Virtual-real combination test method for consistency migration of intelligent algorithm of unmanned equipment

Abstract

The invention provides a virtual-real combination test method for consistency migration of an intelligent algorithm of unmanned equipment. The method comprises the following steps: designing a combat scene; an unmanned aerial vehicle target recognition algorithm is accessed to the deduction platform combat scene; recording intelligent algorithm input and output and related index test data under the full-digital test; carrying out lightweight processing on the algorithm model, and deploying the algorithm model to an unmanned system information processing computer; starting a deduction scene, carrying out virtual-real combination test verification based on the deduction platform digital scene and an intelligent algorithm model operated by an unmanned system information processing computer, and recording intelligent algorithm input and output and related index test data under the virtual-real combination test; and performing test process data comparison and analysis to complete intelligent algorithm verification and evaluation. According to the technical scheme, the technical problems that in the prior art, a physical prototype test is long in period and high in cost, and the functional physical boundary and performance of a mathematical test are difficult to effectively cover are solved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent unmanned equipment test technology, and in particular to a virtual-reality combined test method for consistency migration of intelligent algorithms for unmanned equipment. Background Art

[0002] Intelligent unmanned equipment is unmanned equipment (including drones, unmanned vehicles, satellites, missiles, etc.) that has the ability to perceive the environment, tasks, and its own status and network communication capabilities, and can make limited behavioral decisions based on the environment, tasks, and its own capabilities.

[0003] Intelligent unmanned equipment is driven by intelligent technology. The effectiveness of intelligent technology needs to be trained, optimized, and verified in complex and uncertain scenarios. On the one hand, relying on physical prototypes for internal and external field testing is difficult to fully cover the dynamics and adversarial nature of the test environment. The test cycle is long and the cost is high. In addition, it is difficult to build a verification environment for large-scale collaborative use. The performance of intelligent missiles can only be tested and verified under limited typical conditions. On the other hand, relying on mathematical experiments to build complex combat scenarios in the digital domain, although the dynamics are enriched, most of them are at the functional level, and it is difficult to effectively cover the functional physical boundaries and performance of unmanned equipment. Summary of the invention

[0004] The present invention provides a virtual-reality combined testing method for the consistency migration of intelligent algorithms for unmanned equipment, which can solve the technical problems in the prior art that the testing cycle of physical prototypes is long and costly, and the functional physical boundaries and performance of mathematical experiments are difficult to effectively cover.

[0005] According to one aspect of the present invention, a virtual-reality combined test method for consistency migration of intelligent algorithms of unmanned equipment is provided, and the virtual-reality combined test method for consistency migration of intelligent algorithms of unmanned equipment includes: step 1, completing combat scenario design based on a deduction platform; step 2, connecting the drone target recognition algorithm to the combat scenario of the deduction platform, and publishing the algorithm model as an inference service in the form of a Restful interface; step 3, starting the deduction scenario, and recording the intelligent algorithm input and output and related index test data under the full digital test; step 4, after lightweight processing of the algorithm model, deploying it to the unmanned system information processing computer, the information processing computer is connected to the game deduction platform through a field bus, and is released to the outside in the form of an embedded microservice; step 5, starting the deduction scenario, based on the digital scenario of the deduction platform and the intelligent algorithm model running on the information processing computer of the unmanned system, conducting virtual-reality combined test verification, and recording the intelligent algorithm input and output and related index test data under the virtual-reality combined test; step 6, comparing and analyzing the test data of the intelligent algorithm input and output and related indexes in step 3 with the test data of the intelligent algorithm input and output and related indexes under the virtual-reality combined test in step 5, and completing the intelligent algorithm verification and evaluation.

[0006] Furthermore, in step one, the combat scenario designed based on the simulation platform includes combat programming, position deployment and scenario variable design of the red and blue army equipment models; the red side equipment model includes drones, unmanned vehicles and command vehicles, and the blue side equipment model includes interceptor missiles, launch vehicles and launch sites.

[0007] Furthermore, in step 2, the drone target recognition algorithm model is packaged in docker form together with the environment required for model operation, and the input is a tensor of uniform size (Batchsize, Channel, High, Width), namely batch size, number of channels, image height and image width; the model prediction result output contains two formats of data, which are uniformly saved in a folder named "results", one folder is "text_results", which represents the prediction result file, and the other folder is "visualization" which is the visualization picture of the prediction result.

