Heterogeneous unmanned platform test method and system based on virtual-real combination

Abstract

The invention discloses a heterogeneous unmanned platform test method and system based on virtuality and reality combination, and solves the interconnection, interworking and interoperation problems of various heterogeneous test objects through an LVC integrated joint simulation technology, and further utilizes a virtuality and reality interaction method on the basis of the interconnection, interworking and interoperation problems. The method comprises the following steps of: fusing simulation elements in a test process of a real physical test, adding a physical test link in a virtual simulation test process, performing a virtual simulation test by utilizing data from the real physical test, and broadening and expanding test verification contents by utilizing virtual simulation; and a set of virtual-real combined test environment system is constructed for intelligent unmanned platform integration test verification, and system-level heterogeneous unmanned platform test verification based on a specific scene can be completed. According to the invention, interconnection, intercommunication and interoperation of various heterogeneous test objects are realized, the accuracy and credibility of the test are improved, and the test cost is reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of virtual-reality combined testing of unmanned platforms, and in particular relates to a heterogeneous unmanned platform testing method and system based on virtual-reality combination. Background Art

[0002] With the continuous advancement of artificial intelligence and related information technologies, unmanned platforms such as drones, unmanned vehicles, and bionic robots have developed vigorously, and their information-based, systematized, intelligent, and precise characteristics have become increasingly obvious. This has led to the expansion of test objects from a single platform to a system, the transformation of test assessment content from indicator assessment to performance evaluation, and the transformation of experimental environments from a single environment to a complex environment, thus bringing about a transformation in test methods from focusing only on the external field to a combination of internal and external, and virtual and real.

[0003] The Chinese journal paper "A Review of UAV Autonomous Combat Capability Test and Evaluation Technology" records the discussion and research on the test and evaluation technology of unmanned platforms with intelligent autonomous capabilities, but this method only considers the integrated test between drone clusters and is difficult to expand to system tests composed of multiple heterogeneous unmanned platforms such as drones, unmanned vehicles, and bionic robots; Chinese patent document CN113721481A discloses a virtual-real integrated bridge equipment test platform and test method, the integrated bridge equipment test platform includes a database system and a test system, but the method's research on virtual-real simulation technology is mostly focused on interoperability research, and generally does not involve how to use virtual simulation experiments (including V and C resources) in combination with real physical experiments to expand test subjects, and how to use real physical test data to improve the accuracy and credibility of virtual simulation experiments. Summary of the invention

[0004] The purpose of the present invention is to provide a heterogeneous unmanned platform test method and system based on the combination of virtual and real, realize the interconnection and interoperability of various heterogeneous test objects through LVC integrated joint simulation technology, and further utilize virtual and real interaction to improve the accuracy and credibility of the test and reduce the test cost.

[0005] In order to achieve the purpose of the present invention, on the one hand, the present invention provides a heterogeneous unmanned platform test method based on the combination of virtual and real, the method comprising the following steps:

[0006] Step 1: Decompose the test objectives;

[0007] Step 2: Prepare test scenarios, compile task documents, and determine the composition of the test equipment group;

[0008] Step 3: Build a real-scene environment: Build a real-scene test environment for typical application scenarios;

[0009] Step 4: Build a virtual environment: construct a three-dimensional digital twin scene based on real physical elements;

[0010] Step 5: Joint debugging of participating equipment: Dynamically test the interfaces and matching relationships between various types of participating unmanned platforms, systems and related systems;

[0011] Step 6: Synchronize time and space benchmarks: complete the time synchronization and coordinate integration of physical equipment, semi-physical equipment, simulation environment, test instruments, and basic conditions;

[0012] Step 7, virtual-real mapping: using the optical motion capture system and the UWB positioning system to restore the position, movement, and posture of the heterogeneous unmanned platform;

[0013] Step 8, virtual-real interaction: divide the heterogeneous unmanned platform into physical nodes and virtual nodes, and configure the number of nodes; use resource integration middleware to complete virtual-real interaction;

[0014] Step 9, real-time data acquisition: the optical motion capture system, the UWB positioning system, and the high-speed visual capture system are used to collect equipment data, and the simulation platform operation data is recorded at the same time, and the data management and control system is used to uniformly store and manage the data;

[0015] Step 10: Data fusion processing: Integrate heterogeneous data from different data sources to form a consistent format structure;

[0016] Step 11: Performance evaluation: Use the index method to evaluate, normalize the index parameters, and use the AHP hierarchical analysis method to determine the weight value of the performance evaluation.

