Smart factory personnel positioning and scheduling system based on three-dimensional modeling

Abstract

The invention discloses a smart factory personnel positioning and scheduling system based on three-dimensional modeling, which relates to the technical field of smart factory personnel positioning and scheduling, and comprises a smart factory data acquisition module, a preprocessing module, a three-dimensional positioning module, an environment monitoring module, a model fusion module and a personnel scheduling module, acquiring smart factory environment data, point cloud data of a smart factory and smart factory personnel data by using acquisition equipment; the three-dimensional positioning module updates real-time three-dimensional positioning coordinates of smart factory personnel, constructs a smart factory three-dimensional model, and obtains real-time three-dimensional positioning data of the smart factory personnel. The data acquisition technology, the positioning error calculation technology, the three-dimensional modeling technology, the model construction technology and the model fusion technology in the system are closely combined with the modern information technology, and the intelligent degree in the intelligent factory personnel positioning and scheduling process based on three-dimensional modeling is remarkably enhanced.

Description

[0001] The present invention relates to the field of intelligent factory personnel positioning and scheduling technology, and particularly relates to an intelligent factory personnel positioning and scheduling system based on three-dimensional modeling. Background Art

[0002] In the current rapid development of modern manufacturing, intelligent factories have become a key trend to improve production efficiency and optimize management models. Traditional factories face many challenges in personnel management and are difficult to meet the needs of efficient production. On the one hand, the personnel in the factory are widely distributed and their working states are complex. Only relying on manual records and experience-based scheduling, it is impossible to accurately grasp the real-time positions and work processes of personnel, resulting in unreasonable task allocation and seriously affecting production efficiency. On the other hand, when the factory faces emergencies, it is impossible to quickly locate relevant personnel, resulting in slow emergency response and bringing losses to the factory. At the same time, three-dimensional modeling technology and advanced personnel positioning technology are constantly maturing. Against this background, by integrating three-dimensional modeling and personnel positioning technology, an intelligent factory personnel positioning and scheduling system based on three-dimensional modeling has emerged. Its purpose is to use advanced technology to break through the limitations of traditional personnel management, provide an efficient and accurate personnel positioning and a scientific and reasonable scheduling plan for the factory, improve the overall production efficiency, enhance the factory's ability to respond to emergencies, and promote the transformation of the manufacturing industry towards intelligence and high efficiency. Although the existing technology has made great progress in the direction of intelligent factory personnel positioning and scheduling, there are still some problems to be optimized. The existing technology is difficult to combine the environmental factors of intelligent factories and monitor the interference brought by various environmental factors to the personnel positioning in intelligent factories, which affects the accuracy of real-time positioning of intelligent factory personnel and leads to unreasonable scheduling of intelligent factory personnel. Summary of the Invention

[0003] The purpose of the present invention is to provide an intelligent factory personnel positioning and scheduling system based on three-dimensional modeling to solve the problems raised in the above background art.

[0004] To solve the above technical problems, the technical solution adopted by the present invention is: an intelligent factory personnel positioning and scheduling system based on three-dimensional modeling, including an intelligent factory data acquisition module, a preprocessing module, a three-dimensional positioning module, an environmental monitoring module, a model fusion module, and a personnel scheduling module. Among them, each module is communicatively connected. The intelligent factory data acquisition module uses acquisition devices to collect intelligent factory environmental data, point cloud data of the intelligent factory, and intelligent factory personnel data, providing data support for the realization of the functions of its subsequent modules. The preprocessing module performs data cleaning, data calibration, data standardization, and time synchronization processing on the collected data. The three-dimensional positioning module obtains the real-time positioning error of the personnel in the intelligent factory according to the Euclidean distance formula, constructs a positioning error monitoring model, provides a data basis for accurately determining the real-time three-dimensional positioning coordinates of the personnel in the intelligent factory and constructing a dust concentration monitoring model and an electromagnetic noise monitoring model, updates the real-time three-dimensional positioning coordinates of the personnel in the intelligent factory in combination with the positioning error monitoring model, and constructs a three-dimensional model of the intelligent factory by using the point cloud data of the intelligent factory to obtain the real-time three-dimensional positioning data of the personnel in the intelligent factory; The environmental monitoring module constructs a dust concentration monitoring model and an electromagnetic noise monitoring model respectively through a linear regression algorithm; The model fusion module fuses the dust concentration monitoring model and the electromagnetic noise monitoring model by using a neural network algorithm to obtain an environmental monitoring model; The personnel scheduling module combines the three-dimensional model of the intelligent factory and the environmental monitoring model to obtain the positioning of abnormal environmental data, and schedules the personnel in the intelligent factory to take corresponding measures according to the real-time three-dimensional positioning data of the personnel in the intelligent factory, solving the problem in the prior art that it is difficult to combine the environmental factors of the intelligent factory to monitor the interference caused by various environmental factors to the positioning of the personnel in the intelligent factory, resulting in unreasonable scheduling of the personnel in the intelligent factory.

[0005] A further improvement of the technical solution of the present invention lies in that the process of collecting the environmental data of the intelligent factory and the point cloud data of the intelligent factory by the intelligent factory data collection module by using the collection equipment includes: The collection equipment includes a dust concentration sensor, an electromagnetic interference measuring instrument, a three-dimensional laser scanner, a UWB positioning device, an inertial navigation positioning device and a laser tracker. The environmental data of the intelligent factory includes the dust concentration in the environment of the intelligent factory and the electromagnetic noise in the environment of the intelligent factory. The personnel data of the intelligent factory includes the real-time three-dimensional positioning coordinates of the personnel in the intelligent factory and the reference three-dimensional positioning coordinates of the personnel in the intelligent factory; Using a dust concentration sensor, based on the principle of light scattering, to collect the dust concentration in the environment of the intelligent factory; Using an electromagnetic interference measuring instrument, based on the principle of electromagnetic induction, to collect the electromagnetic noise in the environment of the intelligent factory; Select a scanning site in the intelligent factory, fix the three-dimensional laser scanner at the scanning site in the intelligent factory, set the scanning parameters to obtain the position and attitude information of the three-dimensional laser scanner. The three-dimensional laser scanner measures the time from the emission of the laser beam to its reflection back to the three-dimensional laser scanner, and combines the position and attitude information of the three-dimensional laser scanner to calculate the distance between the measured point and the three-dimensional laser scanner to obtain the point cloud data of the intelligent factory; The UWB positioning device consists of a UWB tag, a UWB base station, and a positioning engine. The UWB tag is worn on the personnel in the smart factory, and the UWB base station is deployed at the central position of the smart factory to receive UWB pulse signals. The positioning engine measures the distance between the UWB tag and the UWB base station based on the ToF ranging principle, and obtains the real-time three-dimensional positioning coordinates of the personnel in the smart factory through a multi-variable positioning algorithm. Using a laser tracker, based on the principle of triangulation, the three-dimensional initial positioning coordinates of the personnel in the smart factory are collected. The inertial navigation positioning device consists of an inertial measurement unit and a data processing unit. Among them, the inertial measurement unit measures the real-time acceleration of the personnel in the smart factory in three-dimensional space and the real-time angular velocity of the personnel in the smart factory in three-dimensional space. The data processing unit receives the real-time acceleration and real-time angular velocity of the personnel in the smart factory in three-dimensional space, performs integral operations on the received data, obtains the displacement of the personnel in the smart factory in three-dimensional space, and calculates the sum of the three-dimensional initial positioning coordinates of the personnel in the smart factory and the displacement of the personnel in the smart factory in three-dimensional space to obtain the reference three-dimensional positioning coordinates of the personnel in the smart factory.

