A virtual-real fusion agent positioning simulation system in a complex real environment

Abstract

This invention relates to the field of intelligent agent positioning technology and discloses a virtual-real fusion intelligent agent positioning simulation system in complex real-world environments. The system includes: a simulation platform module for initiating the operation and status checks of each module, and for receiving and transmitting data from each module; a display management and demonstration platform for configuring simulation parameters and visualizing and demonstrating the simulation process and results; a physical equipment access module for parsing and transmitting physical equipment data; a data interaction module for receiving and processing multi-source real-world data; a real-world data processing module for establishing a mapping relationship between effective scene data and the simulation model; a high-fidelity positioning simulation module for receiving various data and executing the simulation model; a GIS map module for processing GIS map range parameters; and a real-time weather access module for processing real-time weather data. This invention achieves the construction of a high-fidelity simulation scene and improves the positioning accuracy in this scene.

Description

Technical Field

[0001] This invention relates to the field of intelligent agent positioning technology, specifically to a virtual-real fusion intelligent agent positioning simulation system in complex real-world environments. Background Technology

[0002] With the rapid development of artificial intelligence, unmanned systems, and cooperative control technologies, multi-agent systems (MAS) have been widely applied in several key technology fields, including aerospace, unmanned swarms, and intelligent navigation, due to their distributed collaborative advantages. MAS can accomplish complex tasks such as localization, environmental perception, and task execution that are difficult for a single agent to achieve through the collaborative cooperation of multiple agents. Among these, multi-agent cooperative localization technology, as one of the core supporting technologies for the stable and reliable operation of the entire MAS, directly determines the task execution effect of the MAS.

[0003] In real-world applications, multi-agent systems often operate in complex environments. These environments typically present various challenges, such as communication delays, signal interference, dynamic topology changes, and environmental noise. These factors severely impact the accuracy, robustness, and real-time performance of multi-agent cooperative localization, ultimately preventing the system from completing its intended tasks. Therefore, conducting comprehensive and accurate simulation verification of the cooperative localization method and overall system performance before actual deployment is crucial for ensuring the reliability of the multi-agent system in real-world operation. It is also an important means of reducing system deployment costs and mitigating various risks during actual testing.

[0004] Currently, simulation verification technologies for multi-agent localization in this field mainly fall into two categories: one is simulation methods developed based on general simulation platforms, and the other is dedicated simulation systems designed specifically for multi-agent localization scenarios. However, both types of simulation methods and systems have significant technical shortcomings in practical applications, as follows: 1. The modeling of complex environments suffers from poor adaptability to real-world scenarios, resulting in insufficient realism and comprehensiveness in the simulation environment, making it difficult to accurately reproduce the dynamic characteristics of complex real-world scenarios. Existing simulation systems often employ simple simulations with fixed parameters, only allowing for the setting of single, fixed interference factors such as noise and signal attenuation. They fail to simulate dynamic interference in complex environments, such as sudden signal interference, dynamic topology switching, and real-time fluctuations in environmental parameters. Furthermore, existing simulation systems do not consider the coupling effects of multiple interference factors in complex outdoor terrain and real-time weather scenarios, leading to significant deviations between the constructed simulation environment and the actual application environment. This results in low reference value for simulation results and fails to provide reliable support for the practical deployment of multi-agent positioning systems. The core reason for this is the simplistic design of the environment modeling module in existing simulation systems. They lack a mechanism for real-time interaction with highly realistic complex environment parameters, cannot flexibly adjust environmental interference types and corresponding parameters according to different real-world application environments, and do not introduce dynamic interference generation algorithms, making it difficult to accurately reproduce the dynamic evolution process of complex real-world environments.

[0005] 2. The simulation verification lacks specificity and comprehensiveness, failing to meet the verification requirements of various multi-agent localization algorithms. Existing dedicated simulation systems are mostly designed for specific types of localization methods (such as least squares localization and azimuth localization), only capable of verifying the performance of a single algorithm. They cannot simultaneously adapt to multiple types of multi-agent localization methods such as cooperative localization, visual localization, and relative localization. Furthermore, the verification metrics of existing simulation systems are relatively singular, mostly focusing only on the basic indicator of localization accuracy, without comprehensively verifying key performance indicators such as robustness, real-time performance, and cooperative consistency in the multi-agent localization process. This results in insufficient simulation verification, failing to fully reflect the overall performance of the multi-agent localization system in complex environments. The fundamental reason lies in the fixed method interface design of existing simulation systems, lacking standardized method access modules, making it impossible to flexibly integrate different types of localization methods. Moreover, the verification metric system is incomplete, failing to incorporate the core requirements of multi-agent localization in complex environments to construct a comprehensive and systematic performance evaluation system.

[0006] 3. Low simulation efficiency: Large-scale multi-agent simulation suffers from significant real-time bottlenecks. Existing simulation systems, whether based on customized simulation models built on general platforms or dedicated systems, all experience exponential growth in computational complexity when performing large-scale multi-agent simulations (e.g., simulations involving dozens or more agents). This leads to slow simulation speed, poor real-time performance, and even simulation crashes. Furthermore, existing simulation systems lack rapid analysis and visualization capabilities for simulation results. After the simulation, technicians must manually process large amounts of simulation data, which is not only time-consuming and labor-intensive but also prone to data processing errors, further reducing the overall efficiency of simulation verification.

[0007] 4. Poor reliability and traceability of simulation results. Existing simulation systems lack standardized process control. The setting of simulation parameters and the construction of simulation scenarios lack unified specifications, leading to significant differences in simulation models built by different technicians. This results in poor repeatability and insufficient reliability of simulation results under the same conditions. Furthermore, existing simulation systems do not fully record all parameters, data, and intermediate results during the simulation process. When anomalies occur, technicians cannot trace the specific cause of the anomaly, hindering the optimization and iteration of localization methods and the improvement of the simulation model. The core reason is that existing simulation systems lack standardized process management modules and complete data logging modules, failing to establish the correlation between simulation parameters, simulation processes, and simulation results, thus failing to achieve traceability of the simulation process and reproducibility of simulation results.

[0008] In summary, existing multi-agent localization simulation verification technologies cannot meet the actual needs of multi-agent localization systems in complex environments for comprehensive, high-fidelity, accurate, and efficient simulation verification. They suffer from numerous technical defects, such as unrealistic simulation of complex environments, incomplete simulation verification, low simulation efficiency, and poor reliability and traceability of simulation results, which severely limit the practical deployment and application of multi-agent systems in complex scenarios. Summary of the Invention

[0009] To address the aforementioned shortcomings in existing technologies, this invention provides a virtual-real fusion intelligent agent positioning simulation system for complex real-world environments. This system solves the problems of unrealistic simulation of complex environments, incomplete simulation verification, low simulation efficiency, and poor reliability and traceability of simulation results in existing methods or systems, thereby improving the positioning accuracy of the simulation system.

