Personnel positioning method and system based on multi-source data

Abstract

The invention discloses a personnel positioning method and system based on multi-source data, and relates to the technical field of personnel positioning. The method comprises the following steps: for various site types, determining a corresponding optimal personnel positioning fusion method based on a plurality of real position data of experimenters in corresponding sites and a plurality of multi-source positioning data obtained by measuring the experimenters by a plurality of different personnel positioning subsystems; then, according to a plurality of personnel positioning real-time coordinates obtained by measuring the target personnel by the plurality of personnel positioning subsystems and known positions of all sites, the real-time site of the target personnel is determined, and finally, according to the plurality of personnel positioning real-time coordinates, an optimal personnel positioning fusion method corresponding to the site type of the real-time site is adopted, so that the real-time positioning of the target personnel is realized. And the final real-time coordinates of personnel positioning are obtained through fusion, so that the optimal fusion method can be selected at different sites for multi-source positioning data fusion through site and positioning fusion method decision making, and the personnel positioning accuracy is effectively improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of personnel positioning, and particularly relates to a personnel positioning method and system based on multi-source data. Background Art

[0002] Personnel positioning technology refers to a technical method for determining the position of one or more personnel. With the continuous development of technologies such as intelligentization and the Internet of Things, personnel positioning technology has become increasingly mature and widely used.

[0003] At present, the common personnel positioning technologies mainly include the following: (1) GPS positioning technology. GPS is the abbreviation of Global Positioning System, which is a technology that calculates the position of the positioning terminal through satellite signals. The GPS positioning technology needs to receive signals from at least three satellites and locates by calculating the time delay of these signals. (2) Beidou positioning technology. Beidou is a global satellite navigation and positioning system independently developed by China. It consists of 5 geosynchronous orbit satellites and more than 30 medium circular orbit satellites, and can provide global coverage of positioning, navigation, time distribution and short message services. Compared with GPS, the Beidou positioning technology has more satellites, a larger signal coverage area and higher positioning accuracy. (3) RTK positioning technology. RTK (Real-time kinematic) positioning technology is one of the most advanced high-precision positioning technologies at present. It eliminates the errors brought by satellite signals during the propagation process by introducing the signal of a reference station, so as to achieve centimeter-level high-precision positioning. At the same time, the RTK positioning technology has the characteristics of fast data update, high precision and strong real-time performance, and is widely used in fields such as port ships, surveying and mapping, geological exploration, industrial production, etc., and can meet the precise positioning requirements of different needs. With the development of the global satellite navigation system, the RTK positioning technology has become cheaper and more practical, and has become an indispensable key technology in many industries. (4) RFID positioning technology. Radio Frequency Identification (RFID) is a technology that realizes the identification and positioning of objects through electromagnetic waves. The RFID positioning system consists of a reader, an antenna and a tag, etc. The reader sends electromagnetic waves to the tag through the antenna, and the tag returns the identification information to the reader after receiving the electromagnetic waves to complete the identification and positioning of the object. The RFID positioning technology is widely used in occasions such as logistics, warehousing and libraries. (5) WiFi positioning technology. WiFi positioning uses the wireless network, cooperates with WiFi tags and related mobile terminal devices, and determines the position of relevant personnel by the characteristics of the WiFi signal strength, frequency and signal-to-noise ratio, etc., and combines the corresponding positioning algorithm. The WiFi positioning technology has a wide range of applications, covering fields such as hospitals, shopping malls, hotels and airports. (6) Bluetooth positioning technology. Bluetooth positioning technology is a low-power, easy-to-implement and widely used positioning technology. According to the different positioning terminals, the Bluetooth positioning method is divided into active positioning, passive positioning and active-passive integrated positioning. The Bluetooth positioning technology is widely used and suitable for various indoor positioning scenarios.(7) UWB positioning technology, that is, UWB (Ultra Wide Band) positioning uses a wireless carrier communication technology with a frequency bandwidth above 1 GHz, and transmits data using non-sinusoidal narrow pulses at the nanosecond level. This is a wireless communication technology with extremely high transmission rate and extremely wide bandwidth. Compared with other technologies, UWB positioning technology has the advantages of high positioning accuracy and strong resistance to signal interference, and has a relatively wide range of applications; (8) Ultrasonic positioning technology, that is, ultrasonic positioning mainly uses the reflection ranging method to determine the position of an object through methods such as multi-lateral positioning. The system consists of a main rangefinder and several receivers. The main rangefinder can be placed on the target to be measured, and the receivers are fixed in the indoor environment. When positioning, a signal with the same frequency is transmitted to the receivers, and after being received, the receivers reflect and transmit it back to the main rangefinder. The distance is calculated based on the time difference between the echo and the transmitted wave, so as to determine the position; etc.;

[0004] When performing personnel positioning currently, usually only a single data source is used to achieve personnel positioning. For example: personnel positioning is achieved based on a single data source such as Bluetooth, ultra-wideband, or GPS. However, the site environment where the personnel are located will change with the activities of the personnel, and different data sources have different positioning accuracies for different site environments (such as outdoor open spaces, outdoor non-open spaces, indoor low-rise rooms, indoor high-rise rooms, and basements, etc.). If only a single data source is used to achieve personnel positioning, it will be impossible to ensure high positioning accuracy in different terrain environments, that is, both the positioning reliability and positioning accuracy are relatively poor.

[0005] The existing patent "CN109826668A, Underground Multi-source Precise Personnel Positioning System and Method" provides an underground multi-source precise personnel positioning solution, that is, it includes: a camera device, which is used to collect image information of underground personnel; a positioning identification card, which is used to record the identity information of underground personnel and is carried by underground personnel; a positioning base station, which is used to collect wireless signals sent by the positioning identification card in real time; a host computer, which is used to process the information collected at the camera device and the positioning base station; and a data transmission device, which is used to realize data transmission between the camera device, the positioning base station and the host computer. Although this solution has better underground positioning accuracy, the fusion of its video positioning data and wireless positioning data is used equally, without considering their respective positioning accuracy differences in the underground environment, which limits the improvement of positioning accuracy when used equally.

[0006] The existing patent "CN114339609A, A Personnel Positioning Method and Device" provides a personnel positioning solution, that is, obtaining first personnel positioning data carrying a target positioning card flag accessed from a target manufacturer, and the first personnel positioning data is collected through any data source; converting the first personnel positioning data into a unified data format to obtain second personnel positioning data; determining a target object bound to the target positioning card flag from a three-dimensional positioning system; according to the mapping table and the building number and floor value in the second personnel positioning data, attaching the three-dimensional scene height and floor block number corresponding to the floor value to the second personnel positioning data; using the second personnel positioning data attached with the three-dimensional scene height and floor block number to display the area where the target object is located, the icon of the target object, and the detailed information of the area where the target object is located in the three-dimensional map of the three-dimensional positioning system. Although this solution realizes the processing of personnel positioning data from multiple data sources and can support the display of three-dimensional scenes, improving the reliability and accuracy of positioning, it also does not consider the positioning accuracy differences of different positioning data on different floors, so its positioning accuracy still needs to be improved.

[0007] In summary, how to provide a new solution that integrates multiple positioning technologies such as GPS, Wi-Fi, Bluetooth, and UWB and can accurately locate target personnel in complex indoor and outdoor scenarios, so as to provide a real-time personnel monitoring function for emergency rescue, personnel management, and public safety and other affairs, is an urgent research topic for those skilled in the art. Summary of the Invention

[0008] The object of the present invention is to provide a personnel positioning method, system, computer device, computer-readable storage medium, and computer program product based on multi-source data, so as to solve the problem that the positioning accuracy of the existing personnel positioning technical solution still needs to be improved because it does not consider the positioning accuracy differences of different positioning data in different venues.

[0009] To achieve the above object, the present invention adopts the following technical solutions:

[0010] In the first aspect, a personnel positioning method based on multi-source data is provided, including:

[0011] For various site types in a target area, obtaining multiple real position data of experimental personnel in the corresponding site and multiple pieces of multi-source positioning data respectively measured by different multiple personnel positioning subsystems for the experimental personnel and corresponding one-to-one with the multiple real position data, wherein the multi-source positioning data includes multiple personnel positioning coordinates corresponding one-to-one with the multiple personnel positioning subsystems;

[0012] For the various site types, the corresponding multiple multi-source positioning data are used as multiple test sample input items, and the corresponding multiple true position data are used as multiple test sample output items corresponding one by one to the multiple test sample input items. Then, the multiple test sample input items and the multiple test sample output items are applied to perform fusion verification on multiple personnel positioning fusion methods for fusing personnel positioning final coordinates based on multiple personnel positioning coordinates, obtaining multiple verification scores corresponding one by one to the multiple personnel positioning fusion methods, and based on the multiple verification scores, determining the corresponding optimal personnel positioning fusion method from the multiple personnel positioning fusion methods;

[0013] Receive multiple real-time personnel positioning coordinates measured by the multiple personnel positioning subsystems for a target person in the target area, where the multiple real-time personnel positioning coordinates correspond one by one to the multiple personnel positioning subsystems;

[0014] Based on the multiple real-time personnel positioning coordinates and the known positions of each site in the target area, determine the real-time site where the target person is located in the target area;

[0015] Based on the multiple real-time personnel positioning coordinates, use the optimal personnel positioning fusion method corresponding to the site type of the real-time site where the target person is located to fuse and obtain the final real-time personnel positioning coordinates of the target person.