[0017] On the other hand, the present invention also provides a heterogeneous unmanned platform test system based on the combination of virtual and real, the system comprising the following modules:

[0018] Scenario application module: builds a real-life test environment for typical application scenarios such as cities, blocks, buildings, and underground;

[0019] Functional service module: By taking the real-life test environment as the guide, a digital parallel environment is constructed to provide virtual-real combination simulation, algorithm training and verification, data management and control evaluation, data interconnection and interoperability, and to connect with the infrastructure and general platform construction of the general basic modules downward to form a guidance and constraint relationship in professional fields;

[0020] General basic module: It realizes the virtual-reality simulation function by supporting the virtual-reality simulation test verification of multiple heterogeneous unmanned platforms at the system level.

[0021] Compared with the prior art, the significant progress of the present invention is: 1. The present invention is based on LVC integrated joint simulation technology and integrates multiple heterogeneous unmanned platforms; 2. The present invention proposes a highly interactive virtual-reality combined test method to achieve a test mode in which real equipment interacts with real scenes, real equipment interacts with real equipment, real equipment interacts with virtual scenes, and real equipment interacts with virtual equipment, thereby improving the accuracy and credibility of the test, reducing the test cost, and improving the test effect.

[0022] In order to more clearly illustrate the functional characteristics and structural parameters of the present invention, further description is given below in conjunction with the accompanying drawings and specific implementation methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0024] Figure 1 It is a flowchart of the present invention;

[0025] Figure 2 It is a framework diagram of the virtual-real combination test environment of the present invention;

[0026] Figure 3 It is a principle block diagram of the virtual-real interactive information flow of the present invention. DETAILED DESCRIPTION

[0027] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0028] Combination Figure 1 and Figure 2 On the one hand, the present invention provides a heterogeneous unmanned platform test method based on the combination of virtual and real, the method comprising the following steps:

[0029] 1. Preparation stage:

[0030] Step 1: Decompose the test objectives: clarify the effectiveness or applicability of the test object, what conclusions to be obtained through the test and to what extent, whether the existing conditions can well support the conduct of the test, etc.;

[0031] Step 2: Prepare test scenarios, compile mission documents, and determine the composition of the test equipment. The mission documents include the composition of the unmanned equipment actual equipment, the composition of the semi-physical equipment, the composition of the virtual entity, the virtual and real environment debugging plan, the guidance and debugging tools, and the detailed test plan.

[0032] Step 3: Build a real-life environment: Build a real-life test environment in accordance with local conditions for typical application scenarios such as cities, blocks, underground, and buildings. The environment consists of realistic urban environments, complex natural environments, complex confrontation environments, and other infrastructure.

[0033] Step 4: Build a virtual environment: Construct an editable three-dimensional digital twin scene based on real physical elements such as test equipment and terrain environment.

[0034] 4-1: Select the geographical environment file and set the weather, meteorological and other factors according to the test requirements;

[0035] 4-2: Expand the virtual scene according to the test requirements. The environmental simulation model mainly constructs the geographical environment model, meteorological environment model and sound and light special effects model;

[0036] 4-3: Set the number of virtual entities, deployment locations, complete the task actions of virtual entities and edit the conditions for automatic end of training;

[0037] 4-4: Provide computational analysis, data support and interactive interface, as well as visual effects.

[0038] (II) Implementation phase:

[0039] Step 5: Joint debugging of participating equipment: Dynamically test the interfaces and matching relationships between various types of participating unmanned platforms, systems and related systems;

[0040] Step 6: Synchronize time and space benchmarks: Complete the time and coordinate integration of physical equipment, semi-physical equipment, simulation environment, test instruments, and basic conditions.