[0006] A further improvement of the technical solution of the present invention lies in that: the process of the preprocessing module performing data cleaning, data calibration, data standardization, and time synchronization on the collected data includes: Perform data cleaning, data calibration, and data standardization processing on the dust concentration in the smart factory environment, the electromagnetic noise in the smart factory environment, the real-time three-dimensional positioning coordinates of the personnel in the smart factory, the reference three-dimensional positioning coordinates of the personnel in the smart factory, and the point cloud data of the smart factory to remove duplicate values and outliers. Add timestamps to the dust concentration in the smart factory environment, the electromagnetic noise in the smart factory environment, the real-time three-dimensional positioning coordinates of the personnel in the smart factory, and the point cloud data of the smart factory collected. According to the timestamps, synchronize the collection times of the dust concentration in the smart factory environment, the real-time three-dimensional positioning coordinates of the personnel in the smart factory, the reference three-dimensional positioning coordinates of the personnel in the smart factory, and the point cloud data of the smart factory, and synchronize the collection times of the electromagnetic noise in the smart factory environment, the real-time three-dimensional positioning coordinates of the personnel in the smart factory, the reference three-dimensional positioning coordinates of the personnel in the smart factory, and the point cloud data of the smart factory.

[0007] A further improvement of the technical solution of the present invention lies in that: the process of the three-dimensional positioning module obtaining the real-time positioning error of the personnel in the smart factory using the Euclidean distance formula includes: The real-time positioning error of the personnel in the smart factory is divided into the real-time positioning error of the personnel in the smart factory caused by dust concentration and the real-time positioning error of the personnel in the smart factory caused by electromagnetic noise. Extract the real-time three-dimensional positioning coordinates of intelligent factory personnel and the reference three-dimensional positioning coordinates of intelligent factory personnel corresponding to the acquisition timestamp of the dust concentration in the intelligent factory environment, and use the Euclidean distance formula to calculate and obtain the real-time positioning error of intelligent factory personnel caused by the dust concentration; Extract the real-time three-dimensional positioning coordinates of intelligent factory personnel and the reference three-dimensional positioning coordinates of intelligent factory personnel corresponding to the acquisition timestamp of the electromagnetic noise in the intelligent factory environment, and use the Euclidean distance formula to calculate and obtain the real-time positioning error of intelligent factory personnel caused by the electromagnetic noise.

[0008] A further improvement of the technical solution of the present invention lies in that: in the three-dimensional positioning module, the process of constructing the positioning error monitoring model includes: Using the random forest algorithm to construct a random forest model, taking the dust concentration in the intelligent factory environment, the electromagnetic noise in the intelligent factory environment, the real-time positioning error of intelligent factory personnel caused by the dust concentration, and the real-time positioning error of intelligent factory personnel caused by the electromagnetic noise as a data set, dividing it into a training set and a test set according to a ratio of 7:3, using the training set data to train the random forest model, learning the non-linear relationship between the dust concentration in the intelligent factory environment and the real-time positioning error of intelligent factory personnel caused by the dust concentration, and the non-linear relationship between the electromagnetic noise in the intelligent factory environment and the real-time positioning error of intelligent factory personnel caused by the electromagnetic noise, to obtain a trained random forest model; using the test set data to evaluate the trained random forest, adjusting the random forest model parameters, optimizing the random forest model, and deploying the optimized random forest model in an intelligent factory personnel positioning and scheduling system based on three-dimensional modeling to obtain the positioning error monitoring model.

[0009] A further improvement of the technical solution of the present invention lies in that: in the three-dimensional positioning module, the process of combining the positioning error monitoring model to update the real-time positioning of intelligent factory personnel, constructing a three-dimensional model of the intelligent factory, and obtaining the real-time three-dimensional positioning data of intelligent factory personnel includes: Input the dust concentration in the intelligent factory environment and the electromagnetic noise in the intelligent factory environment into the positioning error monitoring model, and the positioning error monitoring model respectively outputs the real-time positioning error of intelligent factory personnel caused by the dust concentration and the real-time positioning error of intelligent factory personnel caused by the electromagnetic noise; According to the real-time positioning error of intelligent factory personnel caused by the dust concentration and the real-time positioning error of intelligent factory personnel caused by the electromagnetic noise, update the real-time positioning coordinates of intelligent factory personnel; Extract the neighborhood points of each point in the point cloud data of the smart factory. Using the principal component analysis method, calculate the covariance matrix of each neighborhood point. By performing eigenvalue decomposition on the covariance matrix of each neighborhood point, obtain the eigenvalues and eigenvectors. Through the eigenvalues and eigenvectors, extract the plane features in the point cloud data of the smart factory; calculate the normal vector of each point in the point cloud data of the smart factory. According to the angle between the normal vector of each point in the point cloud data of the smart factory and the normal vector of its adjacent points, extract the edge features in the point cloud data of the smart factory; use the Harris corner detection algorithm to extract the corner features in the point cloud data of the smart factory; Based on the extracted plane features, edge features and corner features, use the Delaunay triangulation algorithm to perform Delaunay triangulation on the point cloud data of the smart factory, connect the discrete points in the point cloud data of each smart factory into triangular patches, and construct a 3D model of the smart factory; Map the updated real-time 3D positioning coordinates of the smart factory personnel into the 3D model of the smart factory to obtain the real-time 3D positioning data of the smart factory personnel.

[0010] A further improvement of the technical solution of the present invention lies in: the environmental monitoring module, in the process of constructing a dust concentration monitoring model through the linear regression algorithm, includes: Take the real-time positioning error of the smart factory personnel caused by the dust concentration as the input variable, and take the influence index of the dust concentration on the positioning of the smart factory personnel as the output variable. Use the linear regression algorithm to construct a linear regression model. The expression of this linear regression model is , where is the influence index of the dust concentration on the positioning of the smart factory personnel, is the real-time positioning error of the smart factory personnel caused by the dust concentration, is the intercept when the real-time positioning error of the smart factory personnel caused by the dust concentration is 0, is the slope, is the error term, and the least squares method is used to estimate and , and obtain the dust concentration monitoring model.

