Production park personnel positioning method and system

Abstract

The invention provides a production park personnel positioning method and system, and the method comprises the steps: training a personnel behavior model through obtaining the historical motion track and historical behavior data of personnel in a production park; acquiring historical environment dynamic change data in the production park, and training an environment prediction model; the method comprises the following steps: acquiring multi-source data through each sensor, and preprocessing the multi-source data, the multi-source data comprising Wi-Fi signal strength indication data for representing position information of personnel in a production park, radio frequency data, visual data and inertial measurement data; according to a Kalman filtering algorithm, fusing the preprocessed multi-source data to obtain a first position; and calculating the position probability distribution of the personnel in the production park by adopting a Bayesian positioning algorithm in combination with the first position and the output results of the personnel behavior model and the environment prediction model, and determining the target position, thereby effectively realizing the accurate and real-time positioning of the personnel in the production park in the complex environment.

Description

Technical Field

[0001] The present invention belongs to the technical field of personnel positioning, and particularly relates to a method and system for positioning personnel in a production park. Background Art

[0002] In a production park, accurate positioning of personnel is of great significance for safe production and efficient management. Traditional positioning technologies, such as GPS (Global Positioning System), have a significant decline in positioning accuracy or even cannot work in indoor or occluded environments; while emerging technologies such as UWB (Ultra Wide Band) have high accuracy, but have problems of high cost and complex deployment. In addition, most of the existing positioning methods do not fully consider the dynamic changes in the production park environment and the behavior patterns of personnel, and it is difficult to achieve stable and accurate positioning in complex scenarios such as rapid personnel movement, multi-story buildings, and densely occluded areas (such as shelves and pipelines). Summary of the Invention

[0003] Based on this, the embodiments of the present invention provide a method and system for positioning personnel in a production park, aiming to achieve accurate and real-time positioning of personnel in the production park by integrating multiple positioning technologies and intelligent algorithms.

[0004] The first aspect of the embodiments of the present invention provides a method for positioning personnel in a production park, and the method includes: Obtaining the historical movement trajectory and historical behavior data of personnel in the production park, and training a personnel behavior model; Obtaining the historical environmental dynamic change data in the production park, and training an environment prediction model; Obtaining multi-source data through various sensors, and preprocessing the multi-source data, where the multi-source data includes Wi-Fi signal strength indication data, radio frequency data, visual data, and inertial measurement data for representing the position information of personnel in the production park; Fusing the preprocessed multi-source data according to the Kalman filtering algorithm to obtain a first position; Adopting the Bayesian positioning algorithm, combining the first position, the result obtained after inputting the first position into the trained personnel behavior model, and the result obtained after inputting the current environmental dynamic change data into the trained environment prediction model, calculating the position probability distribution of personnel in the production park, and determining the target position.

[0005] Further, the step of fusing the preprocessed multi-source data according to the Kalman filtering algorithm to obtain a first position includes: Defining a state vector of personnel movement, and establishing a prediction model, where the prediction model is represented by a state transition equation; The observation models of different types of sensors are established respectively, and the observation values of each sensor are combined into a unified observation vector to form an overall observation equation, which is represented by the observation equations and observation noise vectors corresponding to each sensor; In the state prediction process, the current state is predicted using the state estimate at the previous moment and the state transition equation. Meanwhile, the state covariance matrix is updated; In the update stage, the observation equation is expanded by the first-order Taylor series to calculate the Jacobian matrix; According to the actual observation value and the predicted observation value, the observation residual is calculated; According to the covariance matrix, the Jacobian matrix, and the noise covariance, the Kalman gain is calculated; The predicted state and the observation residual are fused using the Kalman gain to obtain the first position.

[0006] Furthermore, in the step of fusing the preprocessed multi-source data according to the Kalman filtering algorithm to obtain the first position, when the sampling frequencies of each sensor are different, the timestamps are aligned by interpolation or zero-order hold; when there are multiple sensors of the same type, the data source corresponding to the observation value is determined by nearest neighbor matching.