[0041] 6-1: Time reference calibration: Calibrate the time server to ensure that the time of all measuring devices and the devices under test are in the same world time to meet the needs of high-precision testing;

[0042] 6-2: Spatial benchmark calibration: Through the joint measurement and calculation of the benchmark points in the spatial layout, the coordinate system required for the test is obtained, including plane benchmark control joint measurement, elevation benchmark control joint measurement, three-dimensional benchmark control measurement, acquisition and determination of geodetic coordinates, etc.;

[0043] 6-3: Coordinate integration: Before the test begins, the time and coordinate integration of the installed equipment, test instruments, and basic conditions are completed through the time and space synchronization calibration system.

[0044] Step 7, virtual-reality mapping: Use the optical motion capture system and UWB positioning system to restore the position, movement, and posture of the heterogeneous unmanned platform; through the optical motion capture system, the movement position and posture of the intelligent equipment and personnel are captured and calculated, so as to drive the corresponding digital three-dimensional entity model to move synchronously in the virtual three-dimensional scene, and present the test situation in a virtual three-dimensional way. The data flow is that the infrared optical tracking camera is connected to the network switch through the network cable, and the motion capture server is also connected to the same network through the network cable to collect the posture data of the personnel and equipment in the scene in real time; the motion capture server solves the data collected by the camera, and the solved data is distributed to other workstation computers through the client network and the LAN wireless access point; in a macro sense, the display control server synchronizes the three-dimensional situation scene in real time.

[0045] Step 8, virtual-real interaction: Divide the heterogeneous unmanned platform into physical nodes and virtual nodes, and configure the number of nodes; use resource integration middleware to complete virtual-real interaction; the physical nodes include two types of test objects, namely, physical installation and hardware-in-the-loop, which correspond to the two types of test resources, L (physical) and V (semi-physical) in LVC, respectively, with high model accuracy, and directly connect the flight control system, airborne visual computer, networking communication link and other hardware entities to the simulation closed loop to form a hardware-in-the-loop simulation system, and achieve a simulation experiment effect with higher accuracy and credibility. Virtual nodes include various types of heterogeneous unmanned platforms simulated by software-in-the-loop. Under the premise of ensuring the basic simulation accuracy requirements, simplified motion and control models are adopted, which can realize the rapid simulation of large-scale intelligent unmanned platform clusters. In summary, physical nodes have high accuracy, but occupy more hardware and computing resources, and are deployed in small quantities in the overall cluster simulation; while virtual nodes have slightly lower accuracy, but occupy fewer resources and have low costs, and can be deployed in large quantities in cluster simulation. According to the test requirements, the number of physical nodes and virtual nodes of heterogeneous unmanned platforms such as drones, unmanned vehicles, and bionic robots can be reasonably configured. The accuracy and credibility of the test can be improved through the data of physical nodes. The test content (type of participating equipment, test subjects, etc.) can be expanded through the access of virtual nodes, which can achieve a balance between test accuracy and test cost. Figure 3 , to achieve better test verification effect.

[0046] Step 9, real-time data collection: The optical motion capture system, the UWB positioning system, and the high-speed visual capture system are used to collect equipment data, and the simulation platform operation data is recorded at the same time, and the data management and control system is used for unified storage and management to provide data support for algorithm training iteration;

[0047] (III) Summary stage:

[0048] Step 10: Data fusion processing: Integrate heterogeneous data from different data sources to form a consistent format structure;

[0049] Step 11, performance evaluation: The index method is used for evaluation. The index method can make full use of the performance indicators of the unmanned platform itself for quantitative evaluation and enhance the accuracy of the evaluation. It can also comprehensively use expert evaluation methods such as hierarchical analysis method to determine some coefficients; after normalizing the indicator parameters, the AHP hierarchical analysis method is used to determine the weight value of the performance evaluation.