[0010] To achieve the above-mentioned objectives, the technical solution adopted by this invention is as follows: A virtual-real fusion intelligent agent localization simulation system for complex real-world environments includes: The simulation platform module is used to start the operation of each module and check whether the operation status of each module is abnormal. After that, it receives the simulation parameters transmitted by the display management demonstration platform and transmits them to the high-fidelity positioning simulation module. At the same time, it receives the simulation process and result data and transmits them to the display management demonstration platform. The display management demonstration platform is used to receive simulation parameters input or modified by users, as well as simulation process and result data, and to perform visualization display, simulation report generation, and simulation simulation operations. The physical equipment access module is used to receive and parse the data format of the physical equipment, match the interface type protocol, convert the protocol message into a format that the system can recognize or perform protocol conversion configuration to support various physical equipment, and finally transmit the physical equipment data to the data interaction module. The data interaction module is used to receive data from the actual equipment, GIS map data, and real-time meteorological data, perform caching and verification processing, generate verified multi-source real data, and transmit it to the real data processing module. The real data processing module is used to receive real data from multiple sources, perform effective scene data filtering and noise reduction, and establish a mapping relationship between the effective scene data and each simulation model to realize the conversion of effective scene data into a unified format that can be recognized by the simulation model. At the same time, the mapping relationship data is transmitted to the high-fidelity positioning simulation module. The high-fidelity positioning simulation module is used to receive simulation parameters and mapping relationship data, execute the simulation process after initializing the simulation model, monitor whether the running status of the simulation process is abnormal, and output abnormal data if so, otherwise output simulation process and result data. The GIS map module is used to receive simulation parameters, parse and convert the GIS map range parameters within the simulation parameters, generate GIS map data that the system can recognize, construct a 3D map model with terrain features, and transmit the GIS map data to the data interaction module. The real-time weather access module is used to receive real-time weather data, interpret and filter it, convert its format, generate real-time weather data that the system can recognize, and transmit it to the data interaction module.

[0011] The present invention has the following beneficial effects: The present invention proposes a virtual-real fusion intelligent agent positioning simulation system for complex real-world environments. Through the collaborative action of multiple modules, it effectively solves the problems of existing systems, such as environmental simulation distortion, inability to connect to real-world applications, lack of GIS and meteorological support, disconnect between simulation and demonstration, and poor adaptability and scalability. It can also construct highly realistic 3D fusion simulation scenes, achieving integrated simulation deduction, visualization, and report generation, improving simulation real-time performance, result traceability, and system maintainability, and enhancing the positioning accuracy of the simulation system. Furthermore, it can dynamically reproduce multi-source coupling interference and topology evolution conditions, and comprehensively and objectively evaluate the positioning and collaborative performance of multi-agent systems in complex environments, providing reliable data support for positioning method optimization and real-world deployment. Attached Figure Description

[0012] Figure 1 This is a schematic diagram of the structure of a virtual-real fusion intelligent agent positioning simulation system in a complex real environment proposed in this invention; Figure 2 This is a schematic diagram illustrating the connection principle between each module and the standardized bus in the embodiment; Detailed Implementation The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0013] First, to address the technical shortcomings of existing simulation systems, such as inability to directly connect to real-world equipment, unrealistic simulation of complex environments, lack of GIS maps and real-time meteorological data support, and a disconnect between simulation and demonstration, this invention proposes a virtual-real fusion intelligent agent positioning simulation system for complex real-world environments. The system adopts a modular structure as follows: Figure 1 As shown, the system mainly consists of a simulation platform module 1, a display and management demonstration platform 2, a physical equipment access module 3, a data interaction module 4, a real data processing module 5, a high-fidelity positioning simulation module 6, a GIS map module 7, and a real-time weather access module 8. These modules are connected via a standardized bus 9, enabling real-time data transmission and collaborative operation. Based on real data from the physical equipment, GIS map geographic information, and real-time weather data, a high-fidelity simulation scenario is constructed. The display and management demonstration platform visualizes the simulation process and demonstrates the results, ensuring that the simulation results accurately reflect the actual positioning performance of the physical equipment. This allows the simulation model and physical equipment to replicate the operational effects in a real-world environment within the laboratory. Furthermore, the system supports hybrid deployment simulation of two types of equipment: simulation of physical equipment in a laboratory environment and simulation of virtual equipment. The simulation model constructed in this invention is a simulator that supports multiple types of physical equipment. Multiple physical equipment can be virtualized through the simulation model, allowing for parameter settings, sensor data interaction, and positioning algorithm management for the virtual equipment. The actual device is an intelligent device with positioning function. Through this system, the actual device can access real weather, terrain, and other environmental information of a designated location area in a laboratory environment, thereby simulating the function, performance, and robustness of the actual device.

[0014] Meanwhile, each module is connected to the standardized bus via an independent bus interface, such as Figure 2As shown in the figure, the bus interface adopts a unified specification (RJ45 interface / wireless interface, and the wireless interface is preferably installed in the actual device). The interface is marked with the corresponding module number and name. The data transmission direction between the bus and each module is marked with a two-way arrow, clearly reflecting the two-way data transmission characteristics. At the same time, key parameters such as the bus communication protocol and transmission rate are marked. Therefore, the standardized bus is the core link for data transmission of each module of the system, connecting all core modules, realizing two-way data interaction between each module, and ensuring the stability, real-time and compatibility of data transmission. This bus uses the industrial Ethernet bus as the basis, with a transmission rate supporting 10000M / 1000Mbps. The bus interface uniformly adopts the RJ45 interface, and at the same time supports wireless networks and RS485 buses, facilitating the expansion and maintenance of modules.

[0015] Therefore, based on the above solutions to technical problems and principles, the specific embodiments adopted by the present invention are as follows: As Figures 1-2 shown, a virtual-real fusion intelligent agent positioning simulation system in a complex real environment includes: A deduction simulation platform module, which is used to start the operation of each module and check whether the operation status of each module is abnormal, receive the simulation parameters transmitted by the display management and demonstration platform and transmit them to the high-fidelity positioning simulation module, and at the same time receive the simulation process and result data and transmit them to the display management and demonstration platform.

[0016] This module is the control unit of the system simulation deduction, responsible for simulating scenario planning and scheduling to coordinate the work of each module. It controls the start, operation, pause and termination of the simulation model in the high-fidelity positioning simulation module during the simulation process by issuing instructions. At the same time, it receives the real-time simulation process data transmitted by the high-fidelity positioning simulation module, and synchronously transmits the simulation process and result data to the display management and demonstration platform to achieve the linkage between simulation and demonstration. Among them, the simulation scenario planning is to plan information such as the number of devices (virtual devices or actual devices) for the high-fidelity positioning simulation module to execute the simulation and the execution routes of each device in the deduction simulation platform module, and start the simulation operation of the high-fidelity positioning simulation module by issuing instructions. During this process, the display management and demonstration platform will synchronously display information such as the planned devices (virtual devices or actual devices), the operation status of the devices, and the dynamic modification of the operation parameters of the devices.