[0016] Based on the above invention content, a new solution for integrating multiple positioning technologies to accurately locate a target person in complex indoor and outdoor scenarios is provided. That is, first, for various site types, based on the multiple true position data of the experimental personnel in the corresponding site and the multiple multi-source positioning data measured by different multiple personnel positioning subsystems for the experimental personnel respectively, determine the corresponding optimal personnel positioning fusion method. Then, based on the multiple real-time personnel positioning coordinates measured by the multiple personnel positioning subsystems for the target person and the known positions of each site, determine the real-time site where the target person is located. Finally, based on the multiple real-time personnel positioning coordinates, use the optimal personnel positioning fusion method corresponding to the site type of the real-time site where the target person is located to fuse and obtain the final real-time personnel positioning coordinates. In this way, considering the positioning accuracy differences of different positioning data in different sites, by making decisions on the site and its positioning fusion method, the best fusion method can be selected for multi-source positioning data fusion in different sites, achieving the purpose of effectively improving the accuracy of personnel positioning. Furthermore, it can be used to provide a personnel real-time monitoring function for emergency rescue, personnel management, public safety and other affairs, which is convenient for practical application and promotion.

[0017] In a possible design, the multiple personnel positioning subsystems include a first personnel positioning subsystem based on satellite positioning technology, a second personnel positioning subsystem based on wireless positioning technology, a third personnel positioning subsystem based on video recognition technology, a fourth personnel positioning subsystem based on millimeter-wave radar technology, and / or a fifth personnel positioning subsystem based on ultrasonic positioning technology. Among them, the first personnel positioning subsystem includes a satellite locator for binding personnel, the second personnel positioning subsystem includes at least three position-fixed and position-known positioning base stations and a positioning tag for binding personnel, the third personnel positioning subsystem includes a plurality of AI cameras arranged at different positions within the target area, the fourth personnel positioning subsystem includes a plurality of millimeter-wave radars arranged at different positions within the target area, and the fifth personnel positioning subsystem includes at least three position-fixed and position-known ultrasonic reflectors and an ultrasonic main rangefinder for binding personnel.

[0018] In a possible design, the multiple personnel positioning fusion methods include a simple averaging method, a weighted averaging method, and / or a stacking generalization method.

[0019] In a possible design, when the optimal personnel positioning fusion method is the stacking generalization method, after determining the optimal personnel positioning fusion method corresponding to the site type of the real-time location from the multiple personnel positioning fusion methods and before applying this optimal personnel positioning fusion method, the method further includes:

[0020] Applying the multiple pieces of multi-source positioning data corresponding to the site type of the real-time location and the multiple true position data, and optimizing the hyperparameters of the meta-learner in the optimal personnel positioning fusion method corresponding to the site type of the real-time location based on an optimization algorithm to obtain the hyperparameters and the optimal search result for making the final coordinates of personnel positioning the most accurate.

[0021] In a possible design, the optimization algorithm adopts a particle swarm optimization algorithm, a Newton optimization algorithm, a genetic optimization algorithm, a grey wolf optimization algorithm, a whale optimization algorithm, or a tuna swarm optimization algorithm.

[0022] In a possible design, applying the multiple pieces of multi-source positioning data corresponding to the site type of the real-time location and the multiple true position data, and optimizing the hyperparameters of the meta-learner in the optimal personnel positioning fusion method corresponding to the site type of the real-time location based on an optimization algorithm to obtain the hyperparameters and the optimal search result for making the final coordinates of personnel positioning the most accurate includes the following steps S501 to S516:

[0023] S501. Initialize the number of search individuals I, the maximum number of iterations T, and the first learning factor α of the search population containing the search population1 , the second learning factor α 2 , the third learning factor α 3 , the fourth learning factor α 4 , the fifth learning factor α 5 and the sixth learning factor α 6 Optimization algorithm parameters, and randomly generate the initial positions of each search individual in the search population, and then execute step S502, where the initial positions of each search individual are randomly generated based on the following formula:

[0024]

[0025] In the formula, i′ represents a positive integer less than or equal to I, d′ represents a positive integer less than or equal to D′, D′ represents the dimension number of the parameter vector to be optimized, and the parameter vector to be optimized includes the hyperparameters of the meta-learner in the optimal personnel positioning fusion method corresponding to the site type of the real-time location site, represents the component of the initial position of the i′-th search individual in the search population on the d′-th dimension of the parameter vector to be optimized, u c,d′ represents the upper limit of the parameter search space on the d′-th dimension, l c,d′ represents the lower limit of the parameter search space on the d′-th dimension, and rand(0, 1) represents a pure decimal random generation function;

[0026] S502. For each search individual, use the corresponding initial position as the hyperparameters of the meta-learner in the optimal personnel positioning fusion method corresponding to the site type of the real-time location site, and then apply the multiple multi-source positioning data and the multiple real position data corresponding to the site type of the real-time location site to perform fusion verification on this optimal personnel positioning fusion method, obtain the initial value of the corresponding MAPE index, and use this initial value of the MAPE index as the corresponding fitness, and then execute step S503;

[0027] S503. Use the initial position of a search individual with the minimum fitness as the initial global optimal position x best , and also initialize the current iteration number t′ = 0, and then execute step S504;

[0028] S504. Calculate the current average position of the search population based on the current positions of each search individual, and for each search individual, determine the corresponding first new position, and then execute step S505, where the first new position of each search individual is determined according to the following formula:

[0029] x new1,i′,d′ = x best,d′ + α1 ×rand(0,1)×(x mean,d′ -x i′,d′ )

[0030] In the formula, x new1,i′,d′ represents the component of the first new position of the i'-th search individual in the d'-th dimension, x best,d′ represents the component of the global optimal position x best in the d'-th dimension, x mean,d′ represents the component of the current average position of the search population in the d'-th dimension, x i′,d′ represents the component of the current position of the i'-th search individual in the d'-th dimension;

[0031] S505. For each search individual, use the corresponding first new position as the hyperparameter of the meta-learner in the optimal personnel positioning fusion method corresponding to the site type of the real-time location site, and then apply the multiple multi-source positioning data and the multiple real position data corresponding to the site type of the real-time location site to perform fusion verification on this optimal personnel positioning fusion method to obtain the corresponding new value of the MAPE index, and use this new value of the MAPE index as the corresponding and new fitness, and then execute step S506;

[0032] S506. Determine whether the fitness corresponding to the current position of the i'-th search individual is greater than the current fitness of the i'-th search individual. If so, update the current position of the i'-th search individual to the first new position of the i'-th search individual, and then execute step S507. Otherwise, directly execute step S507;

[0033] S507. Determine whether the fitness corresponding to the global optimal position x best is greater than the minimum value among the current fitnesses of each search individual. If so, update the global optimal position x best to the current position of any search individual with this minimum value, and then execute step S508. Otherwise, directly execute step S508;

[0034] S508. Update and calculate the current average position of the search population according to the current positions of each search individual, and determine the corresponding second new position for each search individual, and then execute step S509, where the second new position of each search individual is determined according to the following formula:

[0035]

[0036] In the formula, x new2,i′,d′represents the component of the second new position of the \(i'\)-th search individual in the \(d'\)-th dimension, \(\beta(i')\) represents the first intermediate variable corresponding to the \(i'\)-th search individual, \(\delta(i')\) represents the second intermediate variable corresponding to the \(i'\)-th search individual, \(i''\) represents a positive integer less than or equal to \(I\), \(\text{mod}()\) represents the modulo function, \(x\) i″,d′ represents the component of the current position of the \(i''\)-th search individual in the search population in the \(d'\)-th dimension, \(\beta_r(i')\) represents the third intermediate variable corresponding to the \(i'\)-th search individual, \(\delta_r(i')\) represents the fourth intermediate variable corresponding to the \(i'\)-th search individual, \(\max()\) represents the maximum value function, \(\theta(i')\) represents the fifth intermediate variable corresponding to the \(i'\)-th search individual, \(r(i')\) represents the sixth intermediate variable corresponding to the \(i'\)-th search individual, \(\pi\) represents 180 degrees;

[0037] S509. For each search individual, use the corresponding second new position as the hyperparameter of the meta-learner in the optimal personnel positioning fusion method corresponding to the site type of the real-time location. Then, apply the multiple multi-source positioning data and the multiple true position data corresponding to the site type of the real-time location to perform fusion verification on this optimal personnel positioning fusion method, obtain the new value of the corresponding MAPE index, and use this MAPE index new value as the corresponding and new fitness, and then execute step S510;

[0038] S510. Determine whether the fitness corresponding to the current position of the \(i'\)-th search individual is greater than the current fitness of the \(i'\)-th search individual. If so, update the current position of the \(i'\)-th search individual to the second new position of the \(i'\)-th search individual, and then execute step S511; otherwise, directly execute step S511;

[0039] S511. Determine whether the fitness corresponding to the global optimal position \(x\) best is greater than the minimum value among the current fitnesses of each search individual. If so, update the global optimal position \(x\) best to the current position of any search individual with this minimum value, and then execute step S512; otherwise, directly execute step S512;

[0040] S512. Update and calculate the current average position of the search population based on the current positions of each search individual, and for each search individual, determine the corresponding third new position, and then execute step S513, where the third new position of each search individual is determined according to the following formula:

[0041]

[0042] where \(x\)new3,i′,d′ Denote the component of the third new position of the \(i'\)-th search individual on the \(d'\)-th dimension. Denote the seventh intermediate variable corresponding to the \(i'\)-th search individual. Denote the eighth intermediate variable corresponding to the \(i'\)-th search individual. Denote the ninth intermediate variable corresponding to the \(i'\)-th search individual. Denote the tenth intermediate variable corresponding to the \(i'\)-th search individual. Denote the eleventh intermediate variable corresponding to the \(i'\)-th search individual.

