data input. The environmental monitoring model is used to collect and analyze scene environmental parameters in real time, such as the changes of environmental indicators such as temperature, humidity, and air quality in the system operating environment. The user behavior analysis model obtains abnormal behavioral parameters generated during the interaction between the user and the system, such as whether the user's operation frequency, operation path, operation time, etc. deviate from the normal mode.
[0122] Furthermore, after weighted data fusion of preprocessed environmental parameters and user behavior data, fault features are extracted, and correlation analysis is used to obtain the correlation between each fault feature and whether a fault has occurred. Based on the correlation, it is determined whether the fault features need to be retained.
[0123] Furthermore, a tree-like decision structure is constructed, and branch judgments are made based on different values of fault characteristics to output whether a fault exists. For example, when judging whether a fault exists in a scenario, the branch is first made based on the temperature characteristics in the scenario. If the temperature is too high, the judgment is further subdivided based on other characteristics such as the air conditioner usage rate in the scenario to gradually determine the fault situation.
[0124] The scenario fault values output by the fault model are input into the fault simulation application;
[0125] Furthermore, the fault simulation application creates a virtual environment, using the received scene fault values as the basic data, to simulate the fault scenarios of the scene fault values. Through the virtual environment, the scene linkage system completely reproduces the various states and conditions when the fault occurs, and records the response of the intelligent scene linkage control platform when facing the simulated fault.
[0126] The buffer correction unit is used to receive the results of fault assessment and to correct, optimize, and compensate for system environmental parameters.
[0127] The fault assessment receives the response from the intelligent scene linkage control platform. By training the fault assessment model, it judges whether changes in scene environment data and behavioral data will cause scene faults. If the scene fault value input fault assessment calculation is abnormal, the fault buffer parameters are adjusted until changes in scene environment data and behavioral data do not cause faults. Then, it outputs instructions to adjust system environment parameters or user behavior parameters to avoid actual system faults.
[0128] The fault assessment model inputs scenario fault values into the fault model and uses the fault model recall rate or mean square error to determine whether the scenario fault values will cause platform faults.
[0129] Based on the latest system data and fault records, the model is updated regularly, the actual effects of system operation and fault buffering are collected, and the model and algorithm are optimized based on feedback information.
[0130] Security and privacy include data encryption and access control.
[0131] Data encryption is used to ensure the security of data during transmission and storage.
[0132] Access control is used to set permissions for different users in order to protect system security and privacy.
[0133] Example 2
[0134] like Figure 2 As shown: A control device for intelligent scene linkage, comprising:
[0135] Environmental monitoring interface 1. Environmental parameter control button 2. Environmental change control button 3. Target recognition switch button 4. Target detection parameter adjustment button 5. Fault warning light 6. Fault alarm light 7. Equipment normal indicator light 8. Equipment fault monitoring interface 9. Target recognition and monitoring interface 10.
[0136] The environmental monitoring interface 1, the equipment fault monitoring interface 9, and the target identification and monitoring interface 1 are located on the same plane, with the environmental monitoring interface 1 and the equipment fault monitoring interface 9 located on opposite sides of the target identification and monitoring interface 1, respectively.
[0137] The environmental parameter control button 2 and the environmental change control button 3 are located at the bottom of the environmental monitoring interface 1;
[0138] The target recognition switch button 4 and the target detection parameter adjustment button 5 are located at the bottom of the target recognition and monitoring interface 1;
[0139] The fault warning light 6, the fault alarm light 7, and the equipment normal indicator light 8 are located at the bottom of the equipment fault monitoring interface 9;
[0140] Furthermore, the environmental parameter control button 2 is used to manually control the scene environmental parameters;
[0141] The environment change control button 3 is used to manually switch scenes;
[0142] Target recognition switch button 4 is used to manually switch the target for scene target detection;
[0143] The target detection parameter adjustment button 5 is used to manually control the target detection parameters;
[0144] Fault warning light 6 is used to respond when the platform reaches a critical fault value;
[0145] The fault alarm light 7 is used to issue an alarm when a platform malfunction occurs;
[0146] The device normal indicator light 8 is used to indicate that the platform and equipment are operating normally.