[0007] Furthermore, in the step of obtaining the historical movement trajectory and historical behavior data of the personnel in the production park and training the personnel behavior model, the architecture of the personnel behavior model is an LSTM network. The historical movement trajectory includes at least historical timestamps and historical positions, and the historical behavior data includes at least personnel basic information, inertial measurement data, work calendars, and task work orders. The personnel behavior model is used to output the position probability distribution within a preset time according to the input timestamps, positions, personnel basic information, inertial measurement data, work calendars, and task work orders.

[0008] Furthermore, in the step of obtaining the historical environmental dynamic change data in the production park and training the environmental prediction model, the architecture of the environmental prediction model is a graph neural network. The historical environmental dynamic change data includes environmental parameters, equipment start-stop states, and the intensity fluctuation ranges of Wi-Fi / RF signals in each area. Among them, the environmental prediction model is used to output the expected intensity fluctuation ranges of Wi-Fi / RF signals in each area according to the input environmental parameters and equipment start-stop states.

[0009] Furthermore, the steps of using the Bayesian positioning algorithm to calculate the position probability distribution of the personnel in the production park and determine the target position by combining the first position, the result obtained after inputting the first position into the trained personnel behavior model, and the result obtained after inputting the current environmental dynamic change data into the trained environmental prediction model include: Obtain the initial weights of various sensors according to the first position, and adjust the initial weights based on the inertial measurement data and the expected intensity fluctuation range of Wi-Fi / RF signals in the corresponding area of the first position to obtain the target weights; Calculate the first probability that a person appears at the first position according to historical data, and correct the first probability according to the output result of the person behavior model to obtain the second probability; Construct a sensor joint likelihood model according to the target weights and the observation probability density function, and perform credibility weighted fusion to obtain the likelihood result; Calculate the posterior probability according to the second probability and the likelihood result, and determine the maximum posterior estimate according to the posterior probability to obtain the target position.

[0010] Further, adjust the noise covariance according to the target weights of the sensors.

[0011] The second aspect of the embodiments of the present invention provides a personnel positioning system for a production park, which is used to implement the personnel positioning method for a production park provided by the first aspect of the embodiments of the present invention. The system includes: A first training module, configured to obtain the historical movement trajectories and historical behavior data of personnel in the production park, and train a personnel behavior model; A second training module, configured to obtain the historical environmental dynamic change data in the production park and train an environmental prediction model; An acquisition module, configured to acquire multi-source data through various sensors and preprocess the multi-source data. The multi-source data includes Wi-Fi signal strength indication data, RF data, visual data, and inertial measurement data for representing the position information of personnel in the production park; A fusion module, configured to fuse the preprocessed multi-source data according to the Kalman filtering algorithm to obtain a first position; A calculation module, configured to adopt a Bayesian positioning algorithm, combine the first position, the result obtained after inputting the first position into the trained personnel behavior model, and the result obtained after inputting the current environmental dynamic change data into the trained environmental prediction model, calculate the position probability distribution of personnel in the production park, and determine the target position.

[0012] The third aspect of the embodiments of the present invention provides a computer-readable storage medium, including: The readable storage medium stores one or more programs, which when executed by a processor implement the personnel positioning method for a production park as described in the first aspect.

[0013] The fourth aspect of the embodiments of the present invention provides an electronic device, which includes a memory and a processor, wherein: The memory is used to store a computer program; When the processor is used to execute the computer program stored in the memory, the personnel positioning method for the production park described in the first aspect is implemented.