[0017] Based on this, the specific working process of the deduction simulation platform module is as follows: (1) Initialization: After the system starts, the simulation platform module sends an initialization command through the standardized bus to start each module in sequence (display management demonstration platform, actual equipment access module, data interaction module, real data processing module, high-fidelity positioning simulation module, GIS map module), controls the creation of virtual devices and access of actual devices in the high-fidelity positioning simulation module, and performs simulation scenario planning for each module in the simulation system (including the virtual devices, actual devices, running paths, speeds, times and sensor parameters used in this simulation), and detects the working status of each module. If an abnormality is detected in a module, an abnormality prompt message is sent to the display management demonstration platform to remind the system operator to check the communication connection and loading status of each module until all modules start normally.

[0018] (2) Simulation Control: Receives simulation parameters (including simulation duration, number of agents, positioning method type, etc.) transmitted from the display management demonstration platform, and transmits these simulation parameters to the high-fidelity positioning simulation module. It controls the initialization and startup of the simulation model within the high-fidelity positioning simulation module by issuing commands. During the simulation process, it receives simulation process data (including positioning accuracy, agent status, etc.) transmitted from the high-fidelity positioning simulation module in real time, thereby enabling the adjustment of simulation parameters (such as simulation step size and running rhythm) on the display management demonstration platform, ensuring the stability of the simulation process within the high-fidelity positioning simulation module. The specific simulation simulation process is controlled through interface control commands. These commands are based on the standard Ethernet frame protocol and defined in the network data segment, enabling the transmission of specified information between the high-fidelity positioning simulation module, the display management demonstration platform, and the simulation platform module. This supports users in developing simulation scenario planning based on open interfaces. The frame format of this interface can be defined as shown in Table 1. Table 1 Interface Frame Format Definition Table Based on the interface frames defined in Table 1, before the simulation begins, the user creates various simulation model devices supported by the system's built-in models in the simulation platform module, and plans the running trajectory and related simulation time information of each device (hereinafter referred to as "simulation scenario"). During the operation, the simulation platform module listens for the response of each model creation equipment, sends a simulation start command, starts the simulation process, sends the first frame of time synchronization, periodically synchronizes the planned position information to each device unit, and processes the device's response. It waits for the virtual and physical devices in the high-fidelity positioning simulation module in the system to complete the first frame of simulation and respond correctly before initiating the second frame of time synchronization and starting the second frame of simulation processing. The simulation stops when the path information and time information planned in the simulation scenario are completed or when a simulation stop command is received from the user.

[0019] (3) Data feedback: The simulation platform module receives real-time data (including running status, real-time error, real-time log, real-time weather, real-time terrain, etc.) and simulation result data (including simulation calculation location data, planned path information, error information, log information, weather information and terrain occlusion information, etc.) transmitted from the high-fidelity positioning simulation module. Then, it transmits these data to the display management demonstration platform through a standardized bus so that they can be visualized and generated in the display management demonstration platform. After the simulation in the high-fidelity positioning simulation module ends, the simulation platform module also needs to issue control commands to stop the high-fidelity positioning simulation module from running, so as to summarize the simulation data and complete the simulation simulation loop.

[0020] The display management demonstration platform is used to receive simulation parameters input or modified by users, as well as simulation process and result data, and to perform visualization display, simulation report generation, and simulation simulation operations.

[0021] This module is the core of the system's human-computer interaction, responsible for configuring simulation parameters, visualizing simulation process data, demonstrating simulation results, and generating reports. It solves the problems of cumbersome operation and unintuitive display of simulation results in existing systems, and realizes convenient control of the simulation process and clear presentation of results.

[0022] Based on this, the display management demonstration platform includes a demonstration control unit 22, a data visualization unit 23, a simulation report generation unit 24, and a simulation parameter input unit 29. Both the demonstration control unit and the data visualization unit are developed using the Qt framework to ensure a smooth interface and convenient operation. The data visualization unit is also equipped with a 1920×1080 resolution display screen.

[0023] Therefore, the specific workflow of the demonstration control unit 22, data visualization unit, simulation report generation unit, and simulation parameter input unit is as follows: The demonstration control unit receives simulation parameters transmitted from the simulation parameter input unit, performs legality verification, and then transmits them to the simulation platform module. At the same time, it receives simulation process data transmitted from the simulation platform module and replays the simulation process to demonstrate the simulation process.

[0024] This module first verifies the validity of the simulation parameters input from the simulation parameter input unit, including their value range, parameter configuration, and whether they are mutually exclusive with other parameter configurations. If the verification is invalid, a message indicating the valid parameter range is displayed on the interface. If the verification is successful, the simulation parameters (usable simulation parameters) are transmitted to the simulation platform module to configure the simulation parameters. Simultaneously, this demonstration control unit also receives simulation process data transmitted from the simulation platform module, supporting users to replay the simulation process for subsequent analysis and optimization of various subsystems.

[0025] The data visualization unit receives simulation process data transmitted from the simulation platform module and uses charts and 3D scenes to visualize the agent's positioning status, environmental parameter changes, and positioning errors in the simulation process or simulation result data. It can also receive GIS map data and real-time meteorological data transmitted from the GIS map module and the real-time meteorological access module, respectively. Combined with the simulation result data, it performs QT layered rendering to visualize the simulation results of 3D realistic terrain, real-time environmental data, agent dynamic trajectory, and positioning errors.

[0026] In this module, the data visualization unit receives simulation process data (including running status, simulation calculation location data, planned path information, error information, log information, meteorological information, terrain occlusion information, etc.) and simulation result data. It uses charts, 3D scenes, and other methods to intuitively display data such as the agent's positioning status, environmental parameter changes, and positioning errors during the simulation process or results. On the other hand, it can also receive GIS map data and real-time meteorological data transmitted from the GIS map module and real-time meteorological access module, respectively. Through a QT graphical interface, it achieves multi-dimensional visualization. Specifically, combined with simulation result data, it loads the map online and displays the map information. Using QT's layered rendering capabilities, it supplements the map interface with meteorological information based on coordinate alignment and dynamic mapping, ultimately achieving the superposition of undulating 3D realistic terrain, real-time rain, snow, and fog weather effects, agent dynamic trajectory, and real-time positioning error. It also achieves immersive visualization integrating environment, motion, and accuracy, supporting map zooming in and out, with meteorological information updated in real-time according to the central target coordinates. Furthermore, the data visualization unit can also perform simulation result comparison, comparing the simulated positioning accuracy with the actual accuracy of the installed equipment.

[0027] The simulation report generation unit is used to extrapolate the simulation result data transmitted by the simulation platform module and generate a formatted simulation report; This module first receives and automatically summarizes all simulation result data, generating a formatted simulation report that includes simulation parameters, process data, result analysis, error assessment, and supports report export.

[0028] The simulation parameter input unit is used to receive simulation parameters input by the user and supports the user to dynamically modify the simulation parameters built into the system. At the same time, it transmits all simulation parameters to the demonstration control unit.