[0050] 11-1: Establish a hierarchical structure model to decompose the evaluation objectives into multiple layers of indicators;

[0051] 11-2: Determine the hierarchical structure, and then determine the subordinate relationship between the upper and lower layers; starting from the second layer, for the same layer factors that belong to each factor in the previous layer, compare them two by two to construct a comparison matrix until the lowest layer;

[0052] The comparison matrix B is:

[0053]

[0054] Among them, B1, B2, ..., B n The next layer element dominated by the previous layer element A; b ij is the i-th element B i With the jth element B j The importance ratio between them is; n is the number of indicators in this layer affected by the element A in the previous layer;

[0055] The relative weight calculation algorithms under a single criterion include the sum method, characteristic root method, and root method. The weight calculation using the sum method as an example is as follows:

[0056] (1) Normalized judgment matrix B:

[0057]

[0058] in, is the element in the i-th row and j-th column of the normalized judgment matrix;

[0059] (2) Total by row:

[0060]

[0061] in, is the unnormalized weight coefficient after summing up by row;

[0062] (3) After normalization, the weight coefficient can be obtained:

[0063]

[0064] Among them, w i is the normalized weight coefficient;

[0065] (4) Calculate the maximum eigenvalue λ max :

[0066]

[0067] Among them, λ max is the largest characteristic root; w is the weight column vector composed of weight coefficients; (Bw) i is the i-th item in the column vector after the comparison matrix B is multiplied by the weight column vector w;

[0068] (5) Calculate the consistency index CI:

[0069] CI=(λ max -n) / (n-1);

[0070] (6) Find the average consistency index RI of the corresponding n and calculate the consistency ratio CR; RI is the result of a large number of statistical calculations:

[0071] Table 1 Average random consistency index

[0072] Order (n) 1 2 3 4 5 6 7 RI 0 0 0.52 0.89 1.12 1.26 1.36

[0073] CR = CI / RI;

[0074] When CR is less than 0.1, the consistency of the judgment matrix is ​​acceptable; when CR is greater than or equal to 0.1, the consistency of the judgment matrix does not meet the requirements, and the judgment matrix is ​​modified.

[0075] When the weights are synthesized from bottom to top, the synthetic weights of the elements in each layer for the overall target are obtained.

[0076] On the other hand, combined Figure 1 The present invention also includes a heterogeneous unmanned platform test system based on the combination of virtual and real, and the system includes the following three levels:

[0077] The scenario application layer is oriented towards test needs, and a real-life test environment is constructed for typical application scenarios such as cities, blocks, buildings, and underground. The environment consists of realistic urban environments, complex natural environments, complex adversarial environments, and other infrastructure: the realistic urban environment includes simulated building complexes, streets, complex roads, personnel models, obstacle facilities, and inflatable enclosed venues; the complex natural environment includes lighting simulation systems and weather simulation systems (rainfall simulators, fog machines, smoke machines, industrial fans, etc.); the complex adversarial environment includes electromagnetic interference environment systems and satellite navigation deception systems.

[0078] The functional service layer is guided by the real-life test environment to build a digital parallel environment, providing functional services such as virtual-reality simulation, algorithm training and verification, data management and evaluation, and data interconnection (resource integration middleware). It connects downward to the infrastructure and general platform construction of the general basic layer to form a guidance and constraint relationship in professional fields.

[0079] The general basic layer provides support for the realization of virtual-reality combined simulation functions, including the construction of basic hardware environment and 5G network communication system according to local conditions, supplementing and improving basic simulation resources such as urban environment resources, equipment data resources, simulation task resources, model mapping resources, etc., deploying optical motion capture systems, UWB positioning systems, high-speed visual capture systems and other data acquisition equipment on demand, and uniformly developing and building intelligent unmanned platform test operation environment, connecting the links of intelligent algorithm model training and release, and supporting system-level multi-type heterogeneous unmanned platform virtual-reality combined simulation test verification

[0080] Example

[0081] The present invention involves the following main hardware information: infrared motion capture camera: resolution exceeds 1000W pixels, and camera positioning and calibration accuracy reaches sub-millimeter level; server: CPU Intel i9-13900; graphics card: NVIDIA Geforce RTX4090; memory: Colorful brand, memory type DDR4, memory size 64GB (32GB×2), frequency 3600MHz; switch: TP-LINK brand 10G upload switch and 1G POE switch; space-time calibration timing server: Beidou Bangtai T800.