[0043] S513. For each of the search individuals, use the corresponding third new position as the hyperparameter of the meta-learner in the optimal personnel positioning fusion method corresponding to the site type of the real-time location site, and then apply the multiple multi-source location data and the multiple real location data corresponding to the site type of the real-time location site to perform fusion verification on this optimal personnel positioning fusion method, obtain the new value of the corresponding MAPE index, and use this new value of the MAPE index as the corresponding and new fitness, and then execute step S514;

[0044] S514. Determine whether the fitness corresponding to the current position of the \(i'\)-th search individual is greater than the current fitness of the \(i'\)-th search individual. If so, update the current position of the \(i'\)-th search individual to the third new position of the \(i'\)-th search individual, and then execute step S515. Otherwise, directly execute step S515;

[0045] S515. Determine whether the fitness corresponding to the global optimal position \(x\) best is greater than the minimum value among the current fitnesses of all the search individuals. If so, update the global optimal position \(x\) best to the current position of any search individual with this minimum value, and then execute step S516. Otherwise, directly execute step S516;

[0046] S516. Increment the current iteration number \(t'\) by 1, and determine whether the current iteration number \(t'\) has reached the maximum iteration number \(T\). If so, use the global optimal position \(x\) best as the hyperparameter and the optimal search result for making the final coordinates of personnel positioning the most accurate. Otherwise, return to execute step S504.

[0047] In a possible design, according to the multiple real-time coordinates of personnel positioning and the known positions of each site in the target area, determine the real-time location site of the target personnel in the target area, including:

[0048] Determine the weight coefficients of each personnel positioning subsystem in the multiple personnel positioning subsystems according to the multiple multi-source positioning data corresponding to the various site types and the multiple true position data;

[0049] According to the multiple real-time personnel positioning coordinates and the known positions of each site in the target area, traverse each personnel positioning subsystem in the following manner: If the real-time personnel positioning coordinate corresponding to the currently traversed personnel positioning subsystem among the multiple real-time personnel positioning coordinates matches the known position of a certain site in the target area, then add the weight coefficient of the currently traversed personnel positioning subsystem to the voting score of the certain site, where the voting score of the certain site is initially zero before traversal;

[0050] Judge whether the highest voting score among the voting scores of all sites in the target area exceeds a preset threshold;

[0051] If so, determine any site corresponding to the highest voting score as the real-time location site of the target person in the target area.

[0052] In a possible design, determining the weight coefficients of each personnel positioning subsystem in the multiple personnel positioning subsystems according to the multiple multi-source positioning data corresponding to the various site types and the multiple true position data includes:

[0053] Extract all the personnel positioning coordinates measured by each personnel positioning subsystem in the multiple personnel positioning subsystems and all the true position data corresponding one-to-one to all the personnel positioning coordinates according to the multiple multi-source positioning data corresponding to the various site types and the multiple true position data;

[0054] For each personnel positioning subsystem, according to all the personnel positioning coordinates measured by the corresponding system for the experimental personnel and all the true position data corresponding one-to-one to all the personnel positioning coordinates, statistically obtain the corresponding average positioning deviation;

[0055] For each personnel positioning subsystem, calculate the corresponding weight coefficient according to the following formula:

[0056]

[0057] In the formula, M represents the total number of systems of the multiple personnel positioning subsystems, m represents a positive integer less than or equal to M, η m represents the weight coefficient of the m-th personnel positioning subsystem in the multiple personnel positioning subsystems, represents the average positioning deviation of the m-th personnel positioning subsystem, The normalized value representing the average positioning deviation of the m-th personnel positioning subsystem, where m' represents a positive integer less than or equal to M. Represents the average positioning deviation of the m'-th personnel positioning subsystem among the multiple personnel positioning subsystems. Represents the normalized value of the average positioning deviation of the m'-th personnel positioning subsystem.

[0058] In a possible design, after determining whether the highest voting score exceeds the preset threshold, if it is determined that the highest voting score does not exceed the preset threshold, the method further includes:

[0059] Selecting at least one venue with a non-zero voting score from all the venues;

[0060] For each venue among the at least one venue, according to the multiple real-time personnel positioning coordinates, using the best personnel positioning fusion method corresponding to the corresponding venue type, fusing to obtain the target personnel and the corresponding final real-time personnel positioning coordinates;

[0061] According to the final real-time personnel positioning coordinates of each venue among the at least one venue, the comprehensive final real-time personnel positioning coordinate P of the target personnel is calculated according to the following formula xyz :

[0062]

[0063] In the formula, K represents the total number of venues of the at least one venue, k represents a positive integer less than or equal to K, and P k,xyz Represents the final real-time personnel positioning coordinate of the k-th venue among the at least one venue, and γ k Represents the voting score of the k-th venue, k' represents a positive integer less than or equal to K, and γ k′ Represents the voting score of the k'-th venue among the at least one venue.

[0064] In a second aspect, a personnel positioning system based on multi-source data is provided, including a plurality of personnel positioning subsystems and a computer device respectively communicatively connected to the plurality of personnel positioning subsystems. Among them, the computer device is used to execute the personnel positioning method as described in the first aspect or any possible design in the first aspect.

[0065] In a third aspect, the present invention provides a computer device, including a memory, a processor, and a transceiver communicatively connected in sequence. Among them, the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program and execute the personnel positioning method as described in the first aspect or any possible design in the first aspect.

[0066] Fourthly, the present invention provides a computer-readable storage medium, on which instructions are stored. When the instructions are run on a computer, the personnel positioning method as described in the first aspect or any possible design in the first aspect is executed.

[0067] Fifthly, the present invention provides a computer program product, including a computer program or instructions. When the computer program or the instructions are executed by a computer, the personnel positioning method as described in the first aspect or any possible design in the first aspect is implemented.

[0068] Advantages of the above solutions:

[0069] (1) The present invention provides a new solution that integrates multiple positioning technologies to accurately locate target personnel in complex indoor and outdoor scenarios. That is, first, for various site types, based on multiple real position data of experimental personnel in the corresponding sites and multiple pieces of multi-source positioning data measured by different multiple personnel positioning subsystems for the experimental personnel, the corresponding optimal personnel positioning fusion method is determined. Then, according to the multiple real-time coordinates of the target personnel measured by multiple personnel positioning subsystems and the known positions of each site, the real-time site where the target personnel is located is determined. Finally, according to the multiple real-time coordinates of the personnel positioning, the optimal personnel positioning fusion method corresponding to the site type of the real-time site is used to fuse and obtain the final real-time coordinates of the personnel positioning. Considering the positioning accuracy differences of different positioning data in different sites, by making decisions on the site and its positioning fusion method, the best fusion method can be selected for multi-source positioning data fusion in different sites, achieving the purpose of effectively improving the accuracy of personnel positioning. Furthermore, it can be used to provide a real-time monitoring function for personnel in emergency rescue, personnel management, public safety and other affairs, which is convenient for practical application and promotion;

[0070] (2) The accuracy of personnel positioning can be further improved by optimizing the optimal personnel positioning fusion method, and based on the fusion of the particle swarm optimization algorithm and the whale optimization algorithm, it is beneficial to quickly and accurately obtain the optimal search result;

[0071] (3) When it is impossible to determine the real-time site where the target personnel is located, the comprehensive final real-time coordinates of the personnel positioning of the target personnel can be obtained based on the optimal personnel positioning fusion method and the weight voting score of all suspected sites where the target personnel is located, still effectively improving the accuracy of personnel positioning. Description of the Drawings

[0072] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0073] Figure 1 It is a schematic flowchart of the personnel positioning method based on multi-source data provided by the embodiments of this application.

[0074] Figure 2 It is a schematic flowchart of determining the venue where the target person is located in real time provided by the embodiments of this application.

[0075] Figure 3 It is a schematic flowchart of optimizing the hyperparameters of the meta-learner in the optimal personnel positioning fusion method based on an optimization algorithm provided by the embodiments of this application.

[0076] Figure 4 It is a schematic structural diagram of the personnel positioning system based on multi-source data provided by the embodiments of this application.

[0077] Figure 5 It is a schematic structural diagram of the computer device provided by the embodiments of this application. Detailed implementation manners

[0078] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the present invention in combination with the drawings and the description of the embodiments or the prior art. Obviously, the following description of the drawing structure is only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other embodiments can be obtained based on these embodiments. It should be noted here that the description of these embodiment modes is used to help understand the present invention, but does not constitute a limitation to the present invention.

[0079] It should be understood that although terms such as first and second may be used herein to describe various objects, these objects should not be limited by these terms. These terms are only used to distinguish one object from another. For example, the first object can be called the second object, and similarly, the second object can be called the first object, without departing from the scope of the exemplary embodiments of the present invention.

[0080] It should be understood that for the term "and / or" that may appear in this text, it is merely a description of the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can represent three situations: A exists alone, B exists alone, or both A and B exist simultaneously. Another example, A, B, and / or C can represent any one of A, B, and C or any combination of them. For the term " / and" that may appear in this text, it describes another association object relationship, indicating that two relationships can exist. For example, A / and B can represent two situations: A exists alone or both A and B exist simultaneously. Additionally, for the character " / " that may appear in this text, generally, it indicates that the associated objects before and after are in an "or" relationship.

[0081] Embodiment

[0082] As Figure 1 shown, the personnel positioning method provided in the first aspect of this embodiment and based on multi-source data can be, but is not limited to, executed by a computer device having certain computing resources and respectively communicatively connected to multiple personnel positioning subsystems. For example, it can be executed by an electronic device such as a cloud server, a personal computer (Personal Computer, PC, referring to a multi-purpose computer with a size, price, and performance suitable for personal use; desktop computers, laptops, small laptops, tablet computers, and ultrabooks all belong to personal computers), a smart phone, a personal digital assistant (Personal Digital Assistant, PDA), or a wearable device. As Figure 1 shown, the personnel positioning method can be, but is not limited to, including the following steps S1 to S5.

[0083] S1. For various site types within the target area, obtain multiple real position data of experimental personnel at the corresponding sites and multiple pieces of multi-source positioning data measured by different multiple personnel positioning subsystems for the experimental personnel and corresponding one-to-one with the multiple real position data. Among them, the multi-source positioning data includes, but is not limited to, multiple personnel positioning coordinates corresponding one-to-one with the multiple personnel positioning subsystems.