[0014] A personnel positioning method and system provided in an embodiment of the present invention, by obtaining the historical movement trajectory and historical behavior data of personnel in the production park, training a personnel behavior model; obtaining the historical environmental dynamic change data in the production park, training an environmental prediction model; obtaining multi-source data through various sensors and preprocessing the multi-source data, the multi-source data includes Wi-Fi signal strength indication data, radio frequency data, visual data, and inertial measurement data for representing the position information of personnel in the production park; according to the Kalman filtering algorithm, fusing the preprocessed multi-source data to obtain a first position; using the Bayesian positioning algorithm, combining the first position, the result obtained after inputting the first position into the trained personnel behavior model, and the result obtained after inputting the current environmental dynamic change data into the trained environmental prediction model, calculating the position probability distribution of personnel in the production park, and determining the target position, effectively realizing the accurate and real-time positioning of personnel in the production park in a complex environment. Description of the Drawings

[0015] Figure 1 It is a flowchart for implementing a personnel positioning method for a production park provided in Embodiment 1 of the present invention; Figure 2 It is a structural block diagram of a personnel positioning system for a production park provided in Embodiment 3 of the present invention; Figure 3 It is a structural block diagram of an electronic device provided in Embodiment 4 of the present invention. Detailed Embodiments

[0016] To facilitate the understanding of the present invention, the present invention will be described more comprehensively below with reference to the relevant drawings. Several embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, these embodiments are provided to make the disclosure of the present invention more thorough and comprehensive.

[0017] It should be noted that when an element is referred to as being "fixed to" another element, it can be directly on the other element or there can also be an intermediate element. When an element is considered to be "connected" to another element, it can be directly connected to the other element or there may be an intermediate element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are only for the purpose of illustration.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field to which this invention belongs. The terms used in the specification of this invention are for the purpose of describing specific embodiments only and are not intended to limit the invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0019] Embodiment 1 Embodiment 1 of the present invention provides a method for locating personnel in a production park. Please refer to Figure 1 which is a flowchart for implementing a method for locating personnel in a production park, specifically including steps S01 to S05.

[0020] Step S01: Obtain the historical movement trajectory and historical behavior data of personnel in the production park, and train a personnel behavior model.

[0021] Specifically, the architecture of the personnel behavior model is an LSTM network. The historical movement trajectory at least includes historical timestamps and historical locations. The historical behavior data at least includes basic personnel information, inertial measurement data, work calendars, and task work orders. The personnel behavior model is used to output the position probability distribution within a preset time according to the input timestamps, locations, basic personnel information, inertial measurement data, work calendars, and task work orders. According to the position probability distribution within the preset time, the high-frequency activity areas of personnel in the future period can be understood. Among them, the inertial measurement data can be obtained through a nine-axis sensor (accelerometer, gyroscope, magnetometer) built into the work badge worn by personnel.

[0022] Step S02: Obtain the historical environmental dynamic change data in the production park, and train an environmental prediction model.

[0023] Specifically, the architecture of the environmental prediction model is a graph neural network, with nodes being park grids and edges being signal propagation paths. The historical environmental dynamic change data includes environmental parameters, equipment start-stop status, and the intensity fluctuation range of Wi-Fi / RF signals in each area. Among them, the environmental prediction model is used to output the expected intensity fluctuation range of Wi-Fi / RF signals in each area according to the input environmental parameters and equipment start-stop status. Among them, the environmental parameters at least include temperature, humidity, and electromagnetic interference intensity. The equipment start-stop status includes forklift movement trajectories, machine operating power, etc. It can be understood that in subsequent steps, since the first position can be determined, the environmental dynamic change data at this first position can be obtained when the first position is known. In addition, environmental dynamic changes generally do not affect visual data and inertial measurement data. Therefore, the output of the environmental prediction model is only the intensity fluctuation range of Wi-Fi / RF signals in each area and does not include the intensity fluctuation range of visual and inertial signals in each area.

[0024] Step S03: Obtain multi-source data through various sensors and preprocess the multi-source data. The multi-source data includes Wi-Fi signal strength indication data, radio frequency data, visual data, and inertial measurement data for representing the position information of personnel in the production park.

