Intelligent equipment control method and system based on machine learning

Abstract

The invention discloses an intelligent equipment control method and system based on machine learning, and relates to the technical field of intelligent equipment control. S2, model training; s3, predicting and controlling; s4, interaction control; and S5, performing feedback adjustment. According to the intelligent equipment control method and system based on machine learning, user behaviors and equipment use data are analyzed and learned by utilizing a machine learning technology, so that behavior habits of the user are predicted, intelligent control on the intelligent equipment is realized, user operation steps are simplified, the use difficulty of the user is reduced, and the user experience is improved. The method can also automatically adapt to behavior habits of the user, adjust a control strategy according to behavior changes of the user, optimize operation modes and parameters of equipment, keep synchronization with user demands, make equipment preparation for the user in advance based on predicted user behaviors, improve the use efficiency of the equipment, improve user experience, and improve user experience. And meanwhile, interconnection, intercommunication and cooperative control between the intelligent devices are promoted.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent device control, and in particular to an intelligent device control method and system based on machine learning. Background Art

[0002] Under the current trend of technological development, with the rapid development of artificial intelligence and machine learning technology, smart devices have penetrated into every aspect of our lives, such as smart homes, smart wearable devices, smart medical care, etc., making our lives more convenient and efficient. While these devices facilitate our lives, they also bring challenges in terms of device control complexity. How to effectively manage and control these devices so that they can work together and meet the personalized needs of users is the focus of existing technologies.

[0003] However, in actual use, due to the uncertainty of user behavior, existing control technology can only control the equipment according to user instructions or fixed programs, resulting in cumbersome user operation steps and poor intelligence of equipment control. In addition, since it is impossible to generate corresponding control strategies based on user behavior habits, it is impossible to prepare equipment for users in advance, and it is impossible to meet user usage needs well, affecting user experience.

[0004] Therefore, there is an urgent need to improve this shortcoming. The present invention studies and improves the existing structure and deficiencies, and provides a smart device control method and system based on machine learning. Summary of the invention

[0005] The purpose of the present invention is to provide a smart device control method and system based on machine learning to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions: In a first aspect, the present invention provides a method for controlling an intelligent device based on machine learning, comprising the following steps: S1. Data acquisition: Collect user device usage data through device sensors or user feedback, preprocess and extract features from the collected data, and obtain a training data set; S2, model training: select machine learning algorithms and models according to specific application scenarios and requirements, introduce training data sets, and use machine learning to train models; S3, prediction and control: After the model training is completed, the model is used to predict the user's behavior habits, and the corresponding control strategy is generated according to the prediction results to control the device; S4, interactive control: users view the operating status of the device through the interactive interface, receive system recommendations and prompts, and adjust the device settings according to their needs; S5. Feedback adjustment: Feedback interaction data to the machine learning model to generate new prediction results to adjust the control strategy, optimize the operating mode and parameters of the equipment, and keep pace with user needs.

[0007] Furthermore, in step S1, the user's device usage data is divided into real-time operation data of the device and user's behavior data, including information such as device usage time, usage frequency, and usage scenario.

[0008] Furthermore, in step S1, feature extraction specifically includes: Statistical features: including but not limited to mean, variance, maximum and minimum values, and median, which can well describe the distribution and central tendency of data and are often used for feature extraction of numerical data; Frequency domain features: refers to converting data into the frequency domain for feature extraction by performing Fourier transform and other operations on the data. It can well describe the periodicity and frequency distribution characteristics of the data and is often used in the fields of signal processing and audio processing. Filter characteristics: refers to extracting the frequency response and time domain characteristics of data by designing and applying filters, which can well capture the local characteristics and frequency components of data, and is often used in the fields of image processing and signal processing; Wavelet transform features: By performing wavelet transform on data, the characteristic representation of data at different scales and frequencies is obtained, which can well describe the local characteristics and frequency characteristics of data and is often used in the fields of signal processing and image processing; Principal component analysis features: The original data is mapped to a low-dimensional space through linear transformation to obtain the principal component features of the data, which can well describe the main change direction and correlation of the data and is often used for data compression; Independent component analysis features: By performing independent component analysis on the data, the independent component features of the data are obtained, which can well describe the independence and mixed characteristics of the data and is often used in the fields of signal processing and medical image processing; Dictionary learning features: By learning the sparse representation dictionary of the data, the sparse coding features of the data are obtained, which can well describe the sparsity and local characteristics of the data and are often used in the fields of image processing and pattern recognition.