[0084] In the step S1, the target area is specifically but not limited to areas with complex indoor and outdoor scenarios such as factory areas or parks. Among them, the specific site types can be but are not limited to outdoor open types, outdoor non-open types, indoor low-rise types, indoor high-rise types, and / or basement types, etc. The experimental personnel refer to the objects participating in the personnel positioning experiment, and the real position data are the real personnel coordinates, which can be obtained through conventional measurement techniques. The multiple personnel positioning subsystems specifically include but are not limited to the first personnel positioning subsystem based on satellite positioning technology, the second personnel positioning subsystem based on wireless positioning technology, the third personnel positioning subsystem based on video recognition technology, the fourth personnel positioning subsystem based on millimeter-wave radar technology, and / or the fifth personnel positioning subsystem based on ultrasonic positioning technology, etc. Among them, the first personnel positioning subsystem includes but is not limited to satellite locators for binding personnel, etc.; the second personnel positioning subsystem includes but is not limited to at least three fixed-position and known-position positioning base stations and positioning tags for binding personnel, etc.; the third personnel positioning subsystem includes but is not limited to multiple AI (Artificial Intelligence) cameras arranged at different positions in the target area, etc.; the fourth personnel positioning subsystem includes but is not limited to multiple millimeter-wave radars arranged at different positions in the target area, etc.; the fifth personnel positioning subsystem includes but is not limited to at least three fixed-position and known-position ultrasonic reflectors and ultrasonic main rangefinders for binding personnel, etc. The aforementioned satellite positioning technology, wireless positioning technology, video recognition technology, millimeter-wave radar technology, and ultrasonic positioning technology, etc. are all existing positioning technologies. For example, the satellite positioning technology can adopt existing GPS positioning technology, Beidou positioning technology, and RTK positioning technology, etc.; the wireless positioning technology can adopt existing RFID positioning technology, WiFi positioning technology, Bluetooth positioning technology, and UWB positioning technology, etc.; the video recognition technology can be implemented with reference to the existing patent "CN112396658A, A Method and System for Indoor Personnel Positioning Based on Video", etc.; the millimeter-wave radar technology can be implemented with reference to the existing patent "CN111999726A, A Method for Personnel Positioning Based on Millimeter-Wave Radar", etc. In addition, the real position data can be input by the experimental personnel, and the multi-source positioning data can be uploaded by the multiple personnel positioning subsystems.

[0085] S2. For the various site types, use the corresponding multiple multi-source positioning data as multiple test sample input items, and use the corresponding multiple true location data as multiple test sample output items that correspond one-to-one to the multiple test sample input items. Then, apply the multiple test sample input items and the multiple test sample output items to perform fusion verification on multiple personnel positioning fusion methods for obtaining the final personnel positioning coordinates by fusing multiple personnel positioning coordinates, obtain multiple verification scores corresponding one-to-one to the multiple personnel positioning fusion methods, and determine the corresponding optimal personnel positioning fusion method from the multiple personnel positioning fusion methods according to the multiple verification scores.

[0086] In step S2, the aforementioned personnel positioning fusion method is an integrated positioning technology that improves the overall positioning ability by combining multiple independent positioning results. Specifically, the multiple personnel positioning fusion methods include, but are not limited to, the simple average method, the weighted average method, and / or the stacking generalization method, etc. The aforementioned personnel positioning fusion methods are all existing methods, and are now introduced one by one as follows.

[0087] (A) Simple average method, that is, simple average is the most basic fusion method, and the final positioning result is obtained by taking the average of multiple personnel positioning coordinates:

[0088]

[0089] In the formula, represents the final fused positioning result, N represents the number of personnel positioning coordinates participating in the fusion, i represents a positive integer, represents the i-th personnel positioning coordinate.

[0090] (B) Weighted average method, that is, weighted average assigns different weights to the personnel positioning coordinates obtained by different personnel positioning subsystems, adjusts the weights according to the performance of each personnel positioning subsystem on the validation set, and calculates the weighted average value:

[0091]

[0092] In the formula, r i represents the weight of the i-th personnel positioning subsystem, and satisfies Other symbols are the same as those in the simple average method.

[0093] (C) Stacking generalization method, that is, the stacking generalization method Stacking is a technology that improves the positioning performance by combining the personnel positioning coordinates obtained by different personnel positioning subsystems. Its basic idea is to use the personnel positioning coordinates obtained by multiple different personnel positioning subsystems as feature inputs into a meta-learner for learning, so as to train a new model (usually a simple linear regression model) to optimize the final positioning result:

[0094]

[0095] In the formula, f meta represents the meta-learner, respectively represent the personnel positioning coordinates obtained by different personnel positioning subsystems.

[0096] In the step S2, the specific process of the foregoing fusion verification is exemplified as follows: For various personnel positioning fusion methods in the multiple personnel positioning fusion methods, based on the test sample input items, the corresponding final personnel positioning coordinates are calculated by using the corresponding method, and the test sample output items are used to verify the final personnel positioning coordinates to obtain the corresponding verification scores. The verification scores are used to reflect the positioning accuracy of the corresponding fusion method, and specifically, but not limited to, the MAPE (Mean Absolute Percentage Error, average percentage error, which is a commonly used evaluation index for measuring the average of the percentage errors between the predicted values and the actual values to reflect the relative accuracy of the prediction) index and are conventionally statistically obtained based on the multiple test sample input items and the multiple test sample output items. Since the lower the MAPE index value, the better the corresponding method, when the verification score uses the MAPE index, according to the multiple verification scores, the corresponding best personnel positioning fusion method is determined from the multiple personnel positioning fusion methods, including but not limited to the following steps: determining the lowest score from the multiple verification scores, and then taking a certain fusion method in the multiple personnel positioning fusion methods that corresponds to the lowest score as the best personnel positioning fusion method.

[0097] S3. Receive multiple real-time personnel positioning coordinates measured by the multiple personnel positioning subsystems for the target personnel in the target area, where the multiple real-time personnel positioning coordinates correspond one-to-one to the multiple personnel positioning subsystems.

[0098] In the step S3, the target personnel is the target object for personnel positioning. The real-time personnel positioning coordinates can be conventionally and real-time obtained and uploaded by the corresponding personnel positioning subsystem.

[0099] S4. Determine the real-time location site of the target personnel in the target area according to the multiple real-time personnel positioning coordinates and the known locations of each site in the target area.

[0100] In the step S4, each of the sites may include, but is not limited to, specific outdoor open spaces, outdoor non-open spaces, indoor low-rise rooms, indoor high-rise rooms, basements, etc. Their locations may be known in advance. Thus, based on the matching results between the real-time coordinates of personnel positioning and the locations of the sites, the real-time site where the target person is located within the target area can be comprehensively determined. Specifically, as Figure 2 shown, determining the real-time site where the target person is located within the target area according to the multiple real-time coordinates of personnel positioning and the known locations of each site within the target area includes, but is not limited to, the following steps S41 to S44.

[0101] S41. Determine the weight coefficients of each personnel positioning subsystem among the multiple personnel positioning subsystems according to the multiple multi-source positioning data corresponding to the various site types and the multiple real position data.

[0102] In the step S41, specifically, determining the weight coefficients of each personnel positioning subsystem among the multiple personnel positioning subsystems according to the multiple multi-source positioning data corresponding to the various site types and the multiple real position data includes, but is not limited to, the following steps S411 to S413.

[0103] S411. Extract all the personnel positioning coordinates measured by each personnel positioning subsystem among the multiple personnel positioning subsystems for the experimental personnel and all the real position data corresponding one by one to all the personnel positioning coordinates according to the multiple multi-source positioning data corresponding to the various site types and the multiple real position data.

[0104] S412. For each personnel positioning subsystem, statistically obtain the corresponding average positioning deviation according to all the personnel positioning coordinates measured by the corresponding system for the experimental personnel and all the real position data corresponding one by one to all the personnel positioning coordinates.

[0105] In the step S412, the average positioning deviation is used to reflect the positioning accuracy of the corresponding positioning subsystem. Specifically, the MAPE index can be used. The lower the value of the MAPE index, the higher the positioning accuracy of the corresponding positioning subsystem, and vice versa.

[0106] S413. For each personnel positioning subsystem, calculate the corresponding weight coefficient according to the following formula:

[0107]

[0108] where M represents the total number of systems of the multiple personnel positioning subsystems, m represents a positive integer less than or equal to M, and η mrepresents the weight coefficient of the m-th personnel positioning subsystem among the multiple personnel positioning subsystems, represents the average positioning deviation of the m-th personnel positioning subsystem, represents the normalized value of the average positioning deviation of the m-th personnel positioning subsystem, and m' represents a positive integer less than or equal to M, represents the average positioning deviation of the m'-th personnel positioning subsystem among the multiple personnel positioning subsystems, represents the normalized value of the average positioning deviation of the m'-th personnel positioning subsystem.

[0109] In step S413, based on the above formula, it can be seen that a personnel positioning subsystem with a lower average positioning deviation can have a higher weight coefficient, and vice versa, which meets the actual requirements.

[0110] S42. According to the multiple real-time coordinates of personnel positioning and the known positions of each site in the target area, traverse each personnel positioning subsystem in the following manner: If the real-time coordinate of personnel positioning corresponding to the currently traversed personnel positioning subsystem among the multiple real-time coordinates of personnel positioning matches the known position of a certain site in the target area, then add the weight coefficient of the currently traversed personnel positioning subsystem to the voting score of the certain site, where the voting score of the certain site is initially zero before traversal.