[0025] It should be noted that for the Wi-Fi signal strength indication data, Wi-Fi access points (APs) are deployed in a grid layout (such as every 20 meters × 20 meters) in the production park. Terminal devices (such as work badges) scan the AP signals at a frequency of 5 Hz, record the MAC address, signal strength (dBm), and timestamp. The data preprocessing includes eliminating invalid data with a signal strength lower than a threshold (such as -90 dBm), taking the average of five consecutive sampling values of the same AP to reduce random noise, and converting the Wi-Fi signal strength indication data into a grid coordinate probability distribution through a fingerprint library matching algorithm (such as KNN, SVM); for the radio frequency data, RFID readers can be deployed at key positions such as entrances and exits, the boundaries of dangerous areas, and equipment workstations. Personnel wear active tags, and when the reader is triggered, the tag ID, reading time, and reader coordinates are recorded. The data preprocessing includes filtering duplicate trigger data (such as multiple readings of the same tag within 1 second), associating the personnel identity and position information through the tag ID, and statistically calculating the residence time threshold for each area in combination with historical data; for the visual data, it is collected through industrial cameras, and personnel targets are detected in real time according to the YOLOv8 algorithm, and the pixel coordinates and confidence are output. The data preprocessing includes converting the pixel coordinates into world coordinates using the camera internal parameter matrix (K) and external parameter matrix (R, t), and for occlusion scenarios (such as when a person is partially blocked by a shelf), the DeepSORT algorithm is used to track the trajectory, and the missing data is filled in by interpolating historical coordinates; the preprocessing of inertial measurement data includes using the Mahony complementary filtering algorithm to fuse the accelerometer and gyroscope data, outputting the attitude angles (pitch angle, roll angle, heading angle), suppressing the accelerometer integration drift through the zero velocity update (ZUPT) algorithm, and resetting the displacement error when the personnel is stationary (speed ≤ 0.1 m / s for 1 second).

[0026] Step S04: According to the Kalman filtering algorithm, fuse the preprocessed multi-source data to obtain the first position.

[0027] In the embodiment of the present invention, first, when the sampling frequencies of the sensors are different, the timestamps are aligned through interpolation or zero-order hold; when there are multiple sensors of the same type, the data source corresponding to the observed value is determined through nearest neighbor matching. Further, a state vector of the personnel movement is defined, and a prediction model is established. The prediction model is represented by a state transition equation. Among them, according to the actual two-dimensional or three-dimensional requirements, the state vector is defined. Taking two dimensions as an example, the state vector is , xrepresents the horizontal axis, y represents the vertical coordinate, v x represents the velocity on the horizontal axis, v y represents the speed on the ordinate. Assuming that the personnel move at a uniform speed, the dynamic equation is used to describe the state transfer from time k−1 to k, that is, the state transfer equation is: ; Among them, the state transfer matrix F is: ; Δt is the time interval, is the process noise (obeying Gaussian distribution , is the covariance matrix); Observation models of different types of sensors are established respectively, and the observation values of each sensor are merged into a unified observation vector to form an overall observation equation. The overall observation equation is represented by the observation equation corresponding to each sensor and the observation noise vector. Specifically, the observation model of Wi-Fi signal strength indicator data is expressed as: ; A is the reference signal strength, n is the path loss index, For the AP location, in practice, it is necessary to jointly solve the Wi-Fi signal strength indication data of multiple APs; The observation model of RF data is expressed as: ; is the reader position, r is the effective detection radius of RF; The observation model of visual data is expressed as: ; is the position in the camera coordinate system, is the observation noise; In the inertial measurement data observation model, the acceleration a and angular velocity ω are directly measured, and the position and velocity are updated by integration; Then, according to the observation equation corresponding to each sensor , forming the overall observation equation: ; is the observation noise vector; In the state prediction process, the state estimation of the previous moment is used And the state transition equation predicts the current state, expressed as: ; is the predicted state; Meanwhile, update the state covariance matrix , expressed as: ; In the update stage, perform a first-order Taylor expansion on the predicted state of the observation equation and calculate the Jacobian matrix, expressed as: ; is the Jacobian matrix. Calculate the observation residual based on the actual observation value and the predicted observation value, expressed as: ; is the observation residual, is the actual observation value, is the predicted observation value; Calculate the Kalman gain based on the covariance matrix, the Jacobian matrix, and the noise covariance, expressed as: ; is the Kalman gain, is the noise covariance; Fuse the predicted state and the observation residual using the Kalman gain to obtain the first position, expressed as: ; is the first position. Additionally, the covariance update is expressed as: .