[0029] This module can not only receive simulation parameters input by the user, but also allows the user to dynamically modify the simulation parameters built into the system, including the positioning method type, GIS map range, meteorological data update frequency, environmental interference factors, etc. Then, these simulation parameters are input into the demonstration control unit to perform legality verification.

[0030] The device access module is used to receive and parse the data format of the device, match the interface type protocol, convert the protocol message into a system-recognizable format or perform protocol conversion configuration to support various types of devices, and finally transmit the device data to the data interaction module.

[0031] This module serves as the core interface between the system and the external physical device 10, enabling direct connection between the system and the intelligent agent (i.e., the physical device). It receives real-world scene parameters, positioning feedback data, and device operating status data transmitted by the physical device. Specifically, the physical device access module includes a data format conversion unit 13 and a semantic parsing module 26. This module uses a dual-interface design with both a standardized industrial Ethernet interface (RJ45) and an RS485 interface, supporting wireless network access for the motion intelligent agent physical device. It also supports hot-swapping of interfaces and is compatible with different types of physical devices (such as physical meteorological equipment, physical electromagnetic environment equipment, physical inspection robots, drones, and unmanned swarm terminals). Furthermore, the physical device access module can monitor the access status of the physical device in real time. If abnormal states such as device disconnection or data transmission interruption occur, it sends an abnormal signal to the display management demonstration platform to remind the user to check the device's connection.

[0032] Based on this, the specific workflow of the data format conversion unit and the semantic parsing module is as follows: The data format conversion unit is used to receive and parse the data format of the installed equipment, perform interface type protocol matching, realize the corresponding channel protocol switching, and generate protocol messages.

[0033] Based on the description of the equipment entity in the communication frame, this module matches the corresponding protocol format from the protocol sample library. After converting the verified multi-source data into a protocol format that conforms to the system's standardized bus interface definition through formatting and encapsulation, it generates a protocol message and transmits it to the semantic parsing module. Within the semantic parsing module, data normalization processing is performed (mapping the data range to the 0-1 interval) to eliminate format differences between different data sources, thereby ensuring that the data can be correctly recognized and processed by the real data processing module after transmission. Finally, after the conversion is completed, the data is transmitted to the data interaction module.

[0034] The semantic parsing module is used to receive protocol messages and convert them into a format that the system can recognize or to perform protocol conversion configuration in order to support various types of installed equipment.

[0035] This module defines a unified semantic metadata format based on ontology and embeds it into the interface as a "semantic parsing module." It integrates a "semantic verification engine" capability, automatically verifying semantic validity upon receiving data and returning error codes for invalid values. The module also remotely updates the semantic mapping relationship after rule adjustments via the interface. This solves the fusion error problem caused by semantic distortion of heterogeneous data, achieving a unified semantic understanding between "simulated values ​​and actual signals," and supporting the autonomous and dynamic semantic connection and conversion between actual equipment and simulated virtual equipment.

[0036] In addition, in an optional embodiment of the present invention, the Ethernet communication interface of the physical device access module can be replaced with a fiber optic interface + Bluetooth 5.0 module. The fiber optic interface is used for long-distance physical device access (transmission distance ≤ 10km), and the communication rate is increased to 1000Mbps. The Bluetooth 5.0 module is used for wireless access of small portable physical devices (such as handheld positioning terminals) (transmission distance ≤ 100m). The original interface adapter unit and signal isolation unit are retained to improve interface adaptability and transmission distance, which is suitable for long-distance and multi-type physical device access scenarios.

[0037] The data interaction module is used to receive data from the actual equipment, GIS map data, and real-time meteorological data, perform caching and verification processing, generate verified multi-source real data, and transmit it to the real data processing module.

[0038] This module serves as the core hub for multi-source real data interaction within the system. It receives various data transmitted from the physical equipment access module, GIS map module, and real-time weather access module, performs caching and verification processing to ensure data integrity, accuracy, and compatibility, and then transmits the processed data to the real data processing module. Based on this, the data interaction module includes a data caching unit 11 and a data verification unit 12. The data caching unit employs a FIFO caching mechanism (capacity 1024KB), and the data verification unit uses the CRC-32 verification method.

[0039] Based on this, the specific workflows of the data caching unit and the data verification unit are as follows: The data caching unit is used to receive and cache data from the installed equipment, GIS map data, and real-time meteorological data.

[0040] This module is for caching the received data: after transmitting real data from multiple sources (actual equipment data, GIS map data, real-time meteorological data) to this module, the real-time transmitted data is temporarily stored to avoid data loss due to excessively fast data transmission speed, and to provide a buffer for subsequent verification and format conversion.

[0041] The data verification unit is used to receive cached actual equipment data, GIS map data, and real-time meteorological data, and verify whether the start bit, end bit, data length, and check code of each data frame are correct. If they are, the data is error-free; otherwise, an error command is sent to the corresponding module for data retransmission.

[0042] This module is for verifying the received data: it verifies the cached data one by one to check the integrity and accuracy of the data, and verifies the start bit, end bit, data length and checksum of the data frame. If the verification fails, it sends a data retransmission instruction to the corresponding module (such as the physical equipment access module, GIS map module, real-time weather access module) to request retransmission.

[0043] The real data processing module receives real data from multiple sources, performs effective scene data filtering and noise reduction, and establishes a mapping relationship between the effective scene data and each simulation model to realize the conversion of effective scene data into a unified format that can be recognized by the simulation model. At the same time, the mapping relationship data is transmitted to the high-fidelity positioning simulation module.

[0044] This module is used to filter, denoise, and fuse multi-source real-world data (actual equipment data, GIS map data, and real-time meteorological data) transmitted by the data interaction module, extracting effective scene parameters. This establishes a mapping relationship between effective scene data from the multi-source real-world data and various simulation models, providing data support for the high-fidelity positioning simulation module and solving the problem of large deviations between existing system scene parameters and actual scenes. The real-world data processing module includes a scene parameter fusion unit 14, a data filtering unit 27, and a data denoising unit 28.

[0045] Based on this, the specific workflows of the scene parameter fusion unit, data filtering unit, and data noise reduction unit are as follows: The data filtering unit is used to receive real data from multiple sources and filter out valid scenario data based on simulation verification requirements.

[0046] This module is for the data filtering process: based on the simulation verification requirements, valid scenario data is filtered out and invalid scenario data (such as data with transmission errors or excessive abnormal fluctuations) is removed; the valid scenario data includes the actual geographical parameters of the installed equipment, environmental interference data, positioning feedback data, terrain, latitude and longitude, altitude data of GIS map, and real-time climate data such as temperature, wind speed, and precipitation.

[0047] The data denoising unit receives valid scene data and uses wavelet transform denoising method to decompose the valid scene data into wavelet decomposition, extract the low-frequency valid signal, and then reconstructs the data through inverse wavelet transform to generate denoised valid scene data.