[0111] In step S42, for example, if the weight coefficient of the first personnel positioning subsystem is 0.20, the weight coefficient of the second personnel positioning subsystem is 0.30, the weight coefficient of the third personnel positioning subsystem is 0.16, the weight coefficient of the fourth personnel positioning subsystem is 0.18, the weight coefficient of the fifth personnel positioning subsystem is 0.16, and there is a real-time coordinate of personnel positioning corresponding to the first personnel positioning subsystem that matches the known position of site A, a real-time coordinate of personnel positioning corresponding to the second personnel positioning subsystem that matches the known position of site B, a real-time coordinate of personnel positioning corresponding to the third personnel positioning subsystem that matches the known position of site B, a real-time coordinate of personnel positioning corresponding to the fourth personnel positioning subsystem that matches the known position of site B, and a real-time coordinate of personnel positioning corresponding to the fifth personnel positioning subsystem that matches the known position of site B, then the voting score of site A is 0.20, and the voting score of site B is 0.3 + 0.16 + 0.18 + 0.16 = 0.80.

[0112] S43. Determine whether the highest voting score among the voting scores of all sites in the target area exceeds a preset threshold.

[0113] In the step S43, the preset threshold can be preset according to actual requirements, for example, it is 0.75.

[0114] S44. If so, determine any venue corresponding to the highest voting score as the real-time venue where the target person is located in the target area.

[0115] In the step S44, based on the examples in steps S42 and S43, venue B can be determined as the real-time venue where the target person is located in the target area.

[0116] S5. According to the multiple real-time coordinates of personnel positioning, adopt the best personnel positioning fusion method corresponding to the venue type of the real-time venue, and fuse to obtain the final real-time coordinates of the personnel positioning of the target person.

[0117] Thus, based on the personnel positioning method described in the foregoing steps S1 to S5, a new solution for integrating multiple positioning technologies to accurately locate target personnel in complex indoor and outdoor scenarios is provided. That is, first, for various venue types, based on multiple real position data of experimental personnel in the corresponding venue and multiple multi-source positioning data measured by different multiple personnel positioning subsystems for the experimental personnel respectively, determine the corresponding best personnel positioning fusion method. Then, according to the multiple real-time coordinates of personnel positioning measured by multiple personnel positioning subsystems for the target person and the known positions of each venue, determine the real-time venue where the target person is located. Finally, according to the multiple real-time coordinates of personnel positioning, adopt the best personnel positioning fusion method corresponding to the venue type of the real-time venue, and fuse to obtain the final real-time coordinates of personnel positioning. In this way, considering the positioning accuracy differences of different positioning data in different venues, by making decisions on the venue and its positioning fusion method, the best fusion method can be selected for multi-source positioning data fusion in different venues, achieving the purpose of effectively improving the accuracy of personnel positioning. Furthermore, it can be used to provide a real-time monitoring function for personnel in emergency rescue, personnel management, public safety and other affairs, which is convenient for practical application and promotion.

[0118] Based on the technical solution of the foregoing first aspect, this embodiment further provides a possible design for optimizing the best personnel positioning fusion method. That is, when the best personnel positioning fusion method is the stacking generalization method, after determining the best personnel positioning fusion method corresponding to the site type of the real-time location from the multiple personnel positioning fusion methods and before applying this best personnel positioning fusion method, the method further includes but is not limited to the following steps: applying the multiple pieces of multi-source positioning data corresponding to the site type of the real-time location and the multiple real location data, and optimizing the hyperparameters of the meta-learner in the best personnel positioning fusion method corresponding to the site type of the real-time location based on an optimization algorithm to obtain the hyperparameters and the optimal search result for making the final coordinates of personnel positioning the most accurate. Specifically, the optimization algorithm can but is not limited to adopting a particle swarm optimization algorithm, a Newton optimization algorithm, a genetic optimization algorithm, a grey wolf optimization algorithm, a whale optimization algorithm, or a tuna swarm optimization algorithm, etc. Considering that different optimization algorithms have different advantages and disadvantages, in order to synthesize the performance of various optimization algorithms to achieve the purpose of making the best use of advantages and avoiding disadvantages, preferably, as Figure 3 shown, applying the multiple pieces of multi-source positioning data corresponding to the site type of the real-time location and the multiple real location data, and optimizing the hyperparameters of the meta-learner in the best personnel positioning fusion method corresponding to the site type of the real-time location based on an optimization algorithm to obtain the hyperparameters and the optimal search result for making the final coordinates of personnel positioning the most accurate, includes the following steps S501 to S516.

[0119] S501. Initialize the optimization algorithm parameters including the number of search individuals I in the search population, the maximum number of iterations T, the first learning factor α 1 , the second learning factor α 2 , the third learning factor α 3 , the fourth learning factor α 4 , the fifth learning factor α 5 and the sixth learning factor α 6 , and randomly generate the initial positions of each search individual in the search population, and then execute step S502. Among them, the initial positions of each search individual are randomly generated based on the following formula:

[0120]

[0121] In the formula, i′ represents a positive integer less than or equal to I, d′ represents a positive integer less than or equal to D′, D′ represents the dimension number of the parameter vector to be optimized, and the parameter vector to be optimized includes the hyperparameters of the meta-learner in the best personnel positioning fusion method corresponding to the site type of the real-time location, Denote the component of the initial position of the \(i'\)-th search individual in the search population on the \(d'\)-th dimension of the parameter vector to be optimized as \(u\). c,d′ Denote the upper limit of the parameter search space on the \(d'\)-th dimension as \(u\). c,d′ Denote the lower limit of the parameter search space on the \(d'\)-th dimension as \(l\), and \(rand(0, 1)\) represents a pure decimal random generation function.

[0122] S502. For each search individual, use the corresponding initial position as the hyperparameter of the meta-learner in the optimal personnel positioning fusion method corresponding to the site type of the real-time location. Then, apply the multiple multi-source positioning data and the multiple true position data corresponding to the site type of the real-time location to perform fusion verification on this optimal personnel positioning fusion method, obtain the initial value of the corresponding MAPE index, and use this initial value of the MAPE index as the corresponding fitness. Then, execute step S503.

[0123] In step S502, the specific technical details can be obtained by referring to the conventional derivation in the aforementioned step S2, which will not be elaborated here.

[0124] S503. Use the initial position of a search individual with the minimum fitness as the initial global optimal position \(x\). best And also initialize the current iteration number \(t' = 0\), then execute step S504.

[0125] S504. Calculate the current average position of the search population based on the current positions of each search individual, and for each search individual, determine the corresponding first new position. Then, execute step S505, where the first new position of each search individual is determined according to the following formula:

[0126] \(x\) new1,i′,d′ \(=\) best,d′ \(x\) 1 \(+\alpha\) mean,d′ \(\times rand(0, 1)\times(x\) i′,d′ \(- x\)

[0127] In the formula, \(x\) new1,i′,d′ represents the component of the first new position of the \(i'\)-th search individual on the \(d'\)-th dimension, \(x\) best,d′ represents the component of the global optimal position \(x\) best on the \(d'\)-th dimension, \(x\) mean,d′ represents the component of the current average position of the search population on the \(d'\)-th dimension, and \(x\) i′,d′ represents the component of the current position of the \(i'\)-th search individual on the \(d'\)-th dimension.

[0128] S505. For each of the search individuals, use the corresponding first new position as the hyperparameter of the meta-learner in the optimal personnel positioning fusion method corresponding to the site type of the real-time location site, and then apply the multiple multi-source positioning data corresponding to the site type of the real-time location site and the multiple real location data to perform fusion verification on this optimal personnel positioning fusion method to obtain the corresponding new value of the MAPE index, and use this new value of the MAPE index as the corresponding new fitness, and then execute step S506.

[0129] In step S505, the specific technical details can be obtained by referring to the conventional derivation in the aforementioned step S2, and will not be elaborated here.

[0130] S506. Determine whether the fitness corresponding to the current position of the i'-th search individual is greater than the current fitness of the i'-th search individual. If so, update the current position of the i'-th search individual to the first new position of the i'-th search individual, and then execute step S507; otherwise, directly execute step S507.

[0131] S507. Determine whether the fitness corresponding to the global optimal position x best is greater than the minimum value among the current fitnesses of each search individual. If so, update the global optimal position x best to the current position of any search individual with this minimum value, and then execute step S508; otherwise, directly execute step S508.

[0132] S508. Update and calculate the current average position of the search population according to the current positions of each search individual, and determine the corresponding second new position for each search individual, and then execute step S509, where the second new position of each search individual is determined according to the following formula:

[0133]

[0134] In the formula, x new2,i′,d′ represents the component of the second new position of the i'-th search individual in the d'-th dimension, β(i') represents the first intermediate variable corresponding to the i'-th search individual, δ(i') represents the second intermediate variable corresponding to the i'-th search individual, i″ represents a positive integer less than or equal to I, mod() represents the remainder function, x i″,d′The component of the current position of the i″-th search individual in the search population on the d′-th dimension is denoted as, βr(i′) represents the third intermediate variable corresponding to the i′-th search individual, δr(i′) represents the fourth intermediate variable corresponding to the i′-th search individual, max() represents the maximum value function, θ(i′) represents the fifth intermediate variable corresponding to the i′-th search individual, r(i′) represents the sixth intermediate variable corresponding to the i′-th search individual, and π represents 180 degrees.

[0135] S509. For each of the search individuals, use the corresponding second new position as the hyperparameter of the meta-learner in the optimal personnel positioning fusion method corresponding to the site type of the real-time location. Then, apply the multiple multi-source positioning data and the multiple true position data corresponding to the site type of the real-time location to perform fusion verification on this optimal personnel positioning fusion method, obtain the new value of the corresponding MAPE index, and use this new value of the MAPE index as the corresponding and new fitness. Then, execute step S510.

[0136] In step S509, the specific technical details can be obtained by referring to the conventional derivation in the foregoing step S2, and will not be elaborated here.

[0137] S510. Determine whether the fitness corresponding to the current position of the i′-th search individual is greater than the current fitness of the i′-th search individual. If so, update the current position of the i′-th search individual to the second new position of the i′-th search individual, and then execute step S511. Otherwise, directly execute step S511.