[0009] Furthermore, in step S2, the machine learning algorithms and models include but are not limited to linear regression, logistic regression, decision tree, random forest, support vector machine, neural network, and the loss function of machine learning is defined as follows: in, is the number of training samples, is the number of features, is the model parameter vector, is the design matrix, The target value of samples, is the regularization coefficient; And the model parameter vector The optimal solution of is defined as follows: in, is the learning rate, is the loss function with respect to The partial derivatives of the parameters, It is Furthermore, in step S2, the specific process of machine learning training model is as follows: Data training: Use the training data set to train the selected model, and use optimization algorithms such as gradient descent to iteratively update the model parameters to minimize the loss function; Model evaluation: During the model training process, the model needs to be evaluated to determine the performance of the model. The evaluation indicators include but are not limited to precision, recall rate, and F1 value; Model parameter adjustment: According to the results of model evaluation, the model parameters are adjusted to further improve the performance of the model; Model preservation and deployment: Save the trained model and deploy it to actual applications for prediction and classification tasks.

[0010] Furthermore, in step S3, the control strategy includes but is not limited to adjusting the operation mode of the device, optimizing the operation parameters of the device, and triggering the collaborative work between devices. The operation mode of the device includes a motion mode, and the motion mode at least includes a moving distance. and moving angle The instruction parameters; The moving distance The command parameters are defined as follows: The moving angle The command parameters are defined as follows: in, Indicates the moving distance, Indicates the target location, Indicates the current location. In a second aspect, the present invention provides a smart device control system based on machine learning, which is applied to a server, and the system includes: A data acquisition module that collects device and user data and performs preprocessing and feature extraction so that subsequent machine learning algorithms can understand and use it; A machine learning module, wherein the machine learning module selects a machine learning algorithm to learn and analyze the collected data to generate a prediction model; A model prediction module, which uses a model to learn user behavior patterns and device usage rules to predict future device needs and user preferences; A strategy generation module, wherein the strategy generation and execution module generates a corresponding control strategy according to the prediction results of the machine learning model; A policy execution module, wherein the policy execution module controls the intelligent device through a device driver and an actuator according to the control policy; A user interaction module allows the user to view the operating status of the device, adjust the settings of the device, receive system recommendations and prompts, and provide feedback and comments to help optimize and improve the system.

[0011] Furthermore, the data acquisition module specifically includes: Data collection unit: collects real-time operation data of the equipment and user behavior data through sensors and data acquisition devices; Preprocessing unit: preprocess the collected data, including but not limited to data cleaning, data integration, data transformation, and data reduction; Feature extraction unit: selects, converts and combines relevant information from raw data, uses computers or other algorithms to analyze and transform data, and extracts representative and distinctive features.

[0012] Further, including: At least one memory for storing computer instructions; At least one processor in communication with the memory, wherein when the at least one processor executes computer instructions, the at least one processor causes the system to execute: a data acquisition module, a machine learning module, a model prediction module, a strategy generation module, a strategy execution module, and a user interaction module.

[0013] Furthermore, a computer program is stored in the memory, and when the computer program is executed by the processor, the intelligent device control method based on machine learning is implemented.

[0014] The present invention provides a method and system for controlling an intelligent device based on machine learning, which has the following beneficial effects: The present invention uses machine learning technology to analyze and learn user behavior and device usage data, so as to predict user behavior habits and realize intelligent control of smart devices. It not only simplifies user operation steps and reduces user difficulty of use, but also can automatically adapt to user behavior habits, adjust control strategies according to user behavior changes, optimize device operation modes and parameters, keep pace with user needs, and prepare devices for users in advance based on predicted user behavior, thereby improving device usage efficiency and user experience. In addition, due to the combination of machine learning technology, the loss function definition and motion mode parameter selection optimization of machine learning after optimization make the control method more precise, efficient and intelligent, while promoting the interconnection and collaborative control between smart devices, thereby further improving device usage efficiency and user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A schematic diagram of the operation flow of a method for controlling an intelligent device based on machine learning according to the present invention; Figure 2 A schematic diagram of feature extraction of a smart device control method based on machine learning according to the present invention; Figure 3 A schematic diagram of a machine learning training model flow chart of a machine learning-based intelligent device control method of the present invention; Figure 4 This is a schematic diagram of the operation flow of an intelligent device control system based on machine learning in the present invention; Figure 5 This is a schematic diagram of a data acquisition module of an intelligent device control system based on machine learning in the present invention. DETAILED DESCRIPTION