[0011] A further improvement of the technical solution of the present invention lies in: the environmental monitoring module, in the process of constructing an electromagnetic noise monitoring model through the linear regression algorithm, includes: Take the real-time positioning error of the smart factory personnel caused by the electromagnetic noise as the input variable, and take the influence index of the electromagnetic noise on the positioning of the smart factory personnel as the output variable. Use the linear regression algorithm to construct a linear regression model. The expression of this linear regression model is , where is the influence index of the electromagnetic noise on the positioning of the smart factory personnel, The real-time positioning error of personnel in the smart factory caused by electromagnetic noise The intercept when the real-time positioning error of personnel in the smart factory caused by electromagnetic noise is 0 Is the slope Is the error term, estimated using the least squares method And , to obtain an electromagnetic noise monitoring model

[0012] A further improvement of the technical solution of the present invention lies in: the model fusion module uses a neural network algorithm to fuse the dust concentration monitoring model and the electromagnetic noise monitoring model. The process of obtaining the environmental monitoring model includes: Input the real-time positioning error of personnel in the smart factory caused by dust concentration into the dust concentration monitoring model to obtain the influence index of dust concentration on the positioning of personnel in the smart factory; input the real-time positioning error of personnel in the smart factory caused by electromagnetic noise into the electromagnetic noise monitoring model to obtain the influence index of electromagnetic noise on the positioning of personnel in the smart factory; Use a neural network algorithm to construct a neural network model. Take the real-time positioning error of personnel in the smart factory caused by dust concentration, the real-time positioning error of personnel in the smart factory caused by electromagnetic noise, the influence index of dust concentration on the positioning of personnel in the smart factory, and the influence index of electromagnetic noise on the positioning of personnel in the smart factory as a data set, and divide it into a training set and a test set according to a ratio of 7:3. Select MLP as the neural network structure. The input layer includes two neurons, receiving the real-time positioning error of personnel in the smart factory caused by dust concentration and the real-time positioning error of personnel in the smart factory caused by electromagnetic noise. The hidden layer is configured with the MSE function. The output layer includes two neurons, outputting the influence index of dust concentration on the positioning of personnel in the smart factory and the influence index of electromagnetic noise on the positioning of personnel in the smart factory; Input the training set data into the neural network model, set the learning rate to 0.01, and the number of iterative training times to 1000. The training process includes forward propagation and backward propagation. Among them, forward propagation is used to calculate the predicted output data, and backward propagation is used to update the weights and biases of the model. Through repeated iterative training, learn the non-linear relationship between the real-time positioning error of personnel in the smart factory caused by dust concentration and the influence index of dust concentration on the positioning of personnel in the smart factory, and the non-linear relationship between the real-time positioning error of personnel in the smart factory caused by electromagnetic noise and the influence index of electromagnetic noise on the positioning of personnel in the smart factory, until the set number of iterative training times is reached, and obtain the trained neural network model; Input the test set data into the trained neural network model, use the MSE function to evaluate the error between the output value and the actual value of the neural network model, and adjust the parameters of the neural network model according to the evaluation result to obtain the environmental monitoring model

[0013] A further improvement of the technical solution of the present invention lies in that: in the process of the personnel scheduling module scheduling the personnel in the smart factory to take corresponding measures, it includes: Input the real-time positioning error of the personnel in the smart factory caused by the dust concentration and the real-time positioning error of the personnel in the smart factory caused by the electromagnetic noise into the environmental monitoring model, and the environmental monitoring model outputs the influence index of the dust concentration on the positioning of the personnel in the smart factory and the influence index of the electromagnetic noise on the positioning of the personnel in the smart factory; When the influence index of the dust concentration on the positioning of the personnel in the smart factory is lower than 0.4, the influence of the dust concentration on the positioning of the personnel in the smart factory is small; when the influence index of the dust concentration on the positioning of the personnel in the smart factory is higher than 0.4, it indicates that the dust concentration is abnormal and has a great influence on the positioning of the personnel in the smart factory. Extract the three-dimensional positioning coordinates with abnormal dust concentration from the three-dimensional model of the smart factory. According to the real-time three-dimensional positioning data of the personnel in the smart factory, through the Euclidean formula, obtain the distance between the three-dimensional positioning coordinates with abnormal dust concentration and the three-dimensional positioning data of the personnel in the smart factory, and extract the personnel in the smart factory corresponding to the minimum distance value, and schedule this personnel in the smart factory to cope with the interference of the dust concentration; When the influence index of the electromagnetic noise on the positioning of the personnel in the smart factory is lower than 0.3, the influence of the electromagnetic noise on the positioning of the personnel in the smart factory is small; when the influence index of the electromagnetic noise on the positioning of the personnel in the smart factory is higher than 0.3, it indicates that the electromagnetic noise is abnormal and has a great influence on the positioning of the personnel in the smart factory. Extract the three-dimensional positioning coordinates with abnormal electromagnetic noise from the three-dimensional model of the smart factory. According to the real-time three-dimensional positioning data of the personnel in the smart factory, through the Euclidean formula, obtain the distance between the three-dimensional positioning coordinates with abnormal electromagnetic noise and the three-dimensional positioning data of the personnel in the smart factory, and extract the personnel in the smart factory corresponding to the minimum distance value, and schedule this personnel in the smart factory to cope with the interference of the electromagnetic noise.