[0138] S511. Determine whether the fitness corresponding to the global optimal position x best is greater than the minimum value among the current fitnesses of each of the search individuals. If so, update the global optimal position x best to the current position of any search individual with this minimum value, and then execute step S512. Otherwise, directly execute step S512.

[0139] S512. Update and calculate the current average position of the search population according to the current positions of each of the search individuals, and determine the corresponding third new position for each of the search individuals. Then, execute step S513. Among them, the third new position of each of the search individuals is determined according to the following formula:

[0140]

[0141] In the formula, x new3,i′,d′ represents the component of the third new position of the i′-th search individual on the d′-th dimension, represents the seventh intermediate variable corresponding to the i'-th search individual, represents the eighth intermediate variable corresponding to the i'-th search individual, represents the ninth intermediate variable corresponding to the i'-th search individual, represents the tenth intermediate variable corresponding to the i'-th search individual, represents the eleventh intermediate variable corresponding to the i'-th search individual.

[0142] S513. For each of the search individuals, use the corresponding third new position as the hyperparameter of the meta-learner in the optimal personnel positioning fusion method corresponding to the site type of the real-time location site, and then apply the multiple multi-source positioning data and the multiple true position data corresponding to the site type of the real-time location site to perform fusion verification on this optimal personnel positioning fusion method, obtain the new value of the corresponding MAPE index, and use this new value of the MAPE index as the corresponding and new fitness, and then execute step S514.

[0143] In step S513, the specific technical details can be obtained by referring to the conventional derivation in the aforementioned step S2, and will not be elaborated here.

[0144] S514. Determine whether the fitness corresponding to the current position of the i'-th search individual is greater than the current fitness of the i'-th search individual. If so, update the current position of the i'-th search individual to the third new position of the i'-th search individual, and then execute step S515. Otherwise, directly execute step S515.

[0145] S515. Determine whether the fitness corresponding to the global optimal position x best is greater than the minimum value among the current fitnesses of each search individual. If so, update the global optimal position x best to the current position of any search individual with this minimum value, and then execute step S516. Otherwise, directly execute step S516.

[0146] S516. Increment the current iteration number t' by 1, and determine whether the current iteration number t' has reached the maximum iteration number T. If so, use the global optimal position x best as the hyperparameter and the optimal search result for making the final coordinates of personnel positioning the most accurate. Otherwise, return to execute step S504.

[0147] Based on the foregoing steps S501 - S516, since in each iteration, each search individual performs three fitness calculations by means of the spiral and the swarm center respectively, it can be regarded as a fusion of the particle swarm optimization algorithm and the whale optimization algorithm. Therefore, under all the same conditions, its optimization performance will be better than other optimization algorithms, which is conducive to quickly and accurately obtaining the hyperparameters and the optimal search result for making the final coordinates of personnel positioning the most accurate.

[0148] Therefore, based on the foregoing possible design one, the accuracy of personnel positioning can be further improved by optimizing the best personnel positioning fusion method, and based on the fusion of the particle swarm optimization algorithm and the whale optimization algorithm, it is conducive to quickly and accurately obtaining the optimal search result.

[0149] On the basis of the technical solutions of the foregoing first aspect or possible design one, this embodiment further provides a possible design two for how to perform positioning fusion when the real - time location of the target personnel cannot be determined, that is, after determining whether the highest voting score exceeds the preset threshold, if it is determined that the highest voting score does not exceed the preset threshold, the method further includes but is not limited to the following steps S451 - S453.

[0150] S451. Select at least one site from all the sites with non - zero voting scores.

[0151] In step S451, based on the example of step S42, if the preset threshold is 0.9, the at least one site includes site A (because its voting score is 0.2) and site B (because its voting score is 0.8).

[0152] S452. For each site in the at least one site, according to the multiple real - time coordinates of personnel positioning, adopt the best personnel positioning fusion method corresponding to the corresponding site type to fuse and obtain the final real - time coordinates of the target personnel and the corresponding personnel positioning.

[0153] S453. According to the final real - time coordinates of personnel positioning in each site among the at least one site, calculate the comprehensive final real - time coordinates P of the target personnel's personnel positioning according to the following formula xyz :

[0154]

[0155] In the formula, K represents the total number of sites of the at least one site, k represents a positive integer less than or equal to K, P k,xyz represents the final real - time coordinates of personnel positioning in the k - th site among the at least one site, γ k represents the voting score of the k - th site, k′ represents a positive integer less than or equal to K, γ k′Represents the voting score of the k'-th venue among the at least one venue.

[0156] Based on the foregoing possible design two, when it is impossible to determine the real-time location of the target person, the comprehensive final real-time coordinates of the target person's location can be obtained based on the optimal personnel location fusion method and the weighted voting score of all suspected venues where the target person is located, still effectively improving the accuracy of personnel location.

[0157] Based on the technical solution of the foregoing possible design one, this embodiment also provides a possible design three for optimizing the hyperparameters of the meta-learner based on a new heuristic algorithm, that is, applying the multiple multi-source location data corresponding to the venue type of the real-time location venue and the multiple real location data, and optimizing the hyperparameters of the meta-learner in the optimal personnel location fusion method corresponding to the venue type of the real-time location venue based on an optimization algorithm to obtain the hyperparameters and the optimal search result for making the final coordinates of personnel location the most accurate, including but not limited to the following steps S521 to S529.

[0158] S521. Initialize the parameters of the optimization algorithm including the maximum number of iterations T, and randomly generate an initial search value array of the set of parameters to be optimized, and then execute step S522, where the set of parameters to be optimized includes all the hyperparameters of the meta-learner in the optimal personnel location fusion method corresponding to the venue type of the real-time location venue, and the initial search value array y of the set of parameters to be optimized 0 Is expressed as follows:

[0159]

[0160] In the formula, d″ represents a positive integer less than or equal to D″, and D″ represents the total number of parameters in the set of parameters to be optimized. Represents the initial search value corresponding to the d″-th parameter in the set of parameters to be optimized, u c,d″ Represents the upper limit of the parameter search space corresponding to the d″-th parameter, l c,d″ Represents the lower limit of the parameter search space corresponding to the d″-th parameter, and rand(0,1) represents a pure decimal random generation function.

[0161] S522. Use the initial search value array of the set of parameters to be optimized as the hyperparameters of the meta-learner in the optimal personnel location fusion method corresponding to the venue type of the real-time location venue, and then apply the multiple multi-source location data corresponding to the venue type of the real-time location venue and the multiple real location data to perform fusion verification on this optimal personnel location fusion method to obtain the initial value of the MAPE index, and use this initial value of the MAPE index as the fitness corresponding to the initial search value array, and then execute step S523.

[0162] In the step S522, the specific technical details can be obtained by referring to the conventional derivation in the aforementioned step S2, and will not be elaborated here.

[0163] S523. Use the initial search value array of the set of parameters to be optimized as the current optimal search value array, initialize the current iteration count t′ = 0, and also initialize the forbidden change countdown value of each parameter in the set of parameters to be optimized to zero, and then execute step S524.

[0164] S524. Based on the current optimal search value array, perform independent increase processing and decrease processing on each parameter in the set of parameters to be optimized and with a current forbidden change countdown value of zero, to obtain 2×D″′ new search value arrays of the set of parameters to be optimized, and then execute step S525, where D″′ represents the total number of parameters in the set of parameters to be optimized and with a current forbidden change countdown value of zero, and 1≤D″′≤D″.

[0165] In the step S524, the increase processing or the decrease processing needs to be carried out within the corresponding parameter search space, and can be a quantitative step - by - step increase / decrease, or an indefinite random increase / decrease. Also, it can first perform quantitative step - by - step increase / decrease on each parameter in the set of parameters to be optimized and with a current forbidden change countdown value of zero, and then if it is found that the update of the current optimal search value array is not completed (i.e., step S529 is directly executed in the subsequent step S528), then perform indefinite random increase / decrease on each parameter in the set of parameters to be optimized and with a current forbidden change countdown value of zero, so as to avoid falling into a local optimal solution. For example, if there are the following five parameters in the set of parameters to be optimized: parameter A, parameter B, parameter C, parameter D, and parameter E, where the current forbidden change countdown values of parameter A, parameter B, and parameter E are zero respectively (i.e., D″′ is 3), then based on the current optimal search value array, independent increase processing and decrease processing can be performed on parameter A respectively to obtain two different new search value arrays of the set of parameters to be optimized (one obtained based on the increase processing and the other based on the decrease processing); based on the current optimal search value array, independent increase processing and decrease processing can be performed on parameter B respectively to obtain two different new search value arrays of the set of parameters to be optimized; based on the current optimal search value array, independent increase processing and decrease processing can be performed on parameter E respectively to obtain two different new search value arrays of the set of parameters to be optimized; thus, 2×3 = 6 new search value arrays can be obtained, and each array represents a neighborhood search direction in the search space.

[0166] S525. For each of the 2×D″′ new search value arrays, use the corresponding array as the hyperparameter of the meta-learner in the optimal personnel positioning fusion method corresponding to the site type of the real-time location. Then, apply the multiple multi-source positioning data and the multiple true location data corresponding to the site type of the real-time location to perform fusion verification on this optimal personnel positioning fusion method, obtain a new value of the MAPE metric, and use this new MAPE metric value as the fitness of the corresponding array. Then, execute step S526.

[0167] In step S525, the specific technical details can be obtained by referring to the conventional derivation in the aforementioned step S2, and will not be elaborated here.

[0168] S526. For each of the arrays, subtract the fitness of the corresponding array from the fitness corresponding to the current optimal search value array to obtain the fitness difference value of the corresponding array. When the fitness difference value is greater than zero and is the largest fitness difference value in this time, update the freeze countdown value of the corresponding unique variable parameter to a positive integer value that is positively correlated with this fitness difference value. Then, execute step S527, where the unique variable parameter refers to the parameter in the set of parameters to be optimized and is used to obtain the corresponding array through addition processing or subtraction processing.

