Unmanned system autonomy grade evaluation subject generation method based on complexity analysis

Through complexity analysis and hierarchical analysis methods, unmanned system autonomy level assessment subjects were generated, which solved the problem of missing subject generation methods, improved the accuracy and efficiency of evaluation, and adapted to the development and changes of unmanned systems.

CN119067312AInactive Publication Date: 2024-12-03THE QUARTERMASTER RES INST OF THE GENERAL LOGISTICS DEPT OF THE CPLA
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411205926.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-30
Publication Date
2024-12-03
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The lack of a method for generating subjects for the unmanned system autonomy level assessment results in inaccurate and inefficient assessment.

Method used

Based on complexity analysis, through the determination of autonomy level, the division of environmental complexity and task complexity, evaluation scenarios and questions are set, and combined with the hierarchical analysis method, corresponding evaluation subjects are generated to realize the mapping of autonomy level evaluation subjects and levels.

Benefits of technology

It improves the accuracy and efficiency of the level of autonomy of unmanned systems to ensure that the assessment subject truly reflects the level of autonomy of unmanned systems and adapts to their development and changes.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119067312A_ABST
    Figure CN119067312A_ABST
Patent Text Reader

Abstract

The invention relates to the field of unmanned system evaluation, in particular to an unmanned system autonomy grade evaluation subject generation method based on complexity analysis. According to the invention, based on the autonomy level of the unmanned system, on the basis of environmental complexity and task complexity level division, through autonomy level evaluation scene setting, evaluation question setting and evaluation scene and evaluation question combination, mapping of autonomy level evaluation subjects and levels of the unmanned system is realized; and completing subject generation of autonomous level evaluation of the unmanned system. The invention provides an unmanned system autonomy grade evaluation subject generation method based on complexity analysis, and mainly solves the problem of lack of subject setting methods and means for unmanned system autonomy grade evaluation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of unmanned system assessment, and in particular to a method for generating unmanned system autonomy level assessment subjects based on complexity analysis. Background Art

[0002] With the rapid development of artificial intelligence, machine learning, sensors, communications and other technologies, unmanned systems have been widely used in military, civilian and commercial fields. The current limitations of technological development have led to different levels of autonomy in unmanned systems, and the level of autonomy of unmanned systems directly affects their application effectiveness. Therefore, the evaluation of the autonomy level of unmanned systems is of great significance to the application and development of unmanned systems.

[0003] In order to improve the accuracy, efficiency and credibility of the autonomy level assessment of unmanned systems, it is critical to set the autonomy level assessment subjects scientifically and reasonably. At present, there are many studies on the autonomy level assessment methods of unmanned systems at home and abroad, but there are few studies on the generation methods of the autonomy level assessment subjects of unmanned systems. Therefore, the generation of the autonomy level assessment subjects of unmanned systems currently faces the problem of lack of methods, and it is urgent to carry out research on the generation methods of assessment subjects. Summary of the invention

[0004] The purpose of the present invention is to provide a method for generating subjects for the evaluation of the autonomy level of unmanned systems based on complexity analysis, which mainly solves the problem of the lack of subject setting methods and means for the evaluation of the autonomy level of unmanned systems. This method is based on the autonomy level of unmanned systems, based on the level division of environmental complexity and task complexity, and through the setting of autonomy level evaluation scenarios, the setting of evaluation questions, and the combination of evaluation scenarios and evaluation questions, the mapping of subjects and levels of the evaluation of the autonomy level of unmanned systems is achieved, and the generation of subjects for the evaluation of the autonomy level of unmanned systems is completed.

[0005] The specific technical solution adopted by the present invention is as follows:

[0006] The present invention provides a method for generating subjects for autonomous level assessment of unmanned systems based on complexity analysis, comprising the following steps:

[0007] Step 1: Determine the level of autonomy

[0008] Autonomy refers to the level of independence that an unmanned system demonstrates when performing tasks. According to the intelligence level assessment criteria, three levels of autonomy of unmanned systems are determined:

[0009] Machine Assistance (A1): The unmanned system is able to complete the set combat mission within the known fixed mission and environment range; Machine Assistance (A1) level unmanned systems have a low level of autonomy, and tasks beyond the scope are controlled or intervened by humans;

[0010] Human-machine collaboration (A2): The unmanned system can complete the set combat mission under known changing tasks or environments; the unmanned system of human-machine collaboration (A2) level has medium-level autonomy, and tasks beyond the scope are controlled or intervened by humans;

[0011] Machine-based (A3): Unmanned systems are able to complete set combat missions in unknown and changing missions or environments. Unmanned systems at the machine-based (A3) level have a high level of autonomy and can complete missions without human involvement.