[0082] The system is characterized by the construction of a real-life environment platform and a digital parallel environment. The real-life environment platform interacts with the digital parallel environment and stores and analyzes the collected and measured data such as the position and posture of the unmanned system for the evaluation of the performance of the unmanned platform.

[0083] The real-scene environment construction platform consists of three parts: realistic urban environment system, complex natural environment system, and complex confrontation environment system. The realistic urban environment system is used to build real environmental elements, the complex confrontation environment system is used to build confrontation elements during the test, and the complex natural environment construction system is used to simulate light, electricity, wind, rain, fog, smoke and other environments. The above three environments are combined to complete the overall real-scene reproduction.

[0084] The data collection platform records and transmits test process data by building indoor and outdoor data collection equipment. The data collection content includes the position, speed, attitude and other states of the unmanned system, obstacles and terrain features and other information.

[0085] Take the neighborhood fire rescue mission as an example.

[0086] Real scene: The disaster reconnaissance mission was assigned to three drones. After receiving the mission, the drone cluster carried out the reconnaissance mission according to the track points. During the reconnaissance process, the drone cluster found out the damage to the two sides of the street, the center of the street, vehicles and buildings, and remotely detected the concentration of flammable and explosive dangerous chemicals and other related gases; after finding the rescue target, it fired fire bombs at key locations around the target to quickly control the fire source and guide ground fire trucks and firefighters to rescue the target.

[0087] The virtual scene restores the real scene, adds fire disaster simulation scenarios and dangerous gas leakage simulation scenarios, and realizes the superposition of real-scene building environment and virtual scene disaster simulation.

[0088] The virtual equipment adds 12 drones and 1 unmanned fire truck to the fire-fighting force, and expands the test space for drone clusters (real + virtual) to jointly perform autonomous navigation, obstacle avoidance and reconnaissance tasks. At the same time, drone clusters (real + virtual) and virtual unmanned vehicles verify performance indicators such as autonomous identification of disaster situations. After the virtual unmanned fire truck and drone cluster (real + virtual) discover the source of fire, they initiate fire bomb / fire hose strike commands to control fires in buildings and streets, realize the interaction between real equipment and real scenes, real equipment and virtual scenes, virtual equipment and virtual scenes, and virtual equipment and real scenes, and expand the test subjects.

[0089] On the one hand, the data acquisition system collects information about the environment, drone clusters and obstacles in the real scene, and automatically and proportionally restores the identified situation information in the virtual scene. On the other hand, it saves the information of the other party’s vehicle in the virtual environment, realizing the joint display of real equipment, real scene, virtual equipment and virtual scene.

[0090] The data acquisition system captures and calculates the motion posture of the physical drone cluster, drives the corresponding entity model to move synchronously in the virtual environment, and presents the real-time dynamic process of test subjects such as drone cluster (physical + virtual) autonomous navigation and obstacle avoidance, disaster reconnaissance, etc. on multiple display terminals to achieve parallel display.

[0091] The data acquisition system saves the status information of the virtual unmanned vehicle (location, power, reconnaissance video) in real time, interacts with the information of the drone cluster (real + virtual) to jointly complete the rescue activities.

[0092] An evaluation index system was established, and the weight of each factor was determined based on the AHP hierarchical analysis method to complete the performance evaluation.