[0016] The following embodiments of the present invention are described in further detail in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0017] like Figure 1-Figure 5 As shown, a method for controlling an intelligent device based on machine learning includes the following specific steps: S1. Data acquisition: Collect user device usage data through device sensors or user feedback, preprocess and extract features from the collected data, and obtain a training data set; In step S1, the user's device usage data is divided into real-time operation data of the device and user behavior data, including information such as device usage time, usage frequency, and usage scenario; In step S1, feature extraction specifically includes: Statistical features: including but not limited to mean, variance, maximum and minimum values, and median, which can well describe the distribution and central tendency of data and are often used for feature extraction of numerical data; Frequency domain features: refers to converting data into the frequency domain for feature extraction by performing Fourier transform and other operations on the data. It can well describe the periodicity and frequency distribution characteristics of the data and is often used in the fields of signal processing and audio processing. Filter characteristics: refers to extracting the frequency response and time domain characteristics of data by designing and applying filters, which can well capture the local characteristics and frequency components of data, and is often used in the fields of image processing and signal processing; Wavelet transform features: By performing wavelet transform on data, the characteristic representation of data at different scales and frequencies is obtained, which can well describe the local characteristics and frequency characteristics of data and is often used in the fields of signal processing and image processing; Principal component analysis features: The original data is mapped to a low-dimensional space through linear transformation to obtain the principal component features of the data, which can well describe the main change direction and correlation of the data and is often used for data compression; Independent component analysis features: By performing independent component analysis on the data, the independent component features of the data are obtained, which can well describe the independence and mixed characteristics of the data and is often used in the fields of signal processing and medical image processing; Dictionary learning features: By learning the sparse representation dictionary of data, the sparse coding features of the data are obtained, which can well describe the sparsity and local characteristics of the data and are often used in the fields of image processing and pattern recognition; S2, model training: select machine learning algorithms and models according to specific application scenarios and requirements, introduce training data sets, and use machine learning to train models; In step S2, the machine learning algorithms and models include but are not limited to linear regression, logistic regression, decision tree, random forest, support vector machine, neural network, and the loss function of machine learning is defined as follows: in, is the number of training samples, is the number of features, is the model parameter vector, is the design matrix, The target value of samples, is the regularization coefficient; And the model parameter vector The optimal solution of is defined as follows: in, is the learning rate, is the loss function with respect to The partial derivatives of the parameters, It is The symbolic function of the parameters; in step S2, the specific process of machine learning training model is as follows: Data training: Use the training data set to train the selected model, and use optimization algorithms such as gradient descent to iteratively update the model parameters to minimize the loss function; Model evaluation: During the model training process, the model needs to be evaluated to determine the model performance. Evaluation indicators include but are not limited to precision, recall rate, and F1 value; Model parameter adjustment: According to the results of model evaluation, the model parameters are adjusted to further improve the performance of the model; Model preservation and deployment: Save the trained model and deploy it to actual applications for prediction and classification tasks; S3, prediction and control: After the model training is completed, the model is used to predict the user's behavior habits, and the corresponding control strategy is generated according to the prediction results to control the device; In step S3, the control strategy includes but is not limited to adjusting the operation mode of the device, optimizing the operation parameters of the device, and triggering the collaborative work between devices. The operation mode of the device includes a motion mode, and the motion mode at least includes a moving distance. and moving angle The instruction parameters; Moving distance The command parameters are defined as follows: Moving Angle The command parameters are defined as follows: in, Indicates the moving distance, Indicates the target location, Indicates the current location. Indicates the moving angle; S4, interactive control: users view the operating status of the device through the interactive interface, receive system recommendations and prompts, and adjust the device settings according to their own needs; S5. Feedback adjustment: Feedback interaction data to the machine learning model to generate new prediction results to adjust the control strategy, optimize the operating mode and parameters of the equipment, and keep pace with user needs.

[0018] like Figure 1-Figure 5 As shown, a smart device control system based on machine learning is applied to a server, and the system includes: Data acquisition module: The data acquisition module collects device and user data, and performs preprocessing and feature extraction so that the subsequent machine learning algorithm can understand and use it; the data acquisition module specifically includes: Data collection unit: collects real-time operation data of the equipment and user behavior data through sensors and data acquisition devices; Preprocessing unit: preprocess the collected data, including but not limited to data cleaning, data integration, data transformation, and data reduction; Feature extraction unit: selects, converts and combines relevant information from raw data, uses computers or other algorithms to analyze and transform data, and extracts representative and distinguishing features; The machine learning module selects a machine learning algorithm to learn and analyze the collected data and generate a prediction model; Model prediction module: The model prediction module uses models to learn user behavior patterns and device usage rules to predict future device needs and user preferences; Strategy generation module,The strategy generation and execution module generates corresponding control strategies based on the prediction results of the machine learning model; A policy execution module controls the smart device through a device driver and an actuator according to the control policy; User interaction module, the user interaction module allows users to view the operating status of the device, adjust the device settings, receive system recommendations and prompts, and allow users to provide feedback and evaluation to help optimize and improve the system.