[0169] In step S526, the fact that the fitness difference value is greater than zero and is the largest fitness difference value in this time means that the best search value array has been obtained in the neighborhood search direction corresponding to the corresponding parameter in this time. Therefore, it is necessary to update the freeze countdown value of this corresponding parameter to a positive integer that is positively correlated with this fitness difference value to temporarily lock the search result in this neighborhood search direction. In addition, continuing with the example in step S524 above, among the six new search value arrays corresponding to parameter A, parameter B, and parameter E, if the fitness difference value of a certain new search value array corresponding to parameter B is greater than zero and is the largest fitness difference value in this time (i.e., the largest among the six fitness difference values in this time), then update the freeze countdown value of parameter B from zero to a positive integer value that is positively correlated with this fitness difference value, while the freeze countdown values of parameter A and parameter E remain zero.

[0170] S527. Determine whether there is any parameter in the set of parameters to be optimized whose current freeze countdown value is zero. If not, decrement the current freeze countdown value of each parameter in the set of parameters to be optimized by 1, and then return to execute step S527. Otherwise, execute step S528.

[0171] In step S527, since the current freezing countdown values of all parameters in the set of parameters to be optimized are not zero, it means that it is impossible to return to step S524 for subsequent execution. Therefore, they need to be decremented together until the current freezing countdown value of at least one parameter is zero.

[0172] S528. Determine whether there is any fitness difference value greater than zero among the fitness difference values of each array. If so, update the current optimal search value array to the array corresponding to the minimum fitness among the 2×D″′ new search value arrays, and then execute step S529; otherwise, directly execute step S529.

[0173] S529. Increment the current iteration count t′ by 1, and determine whether the current iteration count t′ has reached the maximum iteration count T. If so, use the current optimal search value array as the hyperparameter and the optimal search result for making the final coordinates of personnel positioning the most accurate; otherwise, return to execute step S524.

[0174] Therefore, based on the foregoing possible design three, by using the steps of the foregoing new heuristic algorithm to optimize the hyperparameters of the meta-learner, it is possible to temporarily lock the search results in different numbers of iterations based on different fitness difference values during the optimization process for the search results in each best neighborhood search direction, thereby facilitating the rapid search for the optimal hyperparameters of the meta-learner. Moreover, by using a parameter adjustment method of first quantitatively increasing / decreasing the parameters to be adjusted and then randomly increasing / decreasing them in an indefinite amount, it is also possible to avoid falling into local optimal solutions, further facilitating the rapid and accurate obtaining of the optimal search result.

[0175] As Figure 4 shown, in the second aspect of this embodiment, an entity system for implementing the personnel positioning method described in the first aspect or any possible design in the first aspect is provided, including a plurality of personnel positioning subsystems and a computer device respectively communicatively connected to the plurality of personnel positioning subsystems. Among them, the computer device is used to execute the personnel positioning method described in the first aspect or any possible design in the first aspect.

[0176] For the working process, working details, and technical effects of the foregoing system provided in the second aspect of this embodiment, reference can be made to the personnel positioning method described in the first aspect or any possible design in the first aspect, which will not be elaborated herein.

[0177] As Figure 5As shown, in the third aspect of this embodiment, a computer device for executing the personnel positioning method described in the first aspect or any possible design in the first aspect is provided, including a memory, a processor, and a transceiver that are communicatively connected in sequence. Among them, the memory is used to store computer programs, the transceiver is used to send and receive messages, and the processor is used to read the computer programs and execute the personnel positioning method described in the first aspect or any possible design in the first aspect. Specifically, for example, the memory may include, but is not limited to, a random access memory (RAM), a read-only memory (ROM), a flash memory, a first-in first-out memory (FIFO), and / or a first-in last-out memory (FILO), etc.; the processor may include, but is not limited to, a microprocessor of the STM32F105 series. In addition, the computer device may also include, but is not limited to, a power module, a display screen, and other necessary components.

[0178] For the working process, working details, and technical effects of the aforementioned computer device provided in the third aspect of this embodiment, reference may be made to the personnel positioning method described in the first aspect or any possible design in the first aspect, which will not be elaborated here.

[0179] In the fourth aspect of this embodiment, a computer-readable storage medium storing instructions including the personnel positioning method described in the first aspect or any possible design in the first aspect is provided, that is, instructions are stored on the computer-readable storage medium. When the instructions run on a computer, the personnel positioning method described in the first aspect or any possible design in the first aspect is executed. Among them, the computer-readable storage medium refers to a carrier for storing data, and may include, but is not limited to, computer-readable storage media such as floppy disks, optical discs, hard disks, flash memories, USB flash drives, and / or memory sticks. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

[0180] For the working process, working details, and technical effects of the aforementioned computer-readable storage medium provided in the fourth aspect of this embodiment, reference may be made to the personnel positioning method described in the first aspect or any possible design in the first aspect, which will not be elaborated here.

[0181] In the fifth aspect of this embodiment, a computer program product is provided, including a computer program or instructions. When the computer program or the instructions are executed by a computer, the personnel positioning method described in the first aspect or any possible design in the first aspect is implemented. Among them, the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

[0182] Finally, it should be noted that the above are only the preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A personnel positioning method based on multi-source data, characterized in that: include: For various types of venues in the target area, multiple real position data of the experimenters in the corresponding venues and multiple multi-source positioning data respectively measured by multiple different personnel positioning subsystems for the experimenters and corresponding to the multiple real position data are obtained, wherein the multi-source positioning data contains multiple personnel positioning coordinates corresponding to the multiple personnel positioning subsystems; For each type of venue, the corresponding multiple copies of multi-source positioning data are used as multiple test sample input items, and the corresponding multiple real location data are used as multiple test sample output items corresponding one-to-one to the multiple test sample input items, and then the multiple test sample input items and the multiple test sample output items are applied to perform fusion verification on multiple personnel positioning fusion methods for obtaining final personnel positioning coordinates by fusing multiple personnel positioning coordinates, and multiple verification scores corresponding one-to-one to the multiple personnel positioning fusion methods are obtained, and according to the multiple verification scores, a corresponding optimal personnel positioning fusion method is determined from the multiple personnel positioning fusion methods; Receiving a plurality of real-time coordinates of personnel positioning obtained by measuring the target personnel in the target area by the plurality of personnel positioning subsystems, wherein the plurality of real-time coordinates of personnel positioning correspond one-to-one to the plurality of personnel positioning subsystems; Determine the real-time location of the target person in the target area according to the real-time coordinates of the multiple personnel positions and the known locations of various venues in the target area; According to the multiple real-time coordinates of personnel positioning, the best personnel positioning fusion method corresponding to the site type of the real-time site is adopted to fuse and obtain the final real-time coordinates of the personnel positioning of the target person.

2. The personnel positioning method according to claim 1, characterized in that: The multiple personnel positioning subsystems include a first personnel positioning subsystem based on satellite positioning technology, a second personnel positioning subsystem based on wireless positioning technology, a third personnel positioning subsystem based on video recognition technology, a fourth personnel positioning subsystem based on millimeter wave radar technology and / or a fifth personnel positioning subsystem based on ultrasonic positioning technology, wherein the first personnel positioning subsystem includes a satellite locator for binding personnel, the second personnel positioning subsystem includes at least three fixed and known positioning base stations and positioning tags for binding personnel, the third personnel positioning subsystem includes multiple AI cameras arranged at different positions in the target area, the fourth personnel positioning subsystem includes multiple millimeter wave radars arranged at different positions in the target area, and the fifth personnel positioning subsystem includes at least three fixed and known ultrasonic reflectors and an ultrasonic main rangefinder for binding personnel.

3. The personnel positioning method according to claim 1, characterized in that: The multiple personnel positioning fusion methods include a simple average method, a weighted average method and / or a stacked generalization method.

4. The personnel positioning method according to claim 1, characterized in that: When the optimal personnel positioning fusion method is a stacked generalization method, after determining the optimal personnel positioning fusion method corresponding to the venue type of the real-time venue from the multiple personnel positioning fusion methods and before applying the optimal personnel positioning fusion method, the method further includes: The multiple multi-source positioning data corresponding to the venue type of the real-time venue and the multiple real location data are applied to optimize the hyperparameters of the meta-learner in the optimal personnel positioning fusion method corresponding to the venue type of the real-time venue based on the optimization algorithm, and the hyperparameters are obtained and used to obtain the most accurate optimization search result for the final coordinates of the personnel positioning.

5. The personnel positioning method according to claim 4, characterized in that: The optimization algorithm adopts a particle swarm optimization algorithm, a Newton optimization algorithm, a genetic optimization algorithm, a gray wolf optimization algorithm, a whale optimization algorithm or a tuna swarm optimization algorithm.