[0147] Example 3
[0148] like Figure 4 As shown, the triangular area represents the correct target location in the scene for scene linkage recognition, and the rectangular area represents the recognition area. Figure 4 In step 1, the bounding box is identified to the right of the correct target position. Therefore, it is adaptively adjusted to the right. In step 2, the bounding box is identified to the left of the correct target position. This process is continuously corrected until step 4, which identifies the correct target position, is completed. After the correct target position is identified, the platform monitors and triggers scene linkage. Through the intelligent scene linkage control system and control equipment, combined with the environmental monitoring model to monitor scene environmental parameters and user personalization factors, the scene target is controlled.
[0149] It is important to note that the constructions and arrangements of this application shown in several different exemplary embodiments are merely illustrative. Although only two embodiments are described in detail in this disclosure, those who consult this disclosure will readily understand that many modifications are possible without substantially departing from the novel teachings and advantages of the subject matter described herein. For example, variations in the size, dimensions, structure, shape, and proportions of various elements, as well as parameter values (e.g., temperature, pressure, etc.), mounting arrangements, use of materials, color, orientation, etc. For instance, an element shown as integrally formed may be composed of multiple parts or elements, the position of elements may be inverted or otherwise altered, and the nature or number or position of discrete elements may be changed or altered. Therefore, all such modifications are intended to be included within the scope of the invention. The order or sequence of any process or method steps may be changed or rearranged according to alternative embodiments. Any "device plus function" clause is intended to cover the structure performing the function described herein, and not only structural equivalents but also equivalent structures. Other substitutions, modifications, alterations, and omissions may be made in the design, operation, and arrangement of the exemplary embodiments without departing from the scope of the invention. Therefore, the present invention is not limited to the specific embodiments, but extends to various modifications that still fall within the scope of the appended claims.
[0150] Furthermore, in order to provide a concise description of exemplary embodiments, not all features of actual embodiments (i.e., those features that are not relevant to the currently considered best mode for carrying out the invention, or those features that are not relevant to implementing the invention) may be omitted.
[0151] It should be understood that numerous specific implementation decisions can be made during the development of any practical implementation, such as in any engineering or design project. Such development efforts may be complex and time-consuming, but for those of ordinary skill in the art who benefit from this disclosure, the development effort will be a routine task in design, manufacturing, and production without requiring extensive experimentation.
[0152] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.
Claims
1. A control platform for intelligent scene linkage, characterized in that, include: The analysis module includes an environmental monitoring model and a user behavior analysis model, wherein the environmental monitoring model is used to predict environmental changes, and the user behavior analysis model uses target detection data to identify scenarios and predict user behavior. The user interface module includes a user interface, a feedback unit, a scene management unit, an intelligent recommendation unit, a monitoring and maintenance unit, and a security and privacy unit. The fault detection module enables fault detection of a control platform for intelligent scene linkage. The fault detection module includes a data buffer, a buffer simulation unit, and a buffer correction unit; The data buffer includes a buffer range and a fault buffer; The buffer simulation unit includes fault models, fault simulation applications, and fault assessment. Fault features are extracted through fault models, and the correlation between each fault feature and whether a fault has occurred is obtained through correlation analysis. Based on the correlation, it is determined whether the fault feature should be retained. Construct a tree-like decision structure, perform branch judgments based on different values of fault characteristics, and output whether a fault exists; The fault model uses environmental monitoring models and user behavior analysis models to monitor fault values as data foundation, and establishes a buffer fault monitoring neural convolutional network model to monitor and predict scene environmental parameters and abnormal user behavior parameters in real time, and output scene fault values. The scenario fault values output by the fault model are input into the fault simulation application; The fault simulation application creates a virtual environment and uses the received scenario fault values as the base data to simulate the fault scenarios of the scenario fault values. The fault assessment receives the response from the intelligent scene linkage control platform. By training the fault assessment model, it determines whether changes in scene environment data and behavioral data will cause scene faults. If the scene fault value input fault assessment calculation is abnormal, the fault buffer parameters are adjusted until changes in scene environment data and behavioral data do not cause faults. Then, the system environment parameters or user behavior parameters adjustment instructions are output. The buffer correction unit is used to receive the results of fault assessment and to correct, optimize, and compensate the system environmental parameters.