[0019] at least one memory for storing computer instructions; At least one processor in communication with the memory, wherein when the at least one processor executes computer instructions, the at least one processor causes the system to execute: a data acquisition module, a machine learning module, a model prediction module, a strategy generation module, a strategy execution module, and a user interaction module.

[0020] A computer program is stored in the memory, and when the computer program is executed by the processor, a smart device control method based on machine learning is implemented.

[0021] In summary, combined Figure 1-Figure 5 As shown, the working principle of the intelligent device control method and system based on machine learning is as follows: S1. Data acquisition: Collect user device usage data through device sensors or user feedback. The user device usage data is divided into real-time device operation data and user behavior data, including information such as device usage time, usage frequency, and usage scenarios. Preprocess and extract features from the collected data. Feature extraction specifically includes: statistical features, frequency domain features, filter features, wavelet transform features, principal component analysis features, independent component analysis features, and dictionary learning features; obtain training data sets; S2. Model training: Select machine learning algorithms and models based on specific application scenarios and requirements. Machine learning algorithms and models include but are not limited to linear regression, logistic regression, decision tree, random forest, support vector machine, neural network, and the loss function of machine learning is defined as follows: in, is the number of training samples, is the number of features, is the model parameter vector, is the design matrix, The target value of samples, is the regularization coefficient; and the model parameter vector The optimal solution of is defined as follows: in, is the learning rate, is the loss function with respect to The partial derivatives of the parameters, It is The specific process of machine learning training model is as follows: data training: use the training data set to train the selected model, and use optimization algorithms such as gradient descent to iteratively update the model parameters to minimize the loss function; model evaluation: during the model training process, the model needs to be evaluated to determine the performance of the model. The evaluation indicators include but are not limited to accuracy, recall rate, and F1 value; model parameter adjustment: according to the results of model evaluation, the model parameters are adjusted to further improve the performance of the model; model preservation and deployment: save the trained model and deploy it to practical applications for prediction and classification tasks; S3. Prediction and control: After the model training is completed, the model is used to predict the user's behavior habits, and the corresponding control strategy is generated based on the prediction results to control the device; the control strategy includes but is not limited to adjusting the device's operating mode, optimizing the device's operating parameters, and triggering collaborative work between devices. The device's operating mode includes a motion mode, and the motion mode at least includes the moving distance and moving angle The instruction parameters; Moving distance The command parameters are defined as follows: Moving Angle The command parameters are defined as follows: in, Indicates the moving distance, Indicates the target location, Indicates the current location. Indicates the moving angle; S4, interactive control: users view the operating status of the device through the interactive interface, receive system recommendations and prompts, and adjust the device settings according to their own needs; S5. Feedback adjustment: Feedback interaction data to the machine learning model to generate new prediction results to adjust the control strategy, optimize the operating mode and parameters of the equipment, and keep pace with user needs.

[0022] A machine learning-based intelligent device control system, applied to a server, comprising: Data acquisition module: The data acquisition module collects device and user data, and performs preprocessing and feature extraction so that the subsequent machine learning algorithm can understand and use it; the data acquisition module specifically includes: data collection unit: collects real-time operation data of the device and user behavior data through sensors and data acquisition devices; preprocessing unit: preprocesses the collected data, including but not limited to data cleaning, data integration, data transformation, and data reduction; feature extraction unit: selects, converts and combines relevant information from the original data, uses computers or other algorithms to analyze and transform data, and extracts representative and distinguishing features The machine learning module selects a machine learning algorithm to learn and analyze the collected data and generate a prediction model; Model prediction module: The model prediction module uses models to learn user behavior patterns and device usage rules to predict future device needs and user preferences; Strategy generation module,The strategy generation and execution module generates corresponding control strategies based on the prediction results of the machine learning model; A policy execution module controls the smart device through a device driver and an actuator according to the control policy; User interaction module, the user interaction module allows users to view the operating status of the device, adjust the device settings, receive system recommendations and prompts, and allow users to provide feedback and evaluation to help optimize and improve the system.

[0023] at least one memory for storing computer instructions; At least one processor in communication with the memory, wherein when the at least one processor executes computer instructions, the at least one processor causes the system to execute: a data acquisition module, a machine learning module, a model prediction module, a strategy generation module, a strategy execution module, and a user interaction module.

[0024] A computer program is stored in the memory, and when the computer program is executed by the processor, a smart device control method based on machine learning is implemented.