[0012] Step 2: Complexity level classification

[0013] According to the standards for the classification of the levels of autonomy of unmanned systems, changes in the two dimensions of environmental complexity and task complexity are used as the basic elements of the subject setting;

[0014] The environmental complexity is divided into three levels: low, medium, and high, corresponding to the three types of known fixed, known changing, and unknown changing environments; the task complexity is also divided into three levels: low, medium, and high, corresponding to the three types of known fixed, known changing, and unknown changing tasks;

[0015] The next level of sub-indicators of environmental complexity factors include geographical, meteorological, hydrological, and electromagnetic environments and changes in indicators;

[0016] The next level of sub-indicators of the task complexity factor include task composition, task requirements, task conditions, task risks, and indicator changes;

[0017] The levels of environmental complexity and task complexity are mapped to each other through experts’ evaluation and scoring based on the data of the next-level sub-indicators of each factor;

[0018] The next level sub-indicators of environmental complexity factors and task complexity factors are used as bottom-level indicators or further decomposed into more level indicators;

[0019] According to the obtained next-level indicator data of environmental complexity and task complexity, the evaluation results of environmental complexity and task complexity are obtained through hierarchical analysis method; the details are as follows:

[0020] According to the comparison results of the importance of the same-level indicators belonging to the same upper-level indicators, the judgment matrices corresponding to the indicators at each level are determined; the judgment matrices are subjected to consistency test according to the maximum characteristic root and random consistency index of each judgment matrix to determine the test result; when the test result is within the preset range, the weight coefficient corresponding to the judgment matrix is ​​determined according to the judgment matrix; the bottom-level indicators are quantified into score values ​​within the range of 0 to 1 according to the variation metric, and the weight coefficients of the indicators at each level obtained and the corresponding indicator score values ​​are calculated to obtain the score values ​​of the upper-level indicators, and the environmental complexity and task complexity values ​​are obtained level by level;

[0021] If the test result is not within the preset range, the importance of each evaluation index corresponding to the judgment matrix is ​​adjusted until the test result is within the preset range, and then subsequent operations are performed according to the situation when the test result is within the preset range;

[0022] The obtained values ​​of environment complexity and task complexity are divided into intervals corresponding to low, medium and high levels in the range of [0,1] to determine the levels of environment complexity and task complexity;

[0023] Since the complexity level classification is an assessment process that combines experience and objective facts, it is impossible to give a clear division boundary through direct and complete quantification. Therefore, in the process of setting the evaluation subjects, the environment and task complexity levels need to be constantly adjusted dynamically based on expert opinions and feedback from the execution of the evaluation subjects to adapt to the development level of the unmanned system and accurately reflect the level of autonomy of the unmanned system.

[0024] Step 3: Subject Setting

[0025] Step 301. Preprocessing

[0026] The pre-processing stage mainly includes mission analysis and intelligent enabling point extraction;

[0027] Mission analysis refers to analyzing the structural characteristics, combat performance, main applicable scenarios, and types of tasks that can be performed, the process of performing tasks, and the principles of information and data transmission during the task execution of the unmanned system to be evaluated before setting the evaluation subjects, so as to determine the scenarios and task scopes of the selected unmanned system subject settings;

[0028] The intelligent enabling point extraction is to analyze the intelligent algorithms or intelligent processing methods required for the tasks to be performed by the unmanned system to be evaluated, and determine the content and strategy of the subject setting of the selected unmanned system;

[0029] Step 302. Scene setup

[0030] According to the environmental complexity classification in step 2, the underlying indicators of environmental complexity are set to obtain evaluation scenarios of high, medium and low levels of environmental complexity for the selected unmanned system;

[0031] Step 303. Topic Setting

[0032] According to the task complexity classification in step 2, the underlying indicators of task complexity are set to obtain evaluation questions of high, medium and low levels of task complexity of the selected unmanned system;

[0033] Step 304: Combining scenes and topics

[0034] By combining the evaluation scenario with the evaluation topic, an evaluation subject of the autonomy level of the selected unmanned system is obtained; the complexity level of the evaluation subject is a higher level among the environmental complexity level and the task complexity level;

[0035] The low, medium and high complexity levels of the assessment subjects correspond to the A1, A2 and A3 levels of the unmanned system autonomy levels, completing the mapping of assessment subjects to the unmanned system autonomy levels.

[0036] Preferably, after step 304, step 305 is performed. Rationality check:

[0037] Check the logical relationship between the underlying indicators of environmental complexity and task complexity of question settings and scenario settings; if there is a logical conflict, reset the underlying indicators; if there is no logical conflict, complete the subject generation.

[0038] Furthermore, the rationality test includes the consistency of upper and lower indicators and the mutual exclusivity of indicators at the same level; the consistency of upper and lower elements is decomposed step by step from the first-level indicators to the bottom-level indicators, and whether there are inconsistencies in the decomposition and assignment process; the mutual exclusivity of indicators at the same level refers to whether there are any contradictions between the indicators at the same level.

[0039] Preferably, the intelligent enabling point extraction is performed by decomposing the unmanned system functions hierarchically until all the minimum functional units of the unmanned system that apply intelligent technology are decomposed, and the intelligent algorithm or intelligent processing method is determined according to each minimum functional unit of the unmanned system.