[0008] Furthermore, in step three, the recorded data is the input of the algorithm model, namely the batch size, number of channels, image height and image width; the output of the algorithm model, namely the prediction result file and the visualization file of the prediction result; the experimental indicator data is mAP (mean average precision).

[0009] Furthermore, in step four, the algorithm model is compressed and quantized to achieve lightweight processing.

[0010] According to another aspect of the present invention, a virtual-reality combined test system for consistency migration of intelligent algorithms for unmanned equipment is provided. The virtual-reality combined test system for consistency migration of intelligent algorithms for unmanned equipment uses the virtual-reality combined test method for consistency migration of intelligent algorithms for unmanned equipment as described above to perform a virtual-reality combined test for consistency migration of intelligent algorithms for unmanned equipment.

[0011] Furthermore, the virtual-reality combined test system for the consistency migration of intelligent algorithms for unmanned equipment includes: a combat scenario design module, which is used to complete the combat scenario design based on the deduction platform; an algorithm access module, which is used to access the drone target recognition algorithm to the deduction platform combat scenario, and publish the algorithm model as an inference service in the form of a Restful interface; a data acquisition module under full digital tests, which is used to start the deduction scenario and record the input and output of the intelligent algorithm and related indicator test data under the full digital test; a lightweight processing module, which is used to perform lightweight processing on the algorithm model and deploy it to the unmanned system information processing computer Computer, the information processing computer is connected to the game simulation platform through the field bus and released to the outside in the form of embedded microservices; the data acquisition module under the virtual-reality combination test, the data acquisition module under the virtual-reality combination test is used to start the simulation scenario, based on the digital scenario of the simulation platform and the intelligent algorithm model run by the information processing computer of the unmanned system, to carry out virtual-reality combination test verification, and record the input and output of the intelligent algorithm and related indicator test data under the virtual-reality combination test; the intelligent algorithm verification and evaluation module, the intelligent algorithm verification and evaluation module is used to compare and analyze the test data of the intelligent algorithm input and output and related indicators with the test data of the intelligent algorithm input and output and related indicators under the virtual-reality combination test, and complete the intelligent algorithm verification and evaluation.

[0012] According to another aspect of the present invention, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the virtual-reality combined test method for consistency migration of intelligent algorithms for unmanned equipment as described above.

[0013] According to another aspect of the present invention, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the steps of the virtual-reality combined test method for consistency migration of intelligent algorithms for unmanned equipment as described above are implemented.

[0014] The technical solution of the present invention provides a virtual-real verification method for intelligent algorithm consistency migration based on a deduction platform. The method combines digital environment testing with semi-real environment testing, and verifies the migration from the virtual-real digital domain environment to the physical environment. On the basis of the verification of the intelligent algorithm model in the digital environment, the virtual-real environment intelligent algorithm consistency comparison test is achieved through lightweight processing and drop-board deployment. Compared with the traditional test verification that only relies on physical prototypes, this method has a short test cycle, low cost, and strong adaptability to complex scenarios; compared with the test method that relies solely on digital domain verification, the algorithm deployment and operation are consistent with the physical equipment, effectively covering the functional physical boundaries of unmanned equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The included drawings are used to provide a further understanding of the embodiments of the present invention, which constitute a part of the specification, are used to illustrate the embodiments of the present invention, and together with the text description, explain the principles of the present invention. Obviously, the drawings in the following description are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0016] Figure 1 A flowchart of a virtual-real combination test method for unmanned equipment intelligent algorithm consistency migration provided according to a specific embodiment of the present invention is shown;

[0017] Figure 2 A structural schematic diagram of a virtual-reality combined test system for unmanned equipment intelligent algorithm consistency migration provided according to a specific embodiment of the present invention is shown. DETAILED DESCRIPTION

[0018] It should be noted that, in the absence of conflict, the embodiments in this application and the features in the embodiments can be combined with each other. 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, not all of the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present invention and its application or use. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0020] Unless otherwise specifically stated, the relative arrangement of the parts and steps described in these embodiments, numerical expressions and numerical values ​​do not limit the scope of the present invention. At the same time, it should be understood that, for ease of description, the sizes of the various parts shown in the accompanying drawings are not drawn according to the actual proportional relationship. The technology, method and equipment known to ordinary technicians in the relevant field may not be discussed in detail, but in appropriate cases, the technology, method and equipment should be regarded as a part of the authorization specification. In all examples shown and discussed here, any specific value should be interpreted as being merely exemplary, rather than as a limitation. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once a certain item is defined in an accompanying drawing, it does not need to be further discussed in subsequent drawings.