[0093] The comprehensive comparison results of the virtual-real combination method, the real physical test method, and the virtual-real combination test method are shown in Table 2:

[0094] Table 2 Comprehensive comparison between virtual-real combination test method and other two test methods

[0095] Test methods cost cycle Credibility Repeatability Implementation complexity Process Hazards Real physical test method high long high Not repeatable high high Virtual simulation test method Low short Low Repeatable middle Low Virtual-real combination test middle middle high Partially repeatable middle Low

[0096] The advantages of the virtual-real combination test method are specifically manifested in:

[0097] Expand test subjects: In general, real physical tests have poor controllability and repeatability, and cannot fully evaluate the performance of equipment under test boundary conditions, complex electromagnetic environments, and complex confrontation conditions. The virtual-real test method can use simulation models or simulators instead of real equipment to ensure high test accuracy. By constructing a blue team and a complex test environment, it can carry out signal game attack and defense confrontation, which are difficult to implement with field real equipment tests alone, expand test subjects, and improve test intensity.

[0098] Reduce test costs: Real physical tests are costly, long, and complex. It is difficult to organize repeated tests in a limited time, and the organization and implementation are difficult. The test space constructed using virtual simulation test technology has greatly reduced costs, cycles, and organizational difficulties, and has strong controllability, which improves the accuracy and credibility of the test, can meet the requirements of multiple repeatable tests, and reduce test costs.

[0099] Improve verification algorithms: Algorithms are difficult to deploy quickly in actual equipment. In addition, the actual training process is subject to site restrictions, and the equipment does not yet have mobility. In this case, an integrated verification platform is needed to provide a simulated virtual environment consistent with the real scene, generate multimodal data sets required for algorithm training, provide algorithm migration and deployment capabilities, and conduct targeted training and verification of algorithms for actual equipment.

[0100] It should be noted that, in this article, the terms "comprises", "includes" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or apparatus that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or apparatus.

[0101] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A heterogeneous unmanned platform test method based on the combination of virtual and real, characterized in that: The following steps are involved: Step 1: Decompose the test objectives; Step 2: Prepare test scenarios, compile task documents, and determine the composition of the test equipment group; Step 3: Build a real-scene environment: Build a real-scene test environment for typical application scenarios; Step 4: Build a virtual environment: construct a three-dimensional digital twin scene based on real physical elements; Step 5: Joint debugging of participating equipment: Dynamically test the interfaces and matching relationships between various types of participating unmanned platforms, systems and related systems; Step 6: Synchronize time and space benchmarks: complete the time synchronization and coordinate integration of physical equipment, semi-physical equipment, simulation environment, test instruments, and basic conditions; Step 7, virtual-real mapping: using the optical motion capture system and the UWB positioning system to restore the position, movement, and posture of the heterogeneous unmanned platform; Step 8, virtual-real interaction: divide the heterogeneous unmanned platform into physical nodes and virtual nodes, and configure the number of nodes; use resource integration middleware to complete virtual-real interaction; Step 9, real-time data acquisition: the optical motion capture system, the UWB positioning system, and the high-speed visual capture system are used to collect equipment data, and the simulation platform operation data is recorded at the same time, and the data management and control system is used to uniformly store and manage the data; Step 10: Data fusion processing: Integrate heterogeneous data from different data sources to form a consistent format structure; Step 11: Performance evaluation: Use the index method to evaluate, normalize the index parameters, and use the AHP hierarchical analysis method to determine the weight value of the performance evaluation.

2. The heterogeneous unmanned platform test method based on the combination of virtual and real according to claim 1 is characterized in that: The task document in step 2 includes the composition of the actual unmanned equipment group, the composition of the semi-physical equipment group, the composition of the virtual entity group, the virtual and real environment debugging plan, the guidance and debugging tools, and the detailed test plan.

3. The heterogeneous unmanned platform test method based on the combination of virtual and real according to claim 1 is characterized in that: The step 4 is specifically as follows: 4-1: Select the geographic environment file and set relevant elements according to the test requirements; 4-2: Expand the virtual scene according to the test requirements. The environmental simulation model mainly constructs the geographical environment model, meteorological environment model and sound and light special effects model; 4-3: Set the number of virtual entities, deployment locations, complete the task actions of virtual entities and edit the conditions for automatic end of training; 4-4: Provide computational analysis, data support and interactive interface, as well as visual effects.