[0048] This module describes the data denoising process: it employs wavelet transform denoising to process the filtered valid scene data, eliminating noise generated during data acquisition and transmission from the actual equipment (such as electrical noise and environmental interference noise), thereby improving data accuracy. Specifically, it performs wavelet decomposition on the valid scene data to extract low-frequency valid signals, suppresses high-frequency noise signals, and then reconstructs the data through inverse wavelet transform, ultimately generating denoised valid scene data.

[0049] The scene parameter fusion unit is used to receive the noise-reduced effective scene data, map the GIS map data in the effective scene data to the simulation terrain model, map the real-time meteorological data in the effective scene data to the simulation environment interference model, and map the actual equipment data in the effective scene to the intelligent agent positioning simulation model, thereby generating the mapping relationship between the effective scene data and each simulation model, so as to realize the conversion of the effective scene data into a unified format that can be recognized by the simulation model.

[0050] This module handles the scene parameter fusion process: it fuses the denoised effective scene data and establishes a mapping relationship. Specifically, it maps GIS map data from the effective scene data to a simulated terrain model, real-time meteorological data from the effective scene data to a simulated environmental interference model, and actual equipment data from the effective scene data to an intelligent agent positioning simulation model (the intelligent agent positioning simulation model is a positioning simulation model of virtual devices). After establishing this mapping relationship, the effective scene data is converted into a unified format recognizable by the simulation model. Then, the mapping relationship data is transmitted to the realism positioning simulation module.

[0051] The high-fidelity positioning simulation module is used to receive simulation parameters and mapping relationship data, execute the simulation process after initializing the simulation model, monitor whether the running status of the simulation process is abnormal, and output abnormal data if so, otherwise output simulation process and result data.

[0052] This module is the core of the system for high-fidelity positioning simulation. It is responsible for receiving the mapping relationship data (including effective scene data) transmitted by the real data processing module, configuring the intelligent agent positioning method, and having built-in mainstream navigation processing flows such as GNNS, inertial navigation, and visual navigation. It supports users to improve and replace the intelligent agent positioning method, thereby constructing typical / customized multi-agent positioning simulation models for simulation calculation. At the same time, it transmits the simulation process and result data to the inference simulation platform module, and then to the display management demonstration platform for visualization and demonstration processing.

[0053] Therefore, the high-fidelity positioning simulation module includes a simulation model generation unit 15, a simulation parameter configuration unit 16, a simulation process monitoring unit 17, and an influence factor quantification unit 25. The simulation model generation unit supports the selection and customization of typical algorithms using mainstream tools such as Matlab and Python, and supports the inference of deep learning models. The simulation parameter configuration unit supports real-time parameter adjustment, and the simulation process monitoring unit monitors the running status of the simulation process in real time.

[0054] Based on this, the specific workflows of the simulation model generation unit, simulation parameter configuration unit, simulation process monitoring unit, and impact factor quantification unit are as follows: The simulation model generation unit is used to receive mapping relationship data, initialize the simulation terrain model, simulation environment interference model and intelligent agent positioning simulation model, and at the same time receive simulation running parameters transmitted by the simulation parameter configuration unit and execute the simulation process.

[0055] This module handles the model initialization and simulation execution process: First, based on the mapping relationship data transmitted by the real data processing module, the 3D simulation scene is initialized, including a simulated terrain model (built based on GIS map data), multiple agent positioning simulation models (built based on actual equipment parameters or virtual equipment parameters), and a simulated environmental interference model (built based on real-time meteorological data and actual equipment environmental data). Simultaneously, based on the initialized simulation models and simulation operation parameters, upon receiving a periodic synchronization command from the simulation platform module, each simulation model is started to run a specified simulation cycle, executing the multi-agent collaborative positioning simulation logic and simulating the agent's positioning accuracy, positioning error, and other data in real time.

[0056] The simulation parameter configuration unit is used to receive simulation parameters transmitted by the simulation platform module and configure the simulation operation parameters of the simulation terrain model, simulation environment interference model, and intelligent agent positioning simulation model.

[0057] This module handles the simulation parameter configuration process as follows: First, it receives simulation parameters (including simulation duration, number of agents, localization method type, etc.) from the simulation platform module. Then, combining this with mapping relationship data transmitted from the real data processing module, it configures the simulation parameters and data for each simulation model to ensure consistency between the simulation models and the actual scenario. It also configures user-defined typical localization methods or custom localization algorithms, as well as the integration of user-defined deep learning inference models. Finally, the configured simulation parameters are transmitted to the simulation model generation unit for execution of the simulation process.

[0058] The simulation process monitoring unit is used to monitor the running status of the simulation terrain model, simulation environment interference model, and intelligent agent positioning simulation model in real time during the simulation process, and to determine whether there is an abnormality in the simulation process. If so, the simulation is paused and an abnormality is indicated, and the abnormality data is sent to the simulation platform module. Otherwise, the simulation process continues to be executed, and the real-time simulation process data is sent to the simulation platform module.

[0059] This module monitors the running status of the simulation process: when the simulation model generation unit is performing simulation, it monitors the running status of each simulation model in the simulation model generation unit in real time. If a model crash or data abnormality occurs, it immediately sends an abnormal signal to the simulation platform module, pauses the simulation and prompts the abnormality. Otherwise, it directly transmits the simulation process data and simulation result data to the simulation platform module.

[0060] The GIS map module receives simulation parameters, parses and converts the GIS map range parameters within the simulation parameters, generates GIS map data that the system can recognize, constructs a 3D map model with terrain features, and transmits the GIS map data to the data interaction module.

[0061] This module plans the location information and route scenarios for each device (actual or virtual) based on the simulation platform module. Combined with the simulation timeline, it loads and parses GIS map data online to build a realistic terrain simulation foundation. It also transmits multi-dimensional terrain information to the system in real time, thereby providing geographic information support for high-fidelity simulation scenarios and solving the problem that the existing system's terrain simulation is not realistic and cannot reflect the actual geographical environment.

[0062] Therefore, the GIS map module includes a GIS data loading unit 18 and a terrain modeling unit 19. The GIS data loading unit supports loading various GIS data formats such as shp and tif, and uses the GDAL open-source library to implement data parsing. The terrain modeling unit uses OpenGL technology to construct a three-dimensional terrain model to ensure the accurate presentation of terrain details.

[0063] Based on this, the specific workflows of the GIS data loading unit and the terrain modeling unit are as follows: The GIS data loading unit is used to receive simulation parameters, load the corresponding area's GIS map data based on the GIS map extent parameters in the simulation parameters, perform data parsing and format conversion operations, and generate GIS map data that the system can recognize.

[0064] This module handles the data loading process: it receives path parameters simulated and planned by the user for each device through the simulation platform module via a standardized bus, loads GIS map data (including latitude and longitude, altitude, terrain slope, landform type, etc.) for the corresponding path area, parses the data, and converts it into a format that the system can recognize.

[0065] The terrain modeling unit is used to receive GIS map data that the system can recognize, to construct a 3D map model containing terrain features, and to transmit the GIS map data to the data interaction module.