2. The intelligent scene linkage control platform according to claim 1, characterized in that: The environmental monitoring model uses a built-in data cleaning unit to filter out useless data in the environmental data. It then monitors and predicts the environmental data in the scene in real time. After obtaining the predicted values of the environmental parameters in the scene, it judges the environment. If the predicted values of the environmental parameters show special environmental data, the scene linkage system compensates for temperature, humidity and light respectively. If the predicted values of the environmental parameters show normal environmental data, the scene linkage system does not output any control commands. The environmental data includes temperature data, humidity data, and light intensity data; The special environmental data includes data on imbalances in environmental parameters within the scene.
3. The intelligent scene linkage control platform according to claim 2, characterized in that: The number of labels for the detected targets within the scene is established. A cross-optimization function is used to optimize the probabilities of the left and right positions of the detected target labels, so that the network distribution focuses on the label values. The calculation expression of the cross-optimization function is as follows: ； in, The network distribution focus value is optimized for the cross-optimization function. Let p be the left-hand position value of the target label count detected by the user behavior analysis model, and p be the correct target label count detected by the user behavior analysis model. The right position value of the target label detected by the user behavior analysis model. The probability to the left of the correct target label quantity is used to focus the network distribution. The probability of focusing the network distribution on the right side of the correct target label quantity; Predict environmental data and user behavior using regression models.
4. The intelligent scene linkage control platform according to claim 3, characterized in that: The scene management unit includes scene creation, editing, and scene triggering conditions; The intelligent recommendation unit includes personalized recommendations and self-learning recommendations; Personalized recommendations are used to suggest scenario settings based on users' historical behavior and preferences; By using environmental monitoring models and user behavior analysis models to identify user behavior patterns and preferences, customized scenario recommendations can be provided. Self-learning recommendations automatically learn user behavior patterns and environmental changes to optimize scenario settings and recommendations.
5. A control device for intelligent scene linkage, applied to the intelligent scene linkage control platform described in any one of claims 1-4, characterized in that, include: Environmental monitoring interface (1), environmental parameter control button (2), environmental change control button (3), target identification switching button (4), target detection parameter adjustment button (5), fault warning light (6), fault alarm light (7), equipment normal indicator light (8), equipment fault monitoring interface (9), target identification and monitoring interface (10).
6. The intelligent scene linkage control device according to claim 5, characterized in that: The environmental monitoring interface (1), the equipment fault monitoring interface (9), and the target identification and monitoring interface (10) are located on the same plane, and the environmental monitoring interface (1) and the equipment fault monitoring interface (9) are located on opposite sides of the target identification and monitoring interface (10); The environmental parameter control button (2) and the environmental change control button (3) are located at the bottom of the environmental monitoring interface (1); The target recognition switch button (4) and the target detection parameter adjustment button (5) are located at the bottom of the target recognition and monitoring interface (10); The fault warning light (6), the fault alarm light (7), and the normal equipment indicator light (8) are located at the bottom of the equipment fault monitoring interface (9).
7. The intelligent scene linkage control device according to claim 6, characterized in that: The environment parameter control button (2) is used to manually control the scene environment parameters; The environment change control button (3) is used to manually switch scenes; The target recognition switch button (4) is used to manually switch the target of scene target detection; The target detection parameter adjustment button (5) is used to manually control the target detection parameters; The fault warning light (6) is used to respond when the platform reaches a fault threshold; The fault alarm light (7) is used to issue an alarm when a platform malfunctions; The equipment normal indicator light (8) is used to indicate that the platform and equipment are operating normally.