[0028] Step S05: Use the Bayesian localization algorithm to combine the first position, the result obtained after inputting the first position into the trained human behavior model, and the result obtained after inputting the current environmental dynamic change data into the trained environmental prediction model, calculate the position probability distribution of the person in the production park, and determine the target position.

[0029] Specifically, based on the first position, obtain the initial weights of various sensors, and adjust the initial weights according to the inertial measurement data and the expected intensity fluctuation range of Wi-Fi / RF signals in the corresponding area of the first position to obtain the target weights. It should be noted that according to the inertial measurement data and the corresponding mapping relationship, determine the behavior pattern matching factor of the person. Exemplarily, when the person is in the patrol mode, the behavior pattern matching factor is 1.2. Then, according to the size of the expected intensity fluctuation range and the corresponding mapping relationship, determine the environmental impact factor. Exemplarily, the larger the expected intensity fluctuation range, the smaller the environmental impact factor. Finally, multiply the initial weight, the behavior pattern matching factor, and the environmental impact factor to adjust the initial weight and perform normalization processing to obtain the target weight to ensure that the sum of the weights is 1; Calculate the first probability that the person appears at the first position according to historical data. It can be understood that the historical residence time of the person at the first position is divided by the total historical residence time, and correct the first probability according to the output result of the person behavior model to obtain the second probability, that is, multiply the probability that the person is located at the first position output by the LSTM network by the first probability to obtain the second probability; Construct a sensor joint likelihood model according to the target weight and the observation probability density function, expressed as: ; is the target weight, is the observation probability density function of sensor i at position s, and n is the total number of sensors; Subsequently, perform credibility weighted fusion to obtain a likelihood result, expressed as: ; Calculate the posterior probability according to the second probability and the likelihood result, expressed as: ; is the second probability, is the likelihood result, and S is the set of all discrete grid points in the production park; Determine the maximum a posteriori estimate according to the posterior probability to obtain the target position, expressed as: ; is the target position. It can be understood that by traversing all grid points, find the position with the largest posterior probability.

[0030] In summary, a method for positioning personnel in a production park proposed in an embodiment of the present invention trains a personnel behavior model by obtaining the historical movement trajectories and historical behavior data of personnel in the production park; obtains the historical environmental dynamic change data in the production park and trains an environmental prediction model; obtains multi-source data through various sensors and preprocesses the multi-source data. The multi-source data includes Wi-Fi signal strength indication data, radio frequency data, visual data, and inertial measurement data for representing the position information of personnel in the production park; fuses the preprocessed multi-source data according to the Kalman filtering algorithm to obtain a first position; uses the Bayesian positioning algorithm to combine the first position, the result obtained after inputting the first position into the trained personnel behavior model, and the result obtained after inputting the current environmental dynamic change data into the trained environmental prediction model, calculates the position probability distribution of personnel in the production park, and determines the target position, effectively realizing accurate and real-time positioning of personnel in the production park in a complex environment.

[0031] Embodiment 2 The second embodiment of the present invention also provides a method for positioning personnel in a production park. The difference from the first embodiment of the present invention is that according to the target weight of the sensor, the noise covariance is adjusted. In the embodiment of the present invention, a mapping relationship between the weight and the noise covariance is established. The higher the credibility of the sensor (the greater the weight), the smaller its observation noise; conversely, the lower the credibility (the smaller the weight), the greater the noise.