[0066] This module describes the terrain modeling process: First, based on the parsed GIS map data, a three-dimensional terrain model is constructed to restore the terrain features of the actual geographical environment, including landforms such as mountains, plains, and water bodies. At the same time, the GIS map data is transmitted to the data interaction module for integration with other data by the real-time quantification module of the influence factor integrated within the high-fidelity positioning simulation module.

[0067] In addition, the GIS map module also supports real-time updates of GIS map data. Users can upload new GIS map data through the display management demonstration platform, and the terrain model will be parsed and updated in real time through the GIS data loading unit to ensure that the simulated terrain is consistent with the actual geographical environment.

[0068] In another optional embodiment of the present invention, the GDAL open-source library of the GIS map module can be replaced with ArcGIS Runtime SDK. ArcGIS Runtime SDK has stronger GIS data parsing and processing capabilities, supports more GIS data formats, can load high-precision satellite image maps, improves the accuracy of terrain modeling, and is suitable for scenarios with high requirements for geographic information accuracy.

[0069] In an optional embodiment of the present invention, the terrain modeling unit of the GIS map module can also use OSG (OpenSceneGraph) technology instead of OpenGL technology. OSG technology supports fast loading and rendering of large-scale terrain data, is suitable for simulation scenarios of large-scale GIS maps, and can improve the loading speed and running stability of terrain models.

[0070] The real-time weather access module is used to receive real-time weather data, interpret and filter it, convert its format, generate real-time weather data that the system can recognize, and transmit it to the data interaction module.

[0071] This module accesses real-time meteorological data through an online meteorological information center interface, simulates meteorological changes in complex environments, provides dynamic environmental interference support for high-fidelity simulation scenarios, and solves the problems of existing systems being unable to simulate real-time meteorology and lacking environmental realism.

[0072] Therefore, the real-time weather access module includes a weather data interface unit 20 and a weather data parsing unit 21. The weather data interface unit adopts an Ethernet interface, which supports connection to the meteorological department's public data interface or dedicated weather sensors.

[0073] Based on this, the specific workflow of the meteorological data interface unit and the meteorological data parsing unit is as follows: The meteorological data interface unit is used to receive real-time meteorological data and supports multiple interface inputs.

[0074] This module handles the data access process: it connects to real-time external meteorological data (including parameters such as temperature, wind speed, precipitation, visibility, and air pressure) via an Ethernet interface, and supports two access methods: connecting to the meteorological department's public API interface (such as the China Meteorological Data Network API), or connecting to on-site deployed meteorological sensors to ensure the real-time performance and accuracy of the data.

[0075] The meteorological data parsing unit is used to parse and filter real-time meteorological data, extract climate parameters related to multi-agent positioning, convert the format, generate real-time meteorological data that the system can recognize, and transmit it to the data interaction module.

[0076] This module is responsible for the data parsing and filtering process: it parses and filters the incoming real-time meteorological data, extracts climate parameters related to multi-agent positioning, removes irrelevant data, converts the parsed meteorological data into a format that the system can recognize, and transmits it to the data interaction module.

[0077] In addition, the real-time weather access module also supports real-time updates of weather data. It can update weather data in real time according to the update frequency set by the user and transmit it synchronously to the data interaction module to ensure that the climate environment in the simulation scenario is consistent with the actual real-time climate and to simulate the complex environment of dynamic change.

[0078] In an alternative embodiment of the present invention, the meteorological data interface unit of the real-time meteorological access module can be replaced with a LoRa wireless interface for accessing remotely deployed meteorological sensors (transmission distance ≤ 3km), eliminating the need for wired connections and making it suitable for outdoor scenarios without network coverage.

[0079] In an optional embodiment of the present invention, the industrial Ethernet bus of the standardized bus 9 can be replaced with a CAN bus. The communication protocol adopts the CANopen protocol, the transmission rate is 500kbps, and it supports multi-node, long-distance transmission (≤1km). The bus interface adopts the DB9 interface, which is suitable for system deployment in industrial sites and strong interference environments. It can improve the anti-interference capability of the system, and the stability and security of data transmission remain unchanged. This interface can also use high-speed buses such as SRIO and PCIE.

[0080] In summary, the virtual-real fusion intelligent agent positioning simulation system proposed in this invention achieves collaborative operation of the simulation platform module, display management and demonstration platform, actual equipment access module, data interaction module, real data processing module, high-fidelity positioning simulation module, GIS map module, and real-time weather access module. It can accurately solve the technical problems of existing simulation systems, such as unrealistic simulation of complex environments, inability to access actual equipment, lack of GIS terrain and real-time climate support, disconnect between simulation and demonstration, and poor adaptability. Specifically: (1) It solves the technical problems of not being able to directly access the actual equipment and the difficulty in obtaining parameters of real and complex environmental scenarios, thereby improving the authenticity and reference value of the simulation data. By setting up an actual equipment access module and adopting actual equipment access based on virtual and real equipment protocol conversion, the simulation system can directly connect with multiple types of actual equipment. It can receive real geographical parameters, environmental interference data, positioning feedback data and equipment operation status data transmitted by actual equipment in real time. At the same time, the data interaction module ensures the integrity, accuracy and compatibility of actual equipment data through data caching, CRC-32 verification, format conversion and other technologies. The real data processing module filters, reduces noise and fuses the data, and establishes a mapping relationship between real data and simulation model. Ultimately, this breakthrough overcomes the limitation of existing simulation systems' difficulty in connecting to actual equipment, enabling simulation parameters to no longer rely on manual presets but instead construct simulation scenarios based on real data from the actual equipment. This significantly reduces the deviation between simulation parameters and actual scenarios, allowing simulation results to accurately reflect the actual positioning performance of the actual equipment. It endows the actual equipment with the ability to perform simulation operations in complex real environments under specified map coordinates within the test field, enhancing the reference value and reliability of simulation results. This provides real and effective data support for the optimization of positioning algorithms and system deployment of actual equipment, solving the problem of existing simulation results being disconnected from practical applications. (2) It solves the technical problem of unrealistic simulation of complex environments and can build highly realistic simulation scenes. Through the synergistic effect of three core technical features, it achieves accurate simulation of complex environments. Specifically, it uses the GIS map module to load GIS data of various formats using the GDAL open source library and constructs a three-dimensional terrain model through OpenGL technology to accurately restore the latitude, longitude, altitude, terrain slope and other features of the actual geographical environment; it uses the real-time meteorological access module to connect to meteorological API or meteorological sensors through Ethernet interface to access meteorological data such as temperature, wind speed and precipitation in real time and update it dynamically at a preset frequency to simulate the dynamic changes of the climate environment; it uses the high-fidelity positioning simulation module to deeply integrate GIS map data, real-time meteorological data and real data of actual equipment to construct a terrain model, environmental interference model and intelligent agent model that fits the reality. Ultimately, a three-dimensional fusion simulation of "geographical environment + real-time climate + actual equipment data" was achieved, which completely solved the shortcomings of existing simulation systems that can only simulate fixed interference and have low environmental realism. The simulation scene fully covers the main influencing factors of the actual complex environment, and can realistically reproduce the working state of the actual equipment under complex geographical and dynamic climate conditions, making the simulation verification more targeted and effectively avoiding the problem of verification results failure due to excessive deviation between the simulation environment and the actual environment. (3) It solves the technical problems of disconnect between simulation and demonstration and cumbersome operation, and improves the convenience and intuitiveness of simulation verification. A display management and demonstration platform is set up, integrating the demonstration control unit, data visualization unit and simulation report generation unit, and equipped with an industrial display screen. It can realize the configuration of simulation parameters, real-time visualization of the simulation process, demonstration of simulation results and automatic generation of standardized reports. At the same time, the simulation platform module coordinates the work of each module to realize the automated control of the simulation process. Users do not need to manually intervene in the startup and operation of each module. They can complete the entire process operation through the display management and demonstration platform. Finally, it realizes the integrated closed loop of "simulation deduction-visualization demonstration-result analysis", which solves the problems of cumbersome operation, unintuitive simulation process and time-consuming and laborious result analysis of the existing simulation system. The data visualization unit intuitively displays data such as positioning accuracy and environmental parameter changes in the simulation process through charts, three-dimensional scenes and other forms, which makes it easy for users to quickly grasp the simulation status. The simulation report generation unit automatically summarizes simulation data, reduces manual processing costs and improves the efficiency of simulation verification. At the same time, it supports simulation process playback, which provides a convenient means for subsequent positioning algorithm optimization and problem investigation.