[0014] The beneficial effects of the present invention are as follows: In the intelligent factory personnel positioning and scheduling system based on 3D modeling of the present invention, compared with the traditional intelligent factory personnel positioning and scheduling system based on 3D modeling, the data acquisition technology, positioning error calculation technology, 3D modeling technology, model construction technology, and model fusion technology in the system of the present invention are closely combined with modern information technology to accurately capture the environmental data of the intelligent factory, the personnel data of the intelligent factory, and the point cloud data of the intelligent factory. Furthermore, the real-time positioning error of the intelligent factory personnel, the 3D model of the intelligent factory, and the real-time 3D positioning data of the intelligent factory personnel are obtained. Using the linear regression algorithm, a dust concentration monitoring model and an electromagnetic noise monitoring model are respectively constructed, and through the neural network algorithm, the dust concentration monitoring model and the electromagnetic noise monitoring model are fused to obtain an environmental monitoring model, achieving real-time and comprehensive monitoring of the positioning of intelligent factory personnel and the environmental conditions, solving the problem that it is difficult to monitor the interference brought by various environmental factors to the positioning of intelligent factory personnel, resulting in low positioning accuracy of intelligent factory personnel and unreasonable scheduling of intelligent factory personnel. It ensures that the system in the present invention can refine the dynamic monitoring standard for the intelligent factory personnel positioning and scheduling system within a more accurate range, making the monitored data more accurate indicators under the same conditions. The research and application of this method significantly enhance the degree of intelligence in the process of intelligent factory personnel positioning and scheduling based on 3D modeling. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0016] Figure 1 It is a block diagram of an intelligent factory personnel positioning and scheduling system based on 3D modeling of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0018] Such as Figure 1As shown in the figure, the present invention provides a personnel positioning and scheduling system for an intelligent factory based on 3D modeling, including an intelligent factory data acquisition module, a preprocessing module, a 3D positioning module, an environmental monitoring module, a model fusion module, and a personnel scheduling module. Among them, each module is communicatively connected; The intelligent factory data acquisition module uses acquisition devices to collect intelligent factory environmental data, point cloud data of the intelligent factory, and intelligent factory personnel data, providing data support for the implementation of the subsequent module functions; The preprocessing module performs data cleaning, data calibration, data standardization, and time synchronization processing on the collected data; The 3D positioning module, according to the Euclidean distance formula, obtains the real-time positioning error of the intelligent factory personnel, constructs a positioning error monitoring model, provides a data basis for accurately determining the real-time 3D positioning coordinates of the intelligent factory personnel and constructing a dust concentration monitoring model and an electromagnetic noise monitoring model. Combining the positioning error monitoring model, it updates the real-time 3D positioning coordinates of the intelligent factory personnel. Using the point cloud data of the intelligent factory, it constructs a 3D model of the intelligent factory and obtains the real-time 3D positioning data of the intelligent factory personnel; The environmental monitoring module constructs a dust concentration monitoring model and an electromagnetic noise monitoring model respectively through a linear regression algorithm; The model fusion module uses a neural network algorithm to fuse the dust concentration monitoring model and the electromagnetic noise monitoring model to obtain an environmental monitoring model; The personnel scheduling module combines the 3D model of the intelligent factory and the environmental monitoring model to obtain the positioning of abnormal environmental data. According to the real-time 3D positioning data of the intelligent factory personnel, it schedules the intelligent factory personnel to take corresponding measures, solving the problem that it is difficult to combine the environmental factors of the intelligent factory in the prior art, monitor the interference caused by various environmental factors to the positioning of the intelligent factory personnel, and resulting in unreasonable scheduling of the intelligent factory personnel.

[0019] Preferably, the process of the intelligent factory data acquisition module using acquisition devices to collect intelligent factory environmental data and point cloud data of the intelligent factory includes: Among them, the acquisition devices are a dust concentration sensor, an electromagnetic interference measuring instrument, a 3D laser scanner, a UWB positioning device, an inertial navigation positioning device, and a laser tracker. The intelligent factory environmental data includes the dust concentration of the intelligent factory environment and the electromagnetic noise of the intelligent factory environment. The intelligent factory personnel data includes the real-time 3D positioning coordinates of the intelligent factory personnel and the reference 3D positioning coordinates of the intelligent factory personnel; Using the dust concentration sensor, based on the light scattering principle, collect the dust concentration of the intelligent factory environment; Adopt an electromagnetic interference measuring instrument to collect the electromagnetic noise of the intelligent factory environment based on the electromagnetic induction principle; Select a scanning site in the intelligent factory, fix the 3D laser scanner on the scanning site in the intelligent factory, set the scanning parameters, obtain the position and attitude information of the 3D laser scanner. The 3D laser scanner measures the time from the emission of the laser beam to its reflection back to the 3D laser scanner by emitting laser beams. Combining the position and attitude information of the 3D laser scanner, calculate the distance between the measured point and the 3D laser scanner to obtain the point cloud data of the intelligent factory; The UWB positioning device consists of a UWB tag, a UWB base station, and a positioning engine. Wear the UWB tag on the personnel in the intelligent factory, deploy the UWB base station at the central position of the intelligent factory to receive UWB pulse signals. The positioning engine measures the distance between the UWB tag and the UWB base station based on the ToF ranging principle, and obtains the real-time three-dimensional positioning coordinates of the personnel in the intelligent factory through a multi-variable positioning algorithm; Use a laser tracker to collect the initial three-dimensional positioning coordinates of the personnel in the intelligent factory based on the triangulation principle. The inertial navigation positioning device consists of an inertial measurement unit and a data processing unit. Among them, use the inertial measurement unit to measure the real-time acceleration of the personnel in the intelligent factory in three-dimensional space and the real-time angular velocity of the personnel in the intelligent factory in three-dimensional space. The data processing unit receives the real-time acceleration and real-time angular velocity of the personnel in the intelligent factory in three-dimensional space, performs integral operations on the received data to obtain the displacement of the personnel in the intelligent factory in three-dimensional space, and calculates the sum of the initial three-dimensional positioning coordinates of the personnel in the intelligent factory and the displacement of the personnel in the intelligent factory in three-dimensional space to obtain the reference three-dimensional positioning coordinates of the personnel in the intelligent factory.

[0020] Preferably, the process of the preprocessing module for data cleaning, data calibration, data standardization, and time synchronization of the collected data includes: Perform data cleaning, data calibration, and data standardization processing on the dust concentration in the intelligent factory environment, the electromagnetic noise in the intelligent factory environment, the real-time three-dimensional positioning coordinates of the personnel in the intelligent factory, the reference three-dimensional positioning coordinates of the personnel in the intelligent factory, and the point cloud data of the intelligent factory to remove duplicate values and outliers; Add timestamps to the dust concentration in the intelligent factory environment, the electromagnetic noise in the intelligent factory environment, the real-time three-dimensional positioning coordinates of the personnel in the intelligent factory, and the point cloud data of the intelligent factory. According to the timestamps, synchronize the collection times of the dust concentration in the intelligent factory environment, the real-time three-dimensional positioning coordinates of the personnel in the intelligent factory, the reference three-dimensional positioning coordinates of the personnel in the intelligent factory, and the point cloud data of the intelligent factory, and synchronize the collection times of the electromagnetic noise in the intelligent factory environment, the real-time three-dimensional positioning coordinates of the personnel in the intelligent factory, the reference three-dimensional positioning coordinates of the personnel in the intelligent factory, and the point cloud data of the intelligent factory.