[0032] It should be noted that since the target weight is determined after obtaining the first position, and during the process of determining the first position, the noise covariance corresponding to the preset initial weight is used in the Kalman filtering algorithm. In order to optimize the first position, the noise covariance corresponding to the target weight is used again to perform the operation of fusing the preprocessed multi-source data according to the Kalman filtering algorithm to obtain a new first position, and then the corresponding target weight is determined according to the new first position, and the above operations are cycled until the first position no longer changes, or stop after cycling a preset number of times, and perform the subsequent operation of determining the target position with the first position obtained after stopping.

[0033] Embodiment 3 The third embodiment of the present invention provides a personnel positioning system 200 for a production park. Please refer to Figure 2 , which is a structural block diagram of a personnel positioning method system for a production park. The personnel positioning system 200 for a production park includes: The first training module 21 is used to obtain the historical movement trajectory and historical behavior data of personnel in the production park, and train a personnel behavior model. The architecture of the personnel behavior model is an LSTM network. The historical movement trajectory includes at least historical timestamps and historical locations. The historical behavior data includes at least personnel basic information, inertial measurement data, work calendars, and task work orders. The personnel behavior model is used to output the position probability distribution within a preset time according to the input timestamps, locations, personnel basic information, inertial measurement data, work calendars, and task work orders. The second training module 22 is used to obtain the historical environmental dynamic change data in the production park and train an environmental prediction model. The architecture of the environmental prediction model is a graph neural network. The historical environmental dynamic change data includes environmental parameters, equipment start-stop states, and the intensity fluctuation ranges of Wi-Fi / radio frequency signals in each area. Among them, the environmental prediction model is used to output the expected intensity fluctuation ranges of Wi-Fi / radio frequency signals in each area according to the input environmental parameters and equipment start-stop states. The acquisition module 23 is used to obtain multi-source data through various sensors and preprocess the multi-source data. The multi-source data includes Wi-Fi signal strength indication data, radio frequency data, visual data, and inertial measurement data for representing the personnel location information in the production park. The fusion module 24 is used to fuse the preprocessed multi-source data according to the Kalman filtering algorithm to obtain the first position. When the sampling frequencies of the sensors are different, the timestamps are aligned by interpolation or zero-order hold. When there are multiple sensors of the same type, the data source corresponding to the observation value is determined by nearest neighbor matching. The calculation module 25 is used to adopt the Bayesian positioning algorithm, combine the first position, the result obtained after inputting the first position into the trained personnel behavior model, and the result obtained after inputting the current environmental dynamic change data into the trained environmental prediction model, calculate the position probability distribution of the personnel in the production park, and determine the target position.

[0034] Further, in some other embodiments of the present invention, the fusion module 24 includes: The first establishment unit is used to define the state vector of personnel movement and establish a prediction model. The prediction model is represented by a state transition equation. The second establishment unit is used to establish the observation models of different types of sensors respectively, and merge the observation values of each sensor into a unified observation vector to form an overall observation equation. The overall observation equation is represented by the observation equations corresponding to each sensor and the observation noise vector. The prediction unit is used to predict the current state by using the previous moment state estimate and the state transition equation during the state prediction process, and at the same time, update the state covariance matrix. A first calculation unit, configured to perform a first-order Taylor expansion on the observation equation and calculate the Jacobian matrix during the update phase; A second calculation unit, configured to calculate an observation residual according to an actual observation value and a predicted observation value; A third calculation unit, configured to calculate a Kalman gain according to a covariance matrix, the Jacobian matrix, and a noise covariance. Additionally, adjust the noise covariance according to a target weight of a sensor; A fourth calculation unit, configured to fuse a predicted state and an observation residual by using the Kalman gain to obtain the first position.