[0021] like Figure 1 As shown, according to a specific embodiment of the present invention, a virtual-reality combined test method for consistency migration of intelligent algorithms for unmanned equipment is provided, and the virtual-reality combined test method for consistency migration of intelligent algorithms for unmanned equipment includes: step 1, completing the combat scenario design based on the deduction platform; step 2, connecting the drone target recognition algorithm to the deduction platform combat scenario, and publishing the algorithm model as an inference service in the form of a Restful interface; step 3, starting the deduction scenario, and recording the intelligent algorithm input and output and related index test data under the full digital test; step 4, after lightweight processing of the algorithm model, deploying it to the unmanned system information processing computer, the information processing computer is connected to the game deduction platform through the field bus, and is released to the outside in the form of an embedded microservice; step 5, starting the deduction scenario, based on the deduction platform digital scenario and the intelligent algorithm model running on the unmanned system information processing computer, conducting virtual-reality combined test verification, and recording the intelligent algorithm input and output and related index test data under the virtual-reality combined test; step 6, comparing and analyzing the test data of the intelligent algorithm input and output and related indexes in step 3 with the test data of the intelligent algorithm input and output and related indexes under the virtual-reality combined test in step 5, and completing the intelligent algorithm verification and evaluation.

[0022] By applying this configuration method, a virtual-real combination verification method for intelligent algorithm consistency migration based on a deduction platform is provided. The method uses the deduction platform as the scenario basis, and the two states of intelligent algorithm cloud deployment and terminal deployment as the test objects; a method for intelligent algorithm consistency migration verification is proposed, which is application-oriented and focuses on verifying its application capabilities and consistency verification capabilities for physical devices; a virtual-real algorithm comparison verification method is proposed, which focuses on functional consistency in the digital environment, and the physical domain focuses on the real-time performance of intelligent algorithms deployed on physical devices on the basis of focusing on functional consistency; an algorithm access method for virtual-real combination intelligent algorithm testing and verification is proposed, which takes information processing computers as objects and service publishing as an interface form. Compared with the prior art, the virtual-real combination test method for intelligent algorithm consistency migration for unmanned equipment provided by the present invention combines digital environment testing with semi-real environment testing, and verifies through the migration of virtual-real digital domain environment to physical environment. On the basis of verifying the intelligent algorithm model in the digital environment, it is lightweight processed and deployed on the board to realize the virtual-real environment intelligent algorithm consistency comparison test. Compared with traditional testing and verification that only relies on physical prototypes, this method has a short test cycle, low cost, and strong adaptability to complex scenarios; compared with the testing method that relies solely on digital domain verification, the algorithm deployment and operation are consistent with the physical equipment, effectively covering the functional physical boundaries of unmanned equipment.

[0023] Specifically, in step one, the combat scenario designed based on the simulation platform includes the combat programming, position deployment and scenario variable design of the red and blue army equipment models; the red side equipment model includes drones, unmanned vehicles and command vehicles, and the blue side equipment model includes interceptor missiles, launch vehicles and launch sites.

[0024] In step 2, the drone target recognition algorithm model is packaged in docker together with the environment required for model operation, and the input is a tensor of uniform size (Batchsize, Channel, High, Width), namely batch size, number of channels, image height and image width; the model prediction result output contains two formats of data, which are uniformly saved in a folder named "results", one folder is "text_results", which represents the prediction result file, and the other folder is "visualization" which is the visualization picture of the prediction result.

[0025] In step 3, the recorded data is the input of the algorithm model, i.e., batch size, number of channels, image height and image width; the output of the algorithm model is the prediction result file and the visualization file of the prediction result; the test indicator data is mAP (mean average precision). In step 4, the algorithm model is compressed and quantized for lightweight processing.

[0026] According to another aspect of the present invention, a virtual-real combination test system for unmanned equipment intelligent algorithm consistency migration is provided. Figure 2 As shown, the virtual-reality combined test system for the consistency migration of intelligent algorithms for unmanned equipment uses the virtual-reality combined test method for the consistency migration of intelligent algorithms for unmanned equipment as described above to perform a virtual-reality combined test for the consistency migration of intelligent algorithms for unmanned equipment.