4. The heterogeneous unmanned platform test method based on the combination of virtual and real according to claim 1 is characterized in that: The step 6 is specifically as follows: 6-1: Time reference calibration: Calibrate the time server to ensure that the time of all measuring devices and the devices under test are in the same world time to meet the needs of high-precision testing; 6-2: Spatial benchmark calibration: Through the joint measurement and calculation of the benchmark points in the spatial layout, the coordinate system required for the test is obtained, including plane benchmark control joint measurement, elevation benchmark control joint measurement, three-dimensional benchmark control measurement, and acquisition and determination of geodetic coordinates; 6-3: Coordinate integration: The time and coordinate integration of the installed equipment, test instruments and basic conditions are completed through the time and space synchronization calibration system.

5. The heterogeneous unmanned platform test method based on virtual-real combination according to claim 1 is characterized in that: The step 11 is specifically as follows: 11-1: Establish a hierarchical structure model to decompose the evaluation objectives into multiple layers of indicators; 11-2: Determine the hierarchical structure, and then determine the affiliation of elements between upper and lower layers; Starting from the second layer, for the factors of the same layer belonging to each factor of the previous layer, pairwise comparisons are made to construct a comparison matrix until the bottom layer.

6. The heterogeneous unmanned platform test method based on the combination of virtual and real according to claim 5 is characterized in that: The comparison matrix B is: Among them, B1, B2, ..., B n The next layer element dominated by the previous layer element A; b ij is the i-th element B i With the jth element B j The importance ratio between them is; n is the number of indicators in this layer affected by the element A in the previous layer.

7. The method for testing heterogeneous unmanned platforms based on the combination of virtual and real according to claim 6 is characterized in that: The relative weight calculation algorithms under a single criterion include the sum method, characteristic root method, and root method.

8. The method for testing heterogeneous unmanned platforms based on the combination of virtual and real according to claim 7 is characterized in that: The weight calculation of the sum method is as follows: (1) Normalized judgment matrix B: in, is the element in the i-th row and j-th column of the normalized judgment matrix; (2) Total by row: in, is the unnormalized weight coefficient after summing up by row; (3) After normalization, the weight coefficient can be obtained: Among them, w i is the normalized weight coefficient; (4) Calculate the maximum eigenvalue λ max : w=(w1,…,wn) T ; Among them, λ max is the largest characteristic root; w is the weight column vector composed of weight coefficients; (Bw) i is the i-th item in the column vector after the comparison matrix B is multiplied by the weight column vector w; (5) Calculate the consistency index CI: C.I=(λ max -n) / (n-1); (6) Find the average consistency index RI of the corresponding n and calculate the consistency ratio CR: Average random consistency index CR = CI / RI; When CR is less than 0.1, the consistency of the judgment matrix is ​​acceptable; when CR is greater than or equal to 0.1, the consistency of the judgment matrix does not meet the requirements, and the judgment matrix is ​​modified.

9. The heterogeneous unmanned platform test method based on the combination of virtual and real according to claim 8 is characterized in that: When the weights are synthesized from bottom to top, the synthetic weights of the elements in each layer for the overall target are obtained.

10. A heterogeneous unmanned platform test system based on the combination of virtual and real, characterized in that: For implementing the heterogeneous unmanned platform test method based on virtual-real combination as described in any one of claims 1 to 9, the heterogeneous unmanned platform test system includes the following modules: Scenario application module: builds a real-life test environment for typical application scenarios such as cities, blocks, buildings, and underground; Functional service module: By taking the real-life test environment as the guide, a digital parallel environment is constructed to provide virtual-real combination simulation, algorithm training and verification, data management and control evaluation, data interconnection and interoperability, and to connect with the infrastructure and general platform construction of the general basic modules downward to form a guidance and constraint relationship in professional fields; General basic module: It realizes the virtual-reality simulation function by supporting the virtual-reality simulation test verification of multiple heterogeneous unmanned platforms at the system level.

Images