[0035] Furthermore, in some other embodiments of the present invention, the calculation module 25 includes: An adjustment unit, configured to obtain initial weights of various sensors according to the first position, and adjust the initial weights according to inertial measurement data and an expected intensity fluctuation range of Wi-Fi / RF signals in a corresponding area of the first position to obtain target weights; A correction unit, configured to calculate a first probability that a person appears at the first position according to historical data, and correct the first probability according to an output result of the person behavior model to obtain a second probability; A fusion unit, configured to construct a sensor joint likelihood model according to the target weights and an observation probability density function, and perform credibility weighted fusion to obtain a likelihood result; A fifth calculation unit, configured to calculate a posterior probability according to the second probability and the likelihood result, and determine a maximum a posteriori estimate according to the posterior probability to obtain the target position.

[0036] Embodiment 4 Embodiment 4 of the present invention provides an electronic device. Please refer to Figure 3 , which is a structural block diagram of an electronic device, including a memory 20, a processor 10, and a computer program 30 stored in the memory and executable on the processor. When the processor 10 executes the computer program 30, the method for positioning personnel in a production park as described above is implemented.

[0037] Among them, in some embodiments, the processor 10 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips, configured to run program codes stored in the memory 20 or process data, such as executing an access restriction program, etc.

[0038] Among them, the memory 20 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (such as SD or DX memory, etc.), magnetic memory, magnetic disk, optical disc, etc. The memory 20 can be an internal storage unit of the electronic device in some embodiments, such as the hard disk of the electronic device. The memory 20 can also be an external storage device of the electronic device in other embodiments, such as a plug-in hard disk, a SmartMedia Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. equipped on the electronic device. Further, the memory 20 can also include both the internal storage unit and the external storage device of the electronic device. The memory 20 can be used not only to store the application software and various types of data of the electronic device, but also to temporarily store the data that has been output or will be output.

[0039] An embodiment of the present invention also proposes a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the personnel positioning method in the production park as described above.

[0040] Those skilled in the art can understand that the logic and / or steps represented in the flowchart or described in other ways herein, for example, can be considered as a definite sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or in combination with these instruction execution systems, apparatus, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

[0041] More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection part (electronic device) having one or more wirings, a portable computer disk cartridge (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, then editing, interpreting, or otherwise processing it as appropriate, and then storing it in a computer memory.

[0042] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one of the following techniques known in the art or a combination thereof can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0043] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0044] The above embodiments only represent several implementation manners of the present invention, and the description thereof is relatively specific and detailed, but it should not be construed as a limitation on the scope of the patent of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the appended claims.

Claims

1. A method for positioning personnel in a production park, characterized in that, The method includes: Obtaining the historical movement trajectories and historical behavior data of personnel in the production park, and training a personnel behavior model; Obtaining the historical environmental dynamic change data in the production park, and training an environment prediction model; Obtaining multi-source data through various sensors, and preprocessing the multi-source data, where the multi-source data includes Wi-Fi signal strength indication data, radio frequency data, visual data, and inertial measurement data for representing the position information of personnel in the production park; Fusing the preprocessed multi-source data according to the Kalman filtering algorithm to obtain a first position; Adopting the Bayesian positioning algorithm, combining the first position, the result obtained after inputting the first position into the trained personnel behavior model, and the result obtained after inputting the current environmental dynamic change data into the trained environment prediction model, calculating the position probability distribution of the personnel in the production park, and determining the target position.

2. The personnel positioning method for a production park according to claim 1, characterized in that, The step of fusing the preprocessed multi-source data according to the Kalman filtering algorithm to obtain a first position includes: Defining a state vector of personnel movement, and establishing a prediction model, where the prediction model is represented by a state transition equation; Respectively establishing observation models of different types of sensors, and combining the observed values of each sensor into a unified observation vector to form an overall observation equation, where the overall observation equation is represented by the observation equations corresponding to each sensor and an observation noise vector; In the state prediction process, predicting the current state by using the state estimation at the previous moment and the state transition equation, and at the same time, updating the state covariance matrix; In the update stage, performing a first-order Taylor expansion on the observation equation and calculating the Jacobian matrix; Calculating the observation residual according to the actual observed value and the predicted observed value; Calculating the Kalman gain according to the covariance matrix, the Jacobian matrix, and the noise covariance; Fusing the predicted state and the observation residual by using the Kalman gain to obtain the first position.