[0027] This configuration method provides a virtual-real combination verification system for intelligent algorithm consistency migration based on a deduction platform. The system combines digital environment testing with semi-real environment testing, and verifies the migration from the virtual-real digital domain environment to the physical environment. On the basis of the verification of the intelligent algorithm model in the digital environment, it is lightweight processed and deployed on the board to achieve virtual-real environment intelligent algorithm consistency comparison test. Compared with the traditional test verification that only relies on physical prototypes, this system has a short test cycle, low cost, and strong adaptability to complex scenarios; compared with the test method that relies solely on digital domain verification, the algorithm deployment and operation are consistent with the physical equipment, effectively covering the functional physical boundaries of unmanned equipment.

[0028] Furthermore, in the present invention, the virtual-reality combined test system for the consistency migration of intelligent algorithms for unmanned equipment includes a combat scenario design module, an algorithm access module, a data acquisition module under a full digital test, a lightweight processing module, a data acquisition module under a virtual-reality combined test, and an intelligent algorithm verification and evaluation module. The combat scenario design module is used to complete the combat scenario design based on the deduction platform; the algorithm access module is used to access the drone target recognition algorithm to the combat scenario of the deduction platform, and publish the algorithm model as an inference service in the form of a Restful interface; the data acquisition module under the full digital test is used to start the deduction scenario, record the input and output of the intelligent algorithm under the full digital test and related indicator test data; the lightweight processing module is used to After lightweight processing of the algorithm model, it is deployed to the unmanned system information processing computer. The information processing computer is connected to the game simulation platform through the field bus and released to the outside in the form of embedded microservices; the data acquisition module under the virtual-reality combination test is used to start the simulation scenario, and carry out virtual-reality combination test verification based on the digital scenario of the simulation platform and the intelligent algorithm model run by the unmanned system information processing computer, and record the input and output of the intelligent algorithm and related indicator test data under the virtual-reality combination test; the intelligent algorithm verification and evaluation module is used to compare and analyze the test data of the intelligent algorithm input and output and related indicators with the test data of the intelligent algorithm input and output and related indicators under the virtual-reality combination test, and complete the intelligent algorithm verification and evaluation.

[0029] According to another aspect of the present invention, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the virtual-reality combined test method for consistency migration of intelligent algorithms for unmanned equipment as described above.

[0030] According to another aspect of the present invention, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the steps of the virtual-reality combined test method for consistency migration of intelligent algorithms for unmanned equipment as described above are implemented.

[0031] In order to further understand the present invention, the following Figure 1 and Figure 2 The virtual-reality combined test method for the consistency migration of intelligent algorithms for unmanned equipment provided by the present invention is described in detail.

[0032] like Figure 1 and Figure 2 As shown, according to a specific embodiment of the present invention, a virtual-reality combined test method for consistency migration of intelligent algorithms for unmanned equipment is provided, and the method specifically includes the following steps.

[0033] Step 1: Complete combat scenario design based on the simulation platform, including combat programming, position deployment, and scenario variable design of the red and blue army equipment models; the red side equipment models include drones, unmanned vehicles, command vehicles, etc., and the blue side equipment models include interceptor missiles, launch vehicles, launch sites, etc.

[0034] Step 2: Taking the drone target recognition algorithm as an example, connect the modified algorithm to the combat scenario of the simulation platform, publish the algorithm model as an inference service, and publish it in the form of a RESTful interface. RESTful is the interface standard in the cloud environment.

[0035] The model of the drone target recognition algorithm is packaged in docker together with the environment required for the model to run. The input is a tensor of uniform size (Batchsize, Channel, High, Width), that is, (batch size, number of channels, image height, image width); the model prediction result output contains two formats of data, which are saved in a folder named "results", one is "text_results", which represents the prediction result file, and the other folder is "visualization", which is the visualization picture of the prediction result.

[0036] Similarly, the above algorithm can be a task decision algorithm such as online trajectory planning in the red team UAV model.