[0021] Preferably, the process of the three-dimensional positioning module for obtaining the real-time positioning error of the personnel in the intelligent factory using the Euclidean distance formula includes: Among them, the real-time positioning error of personnel in the intelligent factory is divided into the real-time positioning error of personnel in the intelligent factory caused by dust concentration and the real-time positioning error of personnel in the intelligent factory caused by electromagnetic noise; Extract the real-time three-dimensional positioning coordinates and reference three-dimensional positioning coordinates of personnel in the intelligent factory corresponding to the acquisition timestamp of the dust concentration in the intelligent factory environment, and use the Euclidean distance formula to calculate and obtain the real-time positioning error of personnel in the intelligent factory caused by dust concentration; Extract the real-time three-dimensional positioning coordinates and reference three-dimensional positioning coordinates of personnel in the intelligent factory corresponding to the acquisition timestamp of the electromagnetic noise in the intelligent factory environment, and use the Euclidean distance formula to calculate and obtain the real-time positioning error of personnel in the intelligent factory caused by electromagnetic noise.

[0022] Preferably, the process of the three-dimensional positioning module for constructing the positioning error monitoring model includes: Using the random forest algorithm to construct a random forest model, taking the dust concentration in the intelligent factory environment, the electromagnetic noise in the intelligent factory environment, the real-time positioning error of personnel in the intelligent factory caused by dust concentration, and the real-time positioning error of personnel in the intelligent factory caused by electromagnetic noise as the data set, dividing it into a training set and a test set according to a ratio of 7:3, using the training set data to train the random forest model, learning the non-linear relationship between the dust concentration in the intelligent factory environment and the real-time positioning error of personnel in the intelligent factory caused by dust concentration, and the non-linear relationship between the electromagnetic noise in the intelligent factory environment and the real-time positioning error of personnel in the intelligent factory caused by electromagnetic noise, to obtain a trained random forest model; using the test set data to evaluate the trained random forest, adjusting the random forest model parameters, optimizing the random forest model, and deploying the optimized random forest model in an intelligent factory personnel positioning and scheduling system based on three-dimensional modeling to obtain the positioning error monitoring model.

[0023] Preferably, the process of the three-dimensional positioning module for updating the real-time positioning of personnel in the intelligent factory, constructing a three-dimensional model of the intelligent factory, and obtaining the real-time three-dimensional positioning data of personnel in the intelligent factory in combination with the positioning error monitoring model includes: Input the dust concentration in the intelligent factory environment and the electromagnetic noise in the intelligent factory environment into the positioning error monitoring model, and the positioning error monitoring model respectively outputs the real-time positioning error of personnel in the intelligent factory caused by dust concentration and the real-time positioning error of personnel in the intelligent factory caused by electromagnetic noise; Update the real-time positioning coordinates of personnel in the intelligent factory according to the real-time positioning error of personnel in the intelligent factory caused by dust concentration and the real-time positioning error of personnel in the intelligent factory caused by electromagnetic noise; Extract the neighborhood points of each point in the point cloud data of the smart factory. Using the principal component analysis method, calculate the covariance matrix of each neighborhood point. By performing eigenvalue decomposition on the covariance matrix of each neighborhood point, obtain the eigenvalues and eigenvectors. Through the eigenvalues and eigenvectors, extract the plane features in the point cloud data of the smart factory; calculate the normal vector of each point in the point cloud data of the smart factory. According to the angle between the normal vector of each point in the point cloud data of the smart factory and the normal vector of its adjacent points, extract the edge features in the point cloud data of the smart factory; use the Harris corner detection algorithm to extract the corner features in the point cloud data of the smart factory; Based on the extracted plane features, edge features and corner features, use the Delaunay triangulation algorithm to perform Delaunay triangulation on the point cloud data of the smart factory, connect the discrete points in each point cloud data of the smart factory into triangular patches, and construct a 3D model of the smart factory; Map the updated real-time 3D positioning coordinates of the smart factory personnel into the 3D model of the smart factory to obtain the real-time 3D positioning data of the smart factory personnel.

[0024] Preferably, the environmental monitoring module. The process of constructing a dust concentration monitoring model through the linear regression algorithm includes: Taking the real-time positioning error of the smart factory personnel caused by the dust concentration as the input variable, and taking the influence index of the dust concentration on the positioning of the smart factory personnel as the output variable. Using the linear regression algorithm, construct a linear regression model. The expression of this linear regression model is , where is the influence index of the dust concentration on the positioning of the smart factory personnel, is the real-time positioning error of the smart factory personnel caused by the dust concentration, is the intercept when the real-time positioning error of the smart factory personnel caused by the dust concentration is 0, is the slope, is the error term, estimated using the least squares method and , and obtain the dust concentration monitoring model.

[0025] Preferably, the environmental monitoring module. The process of constructing an electromagnetic noise monitoring model through the linear regression algorithm includes: Taking the real-time positioning error of the smart factory personnel caused by the electromagnetic noise as the input variable, and taking the influence index of the electromagnetic noise on the positioning of the smart factory personnel as the output variable. Using the linear regression algorithm, construct a linear regression model. The expression of this linear regression model is , where is the influence index of the electromagnetic noise on the positioning of the smart factory personnel, is the real-time positioning error of the smart factory personnel caused by the electromagnetic noise, is the intercept when the real-time positioning error of the intelligent factory personnel caused by electromagnetic noise is 0, is the slope, is the error term, estimated using the least squares method and , to obtain the electromagnetic noise monitoring model.

[0026] Preferably, the model fusion module uses the neural network algorithm to fuse the dust concentration monitoring model and the electromagnetic noise monitoring model. The process of obtaining the environmental monitoring model includes: Input the real-time positioning error of the intelligent factory personnel caused by dust concentration into the dust concentration monitoring model to obtain the influence index of dust concentration on the positioning of intelligent factory personnel; input the real-time positioning error of the intelligent factory personnel caused by electromagnetic noise into the electromagnetic noise monitoring model to obtain the influence index of electromagnetic noise on the positioning of intelligent factory personnel; Use the neural network algorithm to construct a neural network model. Take the real-time positioning error of the intelligent factory personnel caused by dust concentration, the real-time positioning error of the intelligent factory personnel caused by electromagnetic noise, the influence index of dust concentration on the positioning of intelligent factory personnel, and the influence index of electromagnetic noise on the positioning of intelligent factory personnel as the data set, and divide it into a training set and a test set according to a ratio of 7:3. Select MLP as the neural network structure. The input layer includes two neurons, receiving the real-time positioning error of the intelligent factory personnel caused by dust concentration and the real-time positioning error of the intelligent factory personnel caused by electromagnetic noise. The hidden layer configures the MSE function. The output layer includes two neurons, outputting the influence index of dust concentration on the positioning of intelligent factory personnel and the influence index of electromagnetic noise on the positioning of intelligent factory personnel; Input the training set data into the neural network model, set the learning rate to 0.01, and the number of iterative training times to 1000. The training process includes forward propagation and backward propagation. Among them, forward propagation is used to calculate the predicted output data, and backward propagation is used to update the weights and biases of the model. Through repeated iterative training, learn the non-linear relationship between the real-time positioning error of the intelligent factory personnel caused by dust concentration and the influence index of dust concentration on the positioning of intelligent factory personnel, and the non-linear relationship between the real-time positioning error of the intelligent factory personnel caused by electromagnetic noise and the influence index of electromagnetic noise on the positioning of intelligent factory personnel, until the set number of iterative training times is reached, and obtain the trained neural network model; Input the test set data into the trained neural network model, use the MSE function to evaluate the error between the output value and the actual value of the neural network model, and adjust the parameters of the neural network model according to the evaluation results to obtain the environmental monitoring model.