3. The personnel positioning method for a production park according to claim 2, wherein, In the step of fusing the preprocessed multi-source data according to the Kalman filtering algorithm to obtain a first position, when the sampling frequencies of each sensor are different, aligning the timestamps through interpolation or zero-order hold; when there are multiple sensors of the same type, determining the data source corresponding to the observed value through nearest neighbor matching.

4. The method for positioning personnel in a production park according to claim 3, characterized in that, In the step of obtaining the historical movement trajectories and historical behavior data of personnel in the production park and training a personnel behavior model, the architecture of the personnel behavior model is an LSTM network, the historical movement trajectories at least include historical timestamps and historical positions, the historical behavior data at least includes personnel basic information, inertial measurement data, work calendars, and task work orders, and the personnel behavior model is used to output the position probability distribution within a preset time according to the input timestamps, positions, personnel basic information, inertial measurement data, work calendars, and task work orders.

5. The method for positioning personnel in a production park according to claim 4, wherein In the step of obtaining historical environmental dynamic change data in the production park and training an environmental prediction model, the architecture of the environmental prediction model is a graph neural network. The historical environmental dynamic change data includes environmental parameters, equipment start / stop states, and the intensity fluctuation ranges of Wi-Fi / RF signals in each area. Among them, the environmental prediction model is used to output the expected intensity fluctuation ranges of Wi-Fi / RF signals in each area according to the input environmental parameters and equipment start / stop states.

6. The method for positioning personnel in a production park according to claim 5, wherein The steps of using the Bayesian positioning algorithm, combining the first position, the result obtained after inputting the first position into the trained human behavior model, and the result obtained after inputting the current environmental dynamic change data into the trained environmental prediction model, calculating the position probability distribution of the person in the production park, and determining the target position include: According to the first position, obtain the initial weights of various sensors, and adjust the initial weights according to the inertial measurement data and the expected intensity fluctuation range of the Wi-Fi / RF signal in the area corresponding to the first position to obtain the target weights; According to historical data, calculate the first probability that the person appears at the first position, and correct the first probability according to the output result of the human behavior model to obtain the second probability; According to the target weights and the observation probability density function, construct a sensor joint likelihood model and perform credibility weighted fusion to obtain a likelihood result; According to the second probability and the likelihood result, calculate the posterior probability, and according to the posterior probability, determine the maximum a posteriori estimate to obtain the target position.

7. The method for positioning personnel in a production park according to claim 6, wherein, Adjust the noise covariance according to the target weights of the sensors.

8. A personnel positioning system for a production park, characterized in that, For implementing the production park personnel positioning method as described in any one of claims 1-7, the system includes: A first training module for obtaining the historical movement trajectories and historical behavior data of the personnel in the production park and training a human behavior model; A second training module for obtaining the historical environmental dynamic change data in the production park and training an environmental prediction model; An acquisition module for acquiring multi-source data through various sensors and preprocessing the multi-source data. The multi-source data includes Wi-Fi signal strength indication data, RF data, visual data, and inertial measurement data for representing the position information of the personnel in the production park; A fusion module for fusing the preprocessed multi-source data according to the Kalman filtering algorithm to obtain a first position; A calculation module for using the Bayesian positioning algorithm, combining the first position, the result obtained after inputting the first position into the trained human behavior model, and the result obtained after inputting the current environmental dynamic change data into the trained environmental prediction model, calculating the position probability distribution of the person in the production park, and determining the target position.

9. A computer-readable storage medium, characterized in that, Including: The readable storage medium stores one or more programs, and when the program is executed by a processor, it implements the production park personnel positioning method as described in any one of claims 1-7.

10. An electronic device, characterized in that, The electronic device includes a memory and a processor, where: The memory is used to store a computer program; When the processor is used to execute the computer program stored on the memory, the personnel positioning method for the production park described in any one of claims 1-7 is implemented.

Images