[0081] (4) It solves the technical problems of closed system architecture, poor adaptability, and weak scalability, and improves the system's versatility and maintainability. It adopts a modular structure design, with each module having independent functions and working together, and connected through a standardized bus. The bus adopts a unified communication protocol and interface specifications. At the same time, each module supports flexible replacement and expansion. For example, the interface type of the actual equipment access module can be replaced, the simulation engine of the high-fidelity positioning simulation module can be replaced, and the data parsing library of the GIS map module can be replaced, without modifying the overall system structure. Ultimately, it breaks the limitations of the existing simulation system architecture being closed and difficult to expand. Users can flexibly replace or expand the corresponding modules according to different simulation needs (such as small unmanned clusters and aerospace embedded systems), improving the system's versatility and adaptability. The modular design makes system maintenance more convenient. When a module fails, it can be repaired or replaced separately without stopping the entire system, reducing maintenance costs. The standardized bus ensures stable and compatible data transmission between modules, supports parallel operation of multiple modules, and improves the system's operating efficiency and stability.

[0082] (5) It solves the technical problems of increased computational complexity, insufficient real-time performance, and difficulty in traceability of results after combining environmental data, ensuring the computational efficiency and process reproducibility of large-scale simulations. By coordinating the simulation process monitoring unit integrated within the high-fidelity positioning simulation module and the data caching unit in the data interaction module, the data flow and computation scheduling logic of large-scale multi-agent simulation are optimized, ensuring efficient processing of massive concurrent data on a standardized bus. At the same time, by utilizing the simulation report generation unit within the simulation platform module and the display management demonstration platform, a complete data log system is established to globally record and automatically archive the simulation parameter configuration, scene parameter fusion data, and final output results throughout the entire simulation process. Ultimately, this improves the system's concurrent processing capability and real-time computation efficiency, meeting the real-time computation needs of massive data in complex environments. The globally standardized data recording mechanism solves the pain point of untraceable abnormal results, ensuring that the parameter configuration, intermediate states, and final results of each simulation experiment are interconnected and fully reproducible, providing a reliable historical data traceability basis for the iterative optimization and fault diagnosis of the positioning algorithm.

[0083] (6) This solution addresses the technical problem of relying on a single verification indicator and being unable to comprehensively evaluate the overall collaborative performance of multiple agents in complex environments, establishing a multi-dimensional comprehensive performance evaluation system. Based on the deep integration of the high-fidelity positioning simulation module and the simulation platform module, in addition to traditional absolute positioning accuracy calculations, a comprehensive evaluation model for the collaborative characteristics of multiple agents (including key indicators such as anti-interference robustness, communication delay tolerance, and collaborative consistency under dynamic topology) is introduced and set through the simulation model generation unit and simulation parameter configuration unit. Furthermore, the data visualization unit in the display management demonstration platform presents the aforementioned multi-dimensional evaluation data in real-time and intuitive graphical form. Ultimately, this changes the limitation of existing technologies that judge the merits of algorithms solely based on the single standard of "positioning error," constructing a three-dimensional, comprehensive positioning performance evaluation closed loop. The system can accurately characterize the comprehensive survivability and collaborative performance of various positioning algorithms in different complex environments, ensuring that the simulation evaluation results can comprehensively and objectively reflect the overall operational capabilities of the multi-agent system in real complex environments, effectively guiding the strategy selection of the actual deployment system.

[0084] (7) This system solves the technical problem of fixed and singular environmental interference models that cannot realistically simulate dynamic topology evolution and multi-source coupled interference, and deeply restores complex and harsh dynamic forced environments. Through the scene parameter fusion unit built into the real data processing module, the physical space occlusion attributes extracted by the terrain modeling unit in the GIS map module are deeply coupled with the dynamic meteorological attenuation factors extracted by the meteorological data parsing unit in the real-time meteorological access module. At the same time, combined with the high-fidelity positioning simulation module, a dynamic network topology and sudden impact factor model that can evolve in real time with the high-speed movement of multiple agents is constructed. The virtual and real equipment clock, position reference impact factor real-time quantization and adaptive damage algorithm are embedded and injected into the simulation platform module. Finally, it fills the technical gap that the existing simulation system cannot dynamically simulate the superimposed evolution of multiple interferences. The system can not only simulate static signal attenuation, but also realistically reproduce the extreme dynamic conditions such as communication link disconnection, topology reconstruction and positioning jump caused by agents traversing complex terrain (such as physical occlusion in valleys) and superimposed with severe weather (such as signal absorption in rainstorms). This highly realistic simulation of multi-source coupled interference makes the environmental evolution model highly aligned with the real world, improving the rigor and effectiveness of the verification of positioning anti-interference algorithms in response to sudden environmental changes.

[0085] Specific embodiments have been used to illustrate the principles and implementation methods of this invention. The descriptions of the embodiments above are only for the purpose of helping to understand the method and core ideas of this invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.

[0086] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the principles of the invention, and should be understood that the scope of protection of the invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of this invention.