[0027] Preferably, the personnel scheduling module, the process of scheduling the intelligent factory personnel to take corresponding measures includes: Input the real-time positioning error of smart factory personnel caused by dust concentration and the real-time positioning error of smart factory personnel caused by electromagnetic noise into the environmental monitoring model. The environmental monitoring model outputs the influence index of dust concentration on the positioning of smart factory personnel and the influence index of electromagnetic noise on the positioning of smart factory personnel; When the influence index of dust concentration on the positioning of smart factory personnel is lower than 0.4, the influence of dust concentration on the positioning of smart factory personnel is small; when the influence index of dust concentration on the positioning of smart factory personnel is higher than 0.4, it indicates that the dust concentration is abnormal and has a great influence on the positioning of smart factory personnel. Extract the three-dimensional positioning coordinates with abnormal dust concentration from the three-dimensional model of the smart factory. According to the real-time three-dimensional positioning data of smart factory personnel, through the Euclidean formula, obtain the distance between the three-dimensional positioning coordinates with abnormal dust concentration and the three-dimensional positioning data of smart factory personnel, extract the smart factory personnel corresponding to the minimum distance value, and dispatch this smart factory personnel to cope with the interference of dust concentration; When the influence index of electromagnetic noise on the positioning of smart factory personnel is lower than 0.3, the influence of electromagnetic noise on the positioning of smart factory personnel is small; when the influence index of electromagnetic noise on the positioning of smart factory personnel is higher than 0.3, it indicates that the electromagnetic noise is abnormal and has a great influence on the positioning of smart factory personnel. Extract the three-dimensional positioning coordinates with abnormal electromagnetic noise from the three-dimensional model of the smart factory. According to the real-time three-dimensional positioning data of smart factory personnel, through the Euclidean formula, obtain the distance between the three-dimensional positioning coordinates with abnormal electromagnetic noise and the three-dimensional positioning data of smart factory personnel, extract the smart factory personnel corresponding to the minimum distance value, and dispatch this smart factory personnel to cope with the interference of electromagnetic noise.

[0028] First, use a dust concentration sensor to collect the dust concentration in the intelligent factory environment, use an electromagnetic interference measuring instrument to collect the electromagnetic noise in the intelligent factory environment, use a UWB positioning device to collect the real-time three-dimensional positioning coordinates of the personnel in the intelligent factory, use an inertial navigation positioning device and a laser tracker to collect the reference three-dimensional positioning coordinates of the personnel in the intelligent factory, and use a three-dimensional laser scanner to collect the point cloud data of the intelligent factory. Add timestamps to the collected data and achieve data collection synchronization according to the timestamps. Then, preprocess the collected data. Use the Euclidean distance formula to calculate and obtain the real-time positioning error of the personnel in the intelligent factory, and use the random forest algorithm to construct a positioning error monitoring model. Immediately afterwards, combine the positioning error monitoring model, update the real-time positioning coordinates of the personnel in the intelligent factory, construct a three-dimensional model of the intelligent factory, map the updated real-time three-dimensional positioning coordinates of the personnel in the intelligent factory into the three-dimensional model of the intelligent factory, and obtain the real-time three-dimensional positioning data of the personnel in the intelligent factory. Then, use the linear regression algorithm to construct a dust concentration monitoring model and an electromagnetic noise monitoring model, and use the neural network algorithm to fuse the dust concentration monitoring model and the electromagnetic noise monitoring model to obtain an environmental monitoring model. Finally, combine the three-dimensional model of the intelligent factory and the environmental monitoring model to obtain the positioning of abnormal environmental data, and dispatch the personnel in the intelligent factory to take corresponding measures according to the real-time three-dimensional positioning data of the personnel in the intelligent factory.

[0029] As mentioned above, it is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art in the technical field of this application can easily think of changes or substitutions within the technical scope disclosed by this application, and all of them should be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

1. A personnel positioning and scheduling system for intelligent factories based on 3D modeling, comprising an intelligent factory data acquisition module, a preprocessing module, a 3D positioning module, an environmental monitoring module, a model fusion module, and a personnel scheduling module, wherein, Each module is communicatively connected, characterized in that: The intelligent factory data acquisition module uses acquisition devices to acquire intelligent factory environment data, point cloud data of the intelligent factory, and intelligent factory personnel data; The preprocessing module performs data cleaning, data calibration, data standardization, and time synchronization processing on the acquired data; The 3D positioning module obtains the real-time positioning error of intelligent factory personnel according to the Euclidean distance formula, constructs a positioning error monitoring model, updates the real-time 3D positioning coordinates of intelligent factory personnel in combination with the positioning error monitoring model, constructs a 3D model of the intelligent factory using the point cloud data of the intelligent factory, and obtains the real-time 3D positioning data of intelligent factory personnel; The environment monitoring module constructs a dust concentration monitoring model and an electromagnetic noise monitoring model respectively through a linear regression algorithm; The model fusion module uses a neural network algorithm to fuse the dust concentration monitoring model and the electromagnetic noise monitoring model to obtain an environment monitoring model; The personnel scheduling module combines the 3D model of the intelligent factory and the environment monitoring model to obtain the location of abnormal environment data, and schedules intelligent factory personnel to take corresponding measures according to the real-time 3D positioning data of intelligent factory personnel.