[0040] Preferably, the geographical environment includes cities, jungles, grasslands, oceans, straits, deserts, and polar regions; the meteorological environment includes wind, rain, thunder, lightning, fog, snow, clouds, and light; the hydrological environment includes hydrological type, water level, flow rate, flow, and water temperature; and the electromagnetic environment includes electromagnetic type, electromagnetic intensity, and electromagnetic spectrum.

[0041] Preferably, the mission composition includes identification, positioning, reconnaissance, strike, and mapping; the mission requirements include the execution time, concealment, and degree of intervention of the mission; the mission conditions include obstacle conditions, navigation conditions, and electromagnetic interference conditions; and the mission risks include whether the unmanned system will be interfered with and counterattacked by the opponent during execution.

[0042] Furthermore, the task composition is set based on the unmanned system mission and intelligent enabling points.

[0043] Furthermore, the task conditions and task requirements are set based on expert experience.

[0044] Preferably, in step three, according to the unmanned system type, scene settings and topic settings corresponding to the typical unmanned system type are set as a basic database.

[0045] Furthermore, the typical unmanned systems include unmanned aerial vehicles, unmanned vehicles, unmanned ships, and unmanned submarines.

[0046] In a second aspect, the present invention provides a computer-readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the method described in the first aspect is implemented.

[0047] In a third aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in the first aspect when executing the program.

[0048] Beneficial effects achieved by the present invention:

[0049] The present invention classifies the environmental complexity and the task complexity according to the autonomy level of the unmanned system and according to the known fixed, known changing and unknown changing conditions of the environment and the task; sets the evaluation scenarios and evaluation topics respectively by setting the known fixed, known changing and unknown changing conditions of the factors affecting the environmental and task complexity; and generates evaluation subjects with corresponding complexity to the autonomy level of the unmanned system through the combination of the evaluation scenarios and the evaluation topics. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 This is a schematic diagram of the process of Example 1 of the present invention. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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. Based on the embodiments of 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.

[0052] In order to make the above-mentioned objects, features and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be noted that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0053] Example 1

[0054] Step 1: Determine the level of autonomy

[0055] Autonomy refers to the level of independence that an unmanned system demonstrates when performing tasks. It is divided into three levels according to the intelligent evaluation criteria. The specific characteristics are described as follows:

[0056] Table 1 Levels of unmanned autonomy

[0057]

[0058] According to the typical characteristics of the autonomy level of unmanned systems, the autonomy level assessment needs to be carried out in an environment and task with a certain degree of complexity, and is divided into three situations: known fixed, known change, and unknown change. The autonomy levels in the three situations increase in turn. In the setting of autonomy level assessment subjects, the setting of different levels of autonomy level assessment subjects is completed by combining environment and task, based on environment setting and task setting.

[0059] Step 2: Complexity level classification

[0060] Complexity is divided into environmental complexity and task complexity, and is a key indicator for evaluating the autonomy level of unmanned systems. According to the degree of familiarity and the level of change, it is divided into three situations: known fixed, known change, and unknown change, corresponding to low, medium, and high complexity levels respectively. The key to setting the autonomy level assessment subject is to accurately set the environment and tasks of the corresponding complexity level to ensure that the assessment subject can more realistically assess the autonomy level of the unmanned system. The complexity level division is an assessment process that combines experience and objective facts. It is impossible to give a clear division boundary through direct and complete quantification. Therefore, in the process of setting the assessment subject, the environment and task complexity levels need to be constantly adjusted dynamically based on expert opinions and feedback from the execution of the assessment subjects to adapt to the development level of the unmanned system and accurately reflect the autonomy level of the unmanned system.

[0061] Step 201: Classification of environmental complexity levels. Environmental complexity factors include geographical, meteorological, hydrological, electromagnetic and other environments and their changes. In the course setting, reasonable factor selection and change setting are required to meet the environmental complexity requirements of the course setting.

[0062] Step 202: Classification of task complexity. Task complexity factors include task composition, task requirements, task conditions, task risks and other indicators and their changes. According to the mission and intelligent functions of the unmanned system, scientific task goals are set to meet the task complexity requirements set by the subject.

[0063] The details are as follows:

[0064] According to the comparison results of the importance of the same-level indicators under the same upper-level indicator, the judgment matrices corresponding to the indicators at each level are determined; the judgment matrices are subjected to consistency test according to the maximum characteristic root and random consistency index of each judgment matrix to determine the test result; when the test result is within the preset range, the weight coefficient corresponding to the judgment matrix is ​​determined according to the judgment matrix; the bottom-level indicators are quantified into score values ​​within the range of 0 to 1, and the weight coefficients of the indicators at each level obtained and the corresponding indicator score values ​​are calculated to obtain the score values ​​of the upper-level indicators, and the environmental complexity and task complexity values ​​are obtained level by level;

[0065] If the test result is not within