6. The personnel positioning method according to claim 4, characterized in that: Applying the multiple copies of multi-source positioning data corresponding to the venue type of the real venue and the multiple real location data, optimizing the hyperparameters of the meta-learner in the optimal personnel positioning fusion method corresponding to the venue type of the real venue based on the optimization algorithm, obtaining the hyperparameters and the most accurate optimization search result for the final coordinates of the personnel positioning, including the following steps S501 to S516: S501. Initialize the optimization algorithm parameters including the number of search individuals I in the search population, the maximum number of iterations T, the first learning factor α1, the second learning factor α2, the third learning factor α3, the fourth learning factor α4, the fifth learning factor α5 and the sixth learning factor α6, and randomly generate the initial position of each search individual in the search population, and then execute step S502, wherein the initial position of each search individual is randomly generated based on the following formula: In the formula, i′ represents a positive integer less than or equal to I, d′ represents a positive integer less than or equal to D′, D′ represents the number of dimensions of the parameter vector to be optimized, and the parameter vector to be optimized includes the hyperparameters of the meta-learner in the optimal personnel positioning fusion method corresponding to the venue type of the real-time venue. represents the component of the initial position of the i′th search individual in the search population on the d′th dimension of the parameter vector to be optimized, u c,d′ represents the upper limit of the parameter search space in the d′th dimension, l c,d′ represents the lower limit of the parameter search space in the d′th dimension, and rand(0,1) represents a pure decimal random generator function; S502. For each search individual, the corresponding initial position is used as a hyperparameter of a meta-learner in an optimal personnel positioning fusion method corresponding to the venue type of the real-time venue, and then the multiple multi-source positioning data corresponding to the venue type of the real-time venue and the multiple real position data are applied to perform fusion verification on the optimal personnel positioning fusion method to obtain a corresponding initial value of the MAPE indicator, and the initial value of the MAPE indicator is used as the corresponding fitness, and then step S503 is executed; S503. Take the initial position of a search individual with the minimum fitness as the initial global optimal position x best , and also initialize and set the current iteration number t′=0, and then execute step S504; S504. Calculate the current average position of the search population according to the current position of each search individual, and determine the corresponding first new position for each search individual, and then execute step S505, wherein the first new position of each search individual is determined according to the following formula: x new1,i′,d′ =x best,d′ +α1×rand(0,1)×(x mean,d′ -x i′,d′ ) In the formula, x new1,i′,d′ represents the component of the first new position of the i′th search individual in the d′th dimension, x best,d′ Denotes the global optimal position x best The component in the d′th dimension, x mean,d′ represents the component of the current average position of the search population in the d′th dimension, x i′,d′ represents the component of the current position of the i′th search individual in the d′th dimension; S505. For each of the search individuals, the corresponding first new position is used as a hyperparameter of a meta-learner in an optimal personnel positioning fusion method corresponding to the venue type of the real-time venue, and then the multiple multi-source positioning data corresponding to the venue type of the real-time venue and the multiple real position data are applied to perform fusion verification on the optimal personnel positioning fusion method to obtain a corresponding new value of the MAPE indicator, and the new value of the MAPE indicator is used as the corresponding and new fitness, and then step S506 is executed; S506. Determine whether the fitness corresponding to the current position of the i′th search individual is greater than the current fitness of the i′th search individual. If so, update the current position of the i′th search individual to the first new position of the i′th search individual, and then execute step S507. Otherwise, directly execute step S507. S507. Determine the global optimal position x best Is the corresponding fitness greater than the minimum value among the current fitness of each search individual? If so, the global optimal position x best Update to the current position of any search individual with the minimum value, and then execute step S508, otherwise directly execute step S508; S508. The current average position of the search population is calculated and updated according to the current position of each search individual, and the corresponding second new position is determined for each search individual, and then step S509 is executed, wherein the second new position of each search individual is determined according to the following formula: In the formula, x new2,i′,d′ represents the component of the second new position of the i′th search individual in the d′th dimension, β(i′) represents the first intermediate variable corresponding to the i′th search individual, δ(i′) represents the second intermediate variable corresponding to the i′th search individual, i″ represents a positive integer less than or equal to I, mod() represents a remainder function, x i″,d′ represents the component of the current position of the i′-th search individual in the search population on the d′-th dimension, βr(i′) represents the third intermediate variable corresponding to the i′-th search individual, δr(i′) represents the fourth intermediate variable corresponding to the i′-th search individual, max() represents the maximum value function, θ(i′) represents the fifth intermediate variable corresponding to the i′-th search individual, r(i′) represents the sixth intermediate variable corresponding to the i′-th search individual, and π represents 180 degrees; S509. For each of the search individuals, the corresponding second new position is used as a hyperparameter of a meta-learner in an optimal personnel positioning fusion method corresponding to the venue type of the real-time venue, and then the multiple multi-source positioning data corresponding to the venue type of the real-time venue and the multiple real position data are applied to perform fusion verification on the optimal personnel positioning fusion method to obtain a corresponding new value of the MAPE indicator, and the new value of the MAPE indicator is used as the corresponding and new fitness, and then step S510 is executed; S510. Determine whether the fitness corresponding to the current position of the i′th search individual is greater than the current fitness of the i′th search individual. If so, update the current position of the i′th search individual to the second new position of the i′th search individual, and then execute step S511. Otherwise, directly execute step S511. S511. Determine the global optimal position x best Is the corresponding fitness greater than the minimum value among the current fitness of each search individual? If so, the global optimal position x best Update to the current position of any search individual with the minimum value, and then execute step S512, otherwise directly execute step S512; S512. The current average position of the search population is calculated based on the current position of each search individual, and the corresponding third new position is determined for each search individual, and then step S513 is executed, wherein the third new position of each search individual is determined according to the following formula: In the formula, x new3,i′,d′ represents the component of the third new position of the i′th search individual in the d′th dimension, represents the seventh intermediate variable corresponding to the i′th search individual, represents the eighth intermediate variable corresponding to the i′th search individual, represents the ninth intermediate variable corresponding to the i′th search individual, represents the tenth intermediate variable corresponding to the i′th search individual, represents the eleventh intermediate variable corresponding to the i′th search individual; S513. For each of the search individuals, the corresponding third new position is used as a hyperparameter of a meta-learner in an optimal personnel positioning fusion method corresponding to the venue type of the real-time venue, and then the multiple multi-source positioning data corresponding to the venue type of the real-time venue and the multiple real position data are applied to perform fusion verification on the optimal personnel positioning fusion method to obtain a corresponding new value of the MAPE indicator, and the new value of the MAPE indicator is used as the corresponding and new fitness, and then step S514 is executed; S514. Determine whether the fitness corresponding to the current position of the i′th search individual is greater than the current fitness of the i′th search individual. If so, update the current position of the i′th search individual to the third new position of the i′th search individual, and then execute step S515. Otherwise, directly execute step S515. S515. Determine the global optimal position x best Is the corresponding fitness greater than the minimum value among the current fitness of each search individual? If so, the global optimal position x best Update to the current position of any search individual with the minimum value, and then execute step S516, otherwise directly execute step S516; S516. Increment the current number of iterations t′ by 1, and determine whether the current number of iterations t′ reaches the maximum number of iterations T. If so, set the global optimal position x best The optimized search result is used as the hyperparameter to make the personnel locate the final coordinates most accurately, otherwise, the process returns to step S504.

7. The personnel positioning method according to claim 1, characterized in that: Determining the real-time location of the target person in the target area according to the real-time coordinates of the multiple personnel positions and the known locations of various locations in the target area includes: Determining a weight coefficient of each personnel positioning subsystem in the plurality of personnel positioning subsystems according to the plurality of multi-source positioning data corresponding to the various venue types and the plurality of real position data; According to the multiple personnel positioning real-time coordinates and the known positions of the various venues in the target area, the various personnel positioning subsystems are traversed in sequence in the following manner: if the personnel positioning real-time coordinates among the multiple personnel positioning real-time coordinates and corresponding to the currently traversed personnel positioning subsystem match the known position of a certain venue in the target area, the voting score of the certain venue is added with the weight coefficient of the currently traversed personnel positioning subsystem, wherein the voting score of the certain venue is initially zero before traversal; Determining whether the highest voting score among the voting scores of all venues in the target area exceeds a preset threshold; If so, any venue corresponding to the highest voting score is determined as the real-time location of the target person in the target area.

8. The personnel positioning method according to claim 7, characterized in that: Determining a weight coefficient of each personnel positioning subsystem in the plurality of personnel positioning subsystems according to the plurality of multi-source positioning data corresponding to the various venue types and the plurality of real position data, comprises: Extracting, according to the multiple multi-source positioning data corresponding to the various site types and the multiple real position data, all personnel positioning coordinates measured by each personnel positioning subsystem of the multiple personnel positioning subsystems on the experimental personnel and all real position data corresponding to all the personnel positioning coordinates one by one; For each of the personnel positioning subsystems, according to all personnel positioning coordinates measured by the corresponding system on the experimental personnel and all real position data corresponding to all the personnel positioning coordinates, the corresponding average positioning deviation is statistically obtained; For each of the personnel positioning subsystems, the corresponding weight coefficient is calculated according to the following formula: Wherein, M represents the total number of the plurality of personnel positioning subsystems, m represents a positive integer less than or equal to M, and η m represents the weight coefficient of the mth personnel positioning subsystem among the multiple personnel positioning subsystems, represents the average positioning deviation of the mth personnel positioning subsystem, represents the normalized value of the average positioning deviation of the m-th personnel positioning subsystem, m′ represents a positive integer less than or equal to M, represents the average positioning deviation of the m′th personnel positioning subsystem in the multiple personnel positioning subsystems, Represents the normalized value of the average positioning deviation of the m′th personnel positioning subsystem.

9. The personnel positioning method according to claim 7, characterized in that: After determining whether the highest voting score exceeds the preset threshold, if it is determined that the highest voting score does not exceed the preset threshold, the method further includes: Selecting at least one venue with a non-zero voting score from all the venues; For each of the at least one venue, according to the multiple personnel positioning real-time coordinates, an optimal personnel positioning fusion method corresponding to the corresponding venue type is adopted to fuse the target personnel and the corresponding personnel positioning final real-time coordinates; According to the final real-time coordinates of the personnel positioning in each of the at least one venue, the comprehensive final real-time coordinates of the personnel positioning of the target person P are calculated according to the following formula: xyz : Wherein, K represents the total number of the at least one venue, k represents a positive integer less than or equal to K, P k,xyz represents the final real-time coordinates of the personnel in the kth site in the at least one site, γ k represents the voting score of the kth site, k′ represents a positive integer less than or equal to K, γ k′ represents the voting score of the k′th venue in the at least one venue.

10. A personnel positioning system based on multi-source data, characterized in that: The invention comprises a plurality of personnel positioning subsystems and a computer device respectively communicatively connected to the plurality of personnel positioning subsystems, wherein the computer device is used to execute the personnel positioning method as claimed in any one of claims 1 to 9.

Images