2. The intelligent factory personnel positioning and scheduling system based on 3D modeling according to claim 1, characterized in that: The process of the intelligent factory data acquisition module using acquisition devices to acquire intelligent factory environment data and point cloud data of the intelligent factory includes: The acquisition devices are a dust concentration sensor, an electromagnetic interference measuring instrument, a 3D laser scanner, a UWB positioning device, an inertial navigation positioning device, and a laser tracker. The intelligent factory environment data includes the dust concentration of the intelligent factory environment and the electromagnetic noise of the intelligent factory environment. The intelligent factory personnel data includes the real-time 3D positioning coordinates of intelligent factory personnel and the reference 3D positioning coordinates of intelligent factory personnel; Using a dust concentration sensor, based on the light scattering principle, to acquire the dust concentration of the intelligent factory environment; Using an electromagnetic interference measuring instrument, based on the electromagnetic induction principle, to acquire the electromagnetic noise of the intelligent factory environment; Select a scanning site in the intelligent factory, fix the 3D laser scanner at the scanning site in the intelligent factory, set scanning parameters, obtain the position and attitude information of the 3D laser scanner. The 3D laser scanner measures the time from the emission of the laser beam to its reflection back to the 3D laser scanner, and combines the position and attitude information of the 3D laser scanner to calculate the distance between the measured point and the 3D laser scanner, and obtain the point cloud data of the intelligent factory; The UWB positioning device consists of a UWB tag, a UWB base station, and a positioning engine. Wear the UWB tag on the intelligent factory personnel, deploy the UWB base station at the central position of the intelligent factory, receive the UWB pulse signal, and the positioning engine measures the distance between the UWB tag and the UWB base station based on the ToF ranging principle, and obtains the real-time 3D positioning coordinates of intelligent factory personnel through a multi-variable positioning algorithm; Using a laser tracker, based on the principle of triangulation, the three-dimensional initial positioning coordinates of the personnel in the smart factory are collected. The inertial navigation positioning device consists of an inertial measurement unit and a data processing unit. Among them, the inertial measurement unit is used to measure the real-time acceleration of the personnel in the smart factory in three-dimensional space and the real-time angular velocity of the personnel in the smart factory in three-dimensional space. The data processing unit receives the real-time acceleration and real-time angular velocity of the personnel in the smart factory in three-dimensional space, performs integral operations on the received data to obtain the displacement of the personnel in the smart factory in three-dimensional space, and calculates the sum of the three-dimensional initial positioning coordinates of the personnel in the smart factory and the displacement of the personnel in the smart factory in three-dimensional space to obtain the reference three-dimensional positioning coordinates of the personnel in the smart factory.

3. The intelligent factory personnel positioning and scheduling system based on 3D modeling according to claim 2, characterized in that: The process of the preprocessing module for data cleaning, data calibration, data standardization, and time synchronization of the collected data includes: Perform data cleaning, data calibration, and data standardization on the dust concentration in the smart factory environment, the electromagnetic noise in the smart factory environment, the real-time three-dimensional positioning coordinates of the personnel in the smart factory, the reference three-dimensional positioning coordinates of the personnel in the smart factory, and the point cloud data of the smart factory to remove duplicate values and outliers; Add timestamps to the dust concentration in the smart factory environment, the electromagnetic noise in the smart factory environment, the real-time three-dimensional positioning coordinates of the personnel in the smart factory, and the point cloud data of the smart factory. According to the timestamps, synchronize the collection times of the dust concentration in the smart factory environment, the real-time three-dimensional positioning coordinates of the personnel in the smart factory, the reference three-dimensional positioning coordinates of the personnel in the smart factory, and the point cloud data of the smart factory, and synchronize the collection times of the electromagnetic noise in the smart factory environment, the real-time three-dimensional positioning coordinates of the personnel in the smart factory, the reference three-dimensional positioning coordinates of the personnel in the smart factory, and the point cloud data of the smart factory.

4. The intelligent factory personnel positioning and scheduling system based on 3D modeling according to claim 3, characterized in that: The process of the three-dimensional positioning module for obtaining the real-time positioning error of the personnel in the smart factory using the Euclidean distance formula includes: The real-time positioning error of the personnel in the smart factory is divided into the real-time positioning error of the personnel in the smart factory caused by dust concentration and the real-time positioning error of the personnel in the smart factory caused by electromagnetic noise; Extract the real-time three-dimensional positioning coordinates and the reference three-dimensional positioning coordinates of the personnel in the smart factory corresponding to the collection timestamp of the dust concentration in the smart factory environment, and use the Euclidean distance formula to calculate and obtain the real-time positioning error of the personnel in the smart factory caused by dust concentration; Extract the real-time three-dimensional positioning coordinates and the reference three-dimensional positioning coordinates of the personnel in the smart factory corresponding to the collection timestamp of the electromagnetic noise in the smart factory environment, and use the Euclidean distance formula to calculate and obtain the real-time positioning error of the personnel in the smart factory caused by electromagnetic noise.

5. The intelligent factory personnel positioning and scheduling system based on 3D modeling according to claim 4, wherein: The process of the three-dimensional positioning module for constructing a positioning error monitoring model includes: Using the random forest algorithm, a random forest model is constructed. The dust concentration in the intelligent factory environment, the electromagnetic noise in the intelligent factory environment, the real-time positioning error of intelligent factory personnel caused by the dust concentration, and the real-time positioning error of intelligent factory personnel caused by the electromagnetic noise are used as a data set, which is divided into a training set and a test set according to a ratio of 7:

3. The training set data is used to train the random forest model to learn the non-linear relationship between the dust concentration in the intelligent factory environment and the real-time positioning error of intelligent factory personnel caused by the dust concentration, and the non-linear relationship between the electromagnetic noise in the intelligent factory environment and the real-time positioning error of intelligent factory personnel caused by the electromagnetic noise, and a trained random forest model is obtained; the trained random forest is evaluated using the test set data, the random forest model parameters are adjusted, the random forest model is optimized, and the optimized random forest model is deployed in an intelligent factory personnel positioning and scheduling system based on three-dimensional modeling to obtain a positioning error monitoring model.

6. The intelligent factory personnel positioning and scheduling system based on 3D modeling according to claim 5, characterized in that: The process of the three-dimensional positioning module, in combination with the positioning error monitoring model, updating the real-time positioning of intelligent factory personnel and constructing a three-dimensional model of the intelligent factory to obtain real-time three-dimensional positioning data of intelligent factory personnel includes: Input the dust concentration in the intelligent factory environment and the electromagnetic noise in the intelligent factory environment into the positioning error monitoring model, and the positioning error monitoring model respectively outputs the real-time positioning error of intelligent factory personnel caused by the dust concentration and the real-time positioning error of intelligent factory personnel caused by the electromagnetic noise; According to the real-time positioning error of intelligent factory personnel caused by the dust concentration and the real-time positioning error of intelligent factory personnel caused by the electromagnetic noise, update the real-time positioning coordinates of intelligent factory personnel; Extract the neighborhood points of each point in the point cloud data of the intelligent factory, use the principal component analysis method to calculate the covariance matrix of each neighborhood point, perform eigenvalue decomposition on the covariance matrix of each neighborhood point to obtain eigenvalues and eigenvectors, and extract the plane features in the point cloud data of the intelligent factory through the eigenvalues and eigenvectors; calculate the normal vector of each point in the point cloud data of the intelligent factory, and extract the edge features in the point cloud data of the intelligent factory according to the angle between the normal vector of each point in the point cloud data of the intelligent factory and the normal vector of its adjacent points; use the Harris corner detection algorithm to extract the corner features in the point cloud data of the intelligent factory; Based on the extracted plane features, edge features and corner features, use the Delaunay triangulation algorithm to perform Delaunay triangulation on the point cloud data of the intelligent factory, connect the discrete points in the point cloud data of each intelligent factory into triangular patches, and construct a three-dimensional model of the intelligent factory; Map the updated real-time three-dimensional positioning coordinates of intelligent factory personnel into the three-dimensional model of the intelligent factory to obtain real-time three-dimensional positioning data of intelligent factory personnel.