[0037] Step 3: Start the simulation scenario and record the input and output of the intelligent algorithm and related indicator test data under the full digital test: Taking the drone target recognition algorithm as an example, the recorded data is the algorithm model input, that is, (batch size, number of channels, image height, image width); the output of the algorithm model, that is, the prediction result file and the visualization file of the prediction result; the test indicator data is mAP (mean average precision).

[0038] Step 4: After the intelligent algorithm is compressed, quantized and other lightweight processing (the purpose of lightweight processing is to consider the limited computing resources of embedded computers, it needs to be lightweight), it is deployed to the unmanned system information processing computer, which is connected to the game deduction platform through fieldbus and other forms, and released to the outside in the form of embedded microservices. RESTful is the interface standard in the cloud environment. The embedded microservice is in the form of a service-oriented interface, and the purpose is to keep the input and output of the two interfaces consistent.

[0039] Step 5: Start the simulation scenario, conduct virtual-real combination test verification based on the simulation platform digital scenario and the intelligent algorithm model run by the unmanned system information processing computer, and record the intelligent algorithm input and output and related indicator test data under the virtual-real combination test;

[0040] Step 6: Compare and analyze the virtual and real test data, analyze the data of the two test processes, and complete the intelligent algorithm verification and evaluation.

[0041] The present invention aims at a virtual-real combination testing method for intelligent algorithms of intelligent unmanned equipment. By combining digital environment testing with semi-real environment testing, the virtual-real digital environment is verified by migrating the virtual-real digital environment to the physical environment. On the basis of the verification of the intelligent algorithm model in the digital environment, the method is lightweight processed and deployed on the board to realize the consistency comparison test of the intelligent algorithm in the virtual-real environment.

[0042] The object of the present invention is to overcome the deficiencies in the prior art:

[0043] 1) A virtual-real verification method for consistency migration of intelligent algorithms based on a deduction platform is proposed. This method uses the deduction platform as the scenario basis and two types of intelligent algorithm cloud deployment and end-side deployment as test objects. Cloud deployment refers to deploying the trained model in the same cloud server environment as the training environment in the form of inference service; end-side deployment refers to the hardware board of the unmanned system.

[0044] 2) A smart algorithm consistency migration verification method is proposed. This method is application-oriented and focuses on verifying its application capabilities and consistency verification capabilities for physical devices.

[0045] 3) A virtual-real algorithm comparison and verification method is proposed. In the digital environment, the focus is on functional consistency. In the physical domain, the focus is on the real-time performance of the intelligent algorithm deployed on the physical device, based on the focus on functional consistency.

[0046] 4) An algorithm access method for testing and verifying virtual-real intelligent algorithms is proposed. This method takes information processing computers as the object and service publishing as the interface form.

[0047] The beneficial effects of the present invention compared with the prior art are as follows:

[0048] Compared with traditional testing and verification that only relies on physical prototypes, this method has a short test cycle, low cost, and strong adaptability to complex scenarios; compared with the testing method that relies solely on digital domain verification, the algorithm deployment and operation are consistent with the physical equipment, effectively covering the functional physical boundaries of unmanned equipment.

[0049] For ease of description, spatially relative terms such as "above", "above", "on the upper surface of", "above", etc. may be used here to describe the spatial positional relationship between a device or feature and other devices or features as shown in the figure. It should be understood that spatially relative terms are intended to include different orientations of the device in use or operation in addition to the orientation described in the figure. For example, if the device in the accompanying drawings is inverted, the device described as "above other devices or structures" or "above other devices or structures" will be positioned as "below other devices or structures" or "below other devices or structures". Thus, the exemplary term "above" can include both "above" and "below". The device can also be positioned in other different ways (rotated 90 degrees or in other orientations), and the spatially relative descriptions used here are interpreted accordingly.

[0050] In addition, it should be noted that the use of terms such as "first" and "second" to limit components is only for the convenience of distinguishing the corresponding components. If not otherwise stated, the above terms have no special meaning and therefore cannot be understood as limiting the scope of protection of the present invention.