Claims

1. A virtual-real fusion intelligent agent positioning simulation system for complex real-world environments, characterized in that, include: The simulation platform module is used to start the operation of each module and check whether the operation status of each module is abnormal. After that, it receives the simulation parameters transmitted by the display management demonstration platform and transmits them to the high-fidelity positioning simulation module. At the same time, it receives the simulation process and result data and transmits them to the display management demonstration platform. The display management demonstration platform is used to receive simulation parameters input or modified by users, as well as simulation process and result data, and to perform visualization display, simulation report generation, and simulation simulation operations. The physical equipment access module is used to receive and parse the data format of the physical equipment, match the interface type protocol, convert the protocol message into a format that the system can recognize or perform protocol conversion configuration to support various physical equipment, and finally transmit the physical equipment data to the data interaction module. The data interaction module is used to receive data from the actual equipment, GIS map data, and real-time meteorological data, perform caching and verification processing, generate verified multi-source real data, and transmit it to the real data processing module. The real data processing module is used to receive real data from multiple sources, perform effective scene data filtering and noise reduction, and establish a mapping relationship between the effective scene data and each simulation model to realize the conversion of effective scene data into a unified format that can be recognized by the simulation model. At the same time, the mapping relationship data is transmitted to the high-fidelity positioning simulation module. The high-fidelity positioning simulation module is used to receive simulation parameters and mapping relationship data, execute the simulation process after initializing the simulation model, monitor whether the running status of the simulation process is abnormal, and output abnormal data if so, otherwise output simulation process and result data. The GIS map module is used to receive simulation parameters, parse and convert the GIS map range parameters within the simulation parameters, generate GIS map data that the system can recognize, construct a 3D map model with terrain features, and transmit the GIS map data to the data interaction module. The real-time weather access module is used to receive real-time weather data, interpret and filter it, convert its format, generate real-time weather data that the system can recognize, and transmit it to the data interaction module.

2. The virtual-real fusion intelligent agent positioning simulation system in complex real-world environments according to claim 1, characterized in that, The display management demonstration platform includes a demonstration control unit, a data visualization unit, a simulation report generation unit, and a simulation parameter input unit; The demonstration control unit is used to receive simulation parameters transmitted by the simulation parameter input unit, perform legality verification, and then transmit them to the deduction simulation platform module. At the same time, it receives simulation process data transmitted by the deduction simulation platform module and replays the simulation process to demonstrate the simulation process. The data visualization unit is used to receive simulation process data transmitted from the simulation platform module, and to visualize the agent's positioning status, environmental parameter changes and positioning errors in the simulation process or simulation result data using charts and 3D scenes. It can also receive GIS map data and real-time meteorological data transmitted from the GIS map module and the real-time meteorological access module, respectively, and perform QT ​​layered rendering in combination with the simulation result data to visualize the simulation results of 3D real terrain, real-time environmental data, agent dynamic trajectory and positioning error. The simulation report generation unit is used to extrapolate the simulation result data transmitted by the simulation platform module and generate a formatted simulation report; The simulation parameter input unit is used to receive simulation parameters input by the user and supports the user to dynamically modify the simulation parameters built into the system. At the same time, it transmits all simulation parameters to the demonstration control unit.

3. The virtual-real fusion intelligent agent positioning simulation system in complex real-world environments according to claim 1, characterized in that, The actual equipment access module includes a data format conversion unit and a semantic parsing module; The data format conversion unit is used to receive and parse the data format of the actual equipment, perform interface type protocol matching, realize the corresponding channel protocol switching, and generate protocol messages. The semantic parsing module is used to receive protocol messages and convert them into a format that the system can recognize or to perform protocol conversion configuration in order to support various types of installed equipment.

4. The virtual-real fusion intelligent agent positioning simulation system in complex real-world environments according to claim 1, characterized in that, The data interaction module includes a data caching unit and a data verification unit; The data caching unit is used to receive and cache data from the installed equipment, GIS map data, and real-time meteorological data. The data verification unit is used to receive cached actual equipment data, GIS map data, and real-time meteorological data, and verify whether the start bit, end bit, data length, and check code of each data frame are correct. If they are, the data is error-free; otherwise, an error command is sent to the corresponding module for data retransmission.

5. The virtual-real fusion intelligent agent positioning simulation system in complex real-world environments according to claim 1, characterized in that, The real data processing module includes a scene parameter fusion unit, a data filtering unit, and a data noise reduction unit; The data filtering unit is used to receive real data from multiple sources and filter valid scenario data based on simulation verification requirements; The data denoising unit is used to receive valid scene data and use wavelet transform denoising method to decompose the valid scene data into wavelet decomposition, extract low-frequency valid signals, and then reconstruct the data through inverse wavelet transform to generate denoised valid scene data. The scene parameter fusion unit is used to receive the noise-reduced effective scene data, map the GIS map data in the effective scene data to the simulation terrain model, map the real-time meteorological data in the effective scene data to the simulation environment interference model, and map the actual equipment data in the effective scene to the intelligent agent positioning simulation model, thereby generating the mapping relationship between the effective scene data and each simulation model, so as to realize the conversion of the effective scene data into a unified format that can be recognized by the simulation model.

6. The virtual-real fusion intelligent agent positioning simulation system in complex real-world environments according to claim 1, characterized in that, The high-fidelity positioning simulation module includes a simulation model generation unit, a simulation parameter configuration unit, a simulation process monitoring unit, and an impact factor quantification unit. The simulation model generation unit is used to receive mapping relationship data, initialize the simulation terrain model, simulation environment interference model and intelligent agent positioning simulation model, and at the same time receive the simulation running parameters transmitted by the simulation parameter configuration unit and execute the simulation process. The simulation parameter configuration unit is used to receive simulation parameters transmitted by the simulation platform module and configure the simulation operation parameters of the simulation terrain model, simulation environment interference model and intelligent agent positioning simulation model. The simulation process monitoring unit is used to monitor the running status of the simulation terrain model, simulation environment interference model, and intelligent agent positioning simulation model in real time during the simulation process, and to determine whether there is an abnormality in the simulation process. If so, the simulation is paused and an abnormality is indicated, and the abnormality data is sent to the simulation platform module. Otherwise, the simulation process continues to be executed, and the real-time simulation process data is sent to the simulation platform module.

7. The virtual-real fusion intelligent agent positioning simulation system in complex real-world environments according to claim 1, characterized in that, The GIS map module includes a GIS data loading unit and a terrain modeling unit; The GIS data loading unit is used to receive simulation parameters, load the GIS map data of the corresponding area according to the GIS map range parameters in the simulation parameters, perform data parsing and format conversion operations, and generate GIS map data that the system can recognize. The terrain modeling unit is used to receive GIS map data that the system can recognize, to construct a 3D map model containing terrain features, and to transmit the GIS map data to the data interaction module.

8. The virtual-real fusion intelligent agent positioning simulation system in complex real-world environments according to claim 1, characterized in that, The real-time weather access module includes a weather data interface unit and a weather data parsing unit; The meteorological data interface unit is used to receive real-time meteorological data and supports multiple interface inputs. The meteorological data parsing unit is used to parse and filter real-time meteorological data, extract climate parameters related to multi-agent positioning, convert the format, generate real-time meteorological data that the system can recognize, and transmit it to the data interaction module.

Images