7. The intelligent factory personnel positioning and scheduling system based on 3D modeling according to claim 6, characterized in that: The process of the environmental monitoring module constructing a dust concentration monitoring model through the linear regression algorithm includes: Taking the real-time positioning error of personnel in the smart factory caused by the dust concentration as the input variable and the influence index of the dust concentration on the positioning of personnel in the smart factory as the output variable, a linear regression model is constructed using the linear regression algorithm. The expression of this linear regression model is , where is the influence index of the dust concentration on the positioning of personnel in the smart factory, is the real-time positioning error of personnel in the smart factory caused by the dust concentration, is the intercept when the real-time positioning error of personnel in the smart factory caused by the dust concentration is 0, is the slope, is the error term, estimated using the least squares method and , and a dust concentration monitoring model is obtained.

8. The intelligent factory personnel positioning and scheduling system based on 3D modeling according to claim 7, characterized in that: The process of the environmental monitoring module constructing an electromagnetic noise monitoring model through the linear regression algorithm includes: Taking the real-time positioning error of smart factory personnel caused by electromagnetic noise as the input variable and the influence index of electromagnetic noise on the positioning of smart factory personnel as the output variable, a linear regression model is constructed using the linear regression algorithm. The expression of this linear regression model is , where is the influence index of electromagnetic noise on the positioning of smart factory personnel, is the real-time positioning error of smart factory personnel caused by electromagnetic noise, is the intercept when the real-time positioning error of smart factory personnel caused by electromagnetic noise is 0, is the slope, is the error term, estimated using the least squares method and , to obtain the electromagnetic noise monitoring model.

9. The intelligent factory personnel positioning and scheduling system based on 3D modeling according to claim 8, characterized in that: The process of the model fusion module using the neural network algorithm to fuse the dust concentration monitoring model and the electromagnetic noise monitoring model to obtain the environmental monitoring model includes: Input the real-time positioning error of the intelligent factory personnel caused by the dust concentration into the dust concentration monitoring model to obtain the influence index of the dust concentration on the positioning of the intelligent factory personnel; input the real-time positioning error of the intelligent factory personnel caused by the electromagnetic noise into the electromagnetic noise monitoring model to obtain the influence index of the electromagnetic noise on the positioning of the intelligent factory personnel. Use the neural network algorithm to build a neural network model. Take the real-time positioning error of the intelligent factory personnel caused by the dust concentration, the real-time positioning error of the intelligent factory personnel caused by the electromagnetic noise, the influence index of the dust concentration on the positioning of the intelligent factory personnel, and the influence index of the electromagnetic noise on the positioning of the intelligent factory personnel as the data set, and divide it into a training set and a test set according to the ratio of 7:

3. Select MLP as the neural network structure. The input layer includes two neurons, which receive the real-time positioning error of the intelligent factory personnel caused by the dust concentration and the real-time positioning error of the intelligent factory personnel caused by the electromagnetic noise. The hidden layer is configured with the MSE function. The output layer includes two neurons, which output the influence index of the dust concentration on the positioning of the intelligent factory personnel and the influence index of the electromagnetic noise on the positioning of the intelligent factory personnel. Input the training set data into the neural network model, set the learning rate to 0.01, and the number of iterative training times to 1000. The training process includes forward propagation and backward propagation. Among them, forward propagation is used to calculate the predicted output data, and backward propagation is used to update the weights and biases of the model. By repeating the iterative training, learn the non-linear relationship between the real-time positioning error of the intelligent factory personnel caused by the dust concentration and the influence index of the dust concentration on the positioning of the intelligent factory personnel, and the non-linear relationship between the real-time positioning error of the intelligent factory personnel caused by the electromagnetic noise and the influence index of the electromagnetic noise on the positioning of the intelligent factory personnel, until the set number of iterative training times is reached, and obtain the trained neural network model. Input the test set data into the trained neural network model, use the MSE function to evaluate the error between the output value and the actual value of the neural network model, and adjust the parameters of the neural network model according to the evaluation result to obtain the environmental monitoring model.

10. A personnel positioning and scheduling system for an intelligent factory based on 3D modeling according to claim 9, characterized in that: The process of the personnel scheduling module scheduling the intelligent factory personnel to take corresponding measures includes: Input the real-time positioning error of the intelligent factory personnel caused by the dust concentration and the real-time positioning error of the intelligent factory personnel caused by the electromagnetic noise into the environmental monitoring model, and the environmental monitoring model outputs the influence index of the dust concentration on the positioning of the intelligent factory personnel and the influence index of the electromagnetic noise on the positioning of the intelligent factory personnel. When the influence index of dust concentration on the personnel positioning in the smart factory is lower than 0.4, the influence of dust concentration on the personnel positioning in the smart factory is small; when the influence index of dust concentration on the personnel positioning in the smart factory is higher than 0.4, it indicates that the dust concentration is abnormal and has a great influence on the personnel positioning in the smart factory. Extract the three-dimensional positioning coordinates of the abnormal dust concentration from the three-dimensional model of the smart factory. According to the real-time three-dimensional positioning data of the smart factory personnel, through the Euclidean formula, obtain the distance between the three-dimensional positioning coordinates of the abnormal dust concentration and the three-dimensional positioning data of the smart factory personnel, extract the smart factory personnel corresponding to the minimum distance value, and dispatch this smart factory personnel to cope with the interference of dust concentration; When the influence index of electromagnetic noise on the personnel positioning in the smart factory is lower than 0.3, the influence of electromagnetic noise on the personnel positioning in the smart factory is small; when the influence index of electromagnetic noise on the personnel positioning in the smart factory is higher than 0.3, it indicates that the electromagnetic noise is abnormal and has a great influence on the personnel positioning in the smart factory. Extract the three-dimensional positioning coordinates of the abnormal electromagnetic noise from the three-dimensional model of the smart factory. According to the real-time three-dimensional positioning data of the smart factory personnel, through the Euclidean formula, obtain the distance between the three-dimensional positioning coordinates of the abnormal electromagnetic noise and the three-dimensional positioning data of the smart factory personnel, extract the smart factory personnel corresponding to the minimum distance value, and dispatch this smart factory personnel to cope with the interference of electromagnetic noise.

Images