[0051] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A virtual-real combination test method for consistency migration of intelligent algorithms for unmanned equipment, characterized in that: The virtual-real combination test method for the consistency migration of the intelligent algorithm of unmanned equipment includes: Step 1: Complete the combat scenario design based on the simulation platform; Step 2: Connect the drone target recognition algorithm to the combat scenario of the simulation platform, publish the algorithm model as an inference service in the form of a RESTful interface; Step 3: Start the simulation scenario and record the input and output of the intelligent algorithm and related index test data under the full digital test; Step 4: After lightweight processing of the algorithm model, it is deployed to the unmanned system information processing computer, which is connected to the game deduction platform through the field bus and released to the outside in the form of embedded microservices; Step 5: Start the simulation scenario, conduct virtual-real combination test verification based on the simulation platform digital scenario and the intelligent algorithm model run by the unmanned system information processing computer, and record the intelligent algorithm input and output and related indicator test data under the virtual-real combination test; Step six, compare and analyze the test data of the intelligent algorithm input and output and related indicators in step three with the test data of the intelligent algorithm input and output and related indicators in the virtual-reality combination test in step five to complete the intelligent algorithm verification and evaluation.

2. The virtual-real combination test method for consistency migration of intelligent algorithms for unmanned equipment according to claim 1 is characterized in that: In step one, the combat scenario designed based on the simulation platform includes combat programming, position deployment and scenario variable design of the red and blue army equipment models; the red side equipment model includes drones, unmanned vehicles and command vehicles, and the blue side equipment model includes interceptor missiles, launch vehicles and launch sites.

3. The virtual-real combination test method for consistency migration of intelligent algorithms for unmanned equipment according to claim 2 is characterized in that: In the step 2, the drone target recognition algorithm model is packaged in docker together with the environment required for model operation, and the input is a tensor of uniform size (Batchsize, Channel, High, Width), namely batch size, number of channels, image height and image width; the model prediction result output contains two formats of data, which are uniformly saved in a folder named "results", one folder is "text_results", representing the prediction result file, and the other folder is "visualization", which is a visualization picture of the prediction result.

4. The virtual-real combination test method for consistency migration of intelligent algorithms for unmanned equipment according to claim 3 is characterized in that: In step three, the recorded data is the input of the algorithm model, namely the batch size, number of channels, image height and image width; the output of the algorithm model is the prediction result file and the visualization file of the prediction result; the experimental indicator data is mAP (mean average precision).

5. The virtual-real combination test method for consistency migration of intelligent algorithms for unmanned equipment according to claim 4 is characterized in that: In the step 4, the algorithm model is compressed and quantized to achieve lightweight processing.

6. A virtual-real combination test system for unmanned equipment intelligent algorithm consistency migration, characterized in that: The virtual-reality combined test system for the consistency migration of intelligent algorithms for unmanned equipment uses the virtual-reality combined test method for the consistency migration of intelligent algorithms for unmanned equipment as described in any one of claims 1 to 5 to perform a virtual-reality combined test for the consistency migration of intelligent algorithms for unmanned equipment.

7. The virtual-real combination test system for unmanned equipment intelligent algorithm consistency migration according to claim 6 is characterized in that: The virtual-real combination test system for the consistency migration of intelligent algorithms for unmanned equipment includes: A combat scenario design module, which is used to complete combat scenario design based on the deduction platform; An algorithm access module, which is used to access the drone target recognition algorithm into the combat scenario of the simulation platform, and publish the algorithm model as an inference service in the form of a Restful interface; A data acquisition module under full digital test, which is used to start the simulation scenario and record the input and output of the intelligent algorithm and related index test data under full digital test; A lightweight processing module, which is used to perform lightweight processing on the algorithm model and then deploy it to the unmanned system information processing computer. The information processing computer is connected to the game deduction platform through a field bus and is released to the outside in the form of embedded microservices; A data acquisition module for a virtual-real combination test, which is used to start a simulation scenario, conduct virtual-real combination test verification based on the digital scenario of the simulation platform and the intelligent algorithm model run by the information processing computer of the unmanned system, and record the input and output of the intelligent algorithm and related indicator test data under the virtual-real combination test; The intelligent algorithm verification and evaluation module is used to compare and analyze the test data of the intelligent algorithm input and output and related indicators with the test data of the intelligent algorithm input and output and related indicators under the virtual-reality combination test, so as to complete the intelligent algorithm verification and evaluation.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: The processor executes the computer program to implement the steps of the virtual-reality combined test method for consistency migration of intelligent algorithms for unmanned equipment as described in any one of claims 1 to 5.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by the processor, the steps of the virtual-reality combined test method for consistency migration of intelligent algorithms for unmanned equipment as described in any one of claims 1 to 5 are implemented.

Images