[0025] The embodiments of the present invention are given for the purpose of illustration and description, and are not intended to be exhaustive or to limit the invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

Claims

1. A method for controlling an intelligent device based on machine learning, characterized in that: The specific steps include: S1. Data acquisition: Collect user device usage data through device sensors or user feedback, preprocess and extract features from the collected data, and obtain a training data set; S2, model training: select machine learning algorithms and models according to specific application scenarios and requirements, introduce training data sets, and use machine learning to train models; S3, prediction and control: After the model training is completed, the model is used to predict the user's behavior habits, and the corresponding control strategy is generated according to the prediction results to control the device; S4, interactive control: users view the operating status of the device through the interactive interface, receive system recommendations and prompts, and adjust the device settings according to their needs; S5. Feedback adjustment: Feedback interaction data to the machine learning model to generate new prediction results to adjust the control strategy, optimize the operating mode and parameters of the equipment, and keep pace with user needs.

2. The intelligent device control method based on machine learning according to claim 1, characterized in that: In step S1, the user's device usage data is divided into real-time operation data of the device and user's behavior data, including information such as device usage time, usage frequency, and usage scenario.

3. The intelligent device control method based on machine learning according to claim 1, characterized in that: In step S2, the machine learning algorithms and models include but are not limited to linear regression, logistic regression, decision tree, random forest, support vector machine, neural network, and the loss function of machine learning is defined as follows: in, is the number of training samples, is the number of features, is the model parameter vector, is the design matrix, The target value of samples, is the regularization coefficient; and the model parameter vector The optimal solution of is defined as follows: in, is the learning rate, is the loss function with respect to The partial derivatives of the parameters, It is A symbolic function of 1 parameter.

4. The intelligent device control method based on machine learning according to claim 1, characterized in that: In step S2, the specific process of machine learning training model is as follows: Data training: Use the training data set to train the selected model, and use optimization algorithms such as gradient descent to iteratively update the model parameters to minimize the loss function; Model evaluation: During the model training process, the model needs to be evaluated to determine the performance of the model. The evaluation indicators include but are not limited to precision, recall rate, and F1 value; Model parameter adjustment: According to the results of model evaluation, the model parameters are adjusted to further improve the performance of the model; Model preservation and deployment: Save the trained model and deploy it to actual applications for prediction and classification tasks.

5. The intelligent device control method based on machine learning according to claim 1, characterized in that: In step S3, the control strategy includes but is not limited to adjusting the operation mode of the device, optimizing the operation parameters of the device, and triggering the collaborative work between devices. The operation mode of the device includes a motion mode, and the motion mode at least includes a moving distance. and moving angle The instruction parameter; the moving distance The directive parameters are defined as follows: The moving angle The directive parameters are defined as follows: in, Indicates the moving distance, Indicates the target location, Indicates the current location. Indicates the moving angle.

6. A control system applied to the control method according to any one of claims 1 to 5, characterized in that: The system comprises: A data acquisition module that collects device and user data and performs preprocessing and feature extraction so that subsequent machine learning algorithms can understand and use it; A machine learning module, wherein the machine learning module selects a machine learning algorithm to learn and analyze the collected data to generate a prediction model; A model prediction module, which uses a model to learn user behavior patterns and device usage rules to predict future device needs and user preferences; A strategy generation module, wherein the strategy generation and execution module generates a corresponding control strategy according to the prediction results of the machine learning model; A policy execution module, wherein the policy execution module controls the intelligent device through a device driver and an actuator according to the control policy; A user interaction module allows the user to view the operating status of the device, adjust the settings of the device, receive system recommendations and prompts, and provide feedback and comments to help optimize and improve the system.

7. The control system according to claim 6, characterized in that: The data acquisition module specifically includes: Data collection unit: collects real-time operation data of the equipment and user behavior data through sensors and data acquisition devices; Preprocessing unit: preprocess the collected data, including but not limited to data cleaning, data integration, data transformation, and data reduction; Feature extraction unit: selects, converts and combines relevant information from raw data, uses computers or other algorithms to analyze and transform data, and extracts representative and distinctive features.

8. The control system according to claim 7, characterized in that: include: At least one memory for storing computer instructions; At least one processor in communication with the memory, wherein when the at least one processor executes computer instructions, the at least one processor causes the system to execute: a data acquisition module, a machine learning module, a model prediction module, a strategy generation module, a strategy execution module, and a user interaction module.

9. The control system according to claim 8, characterized in that: The memory stores a computer program, which, when executed by the processor, implements the method according to any one of claims 1 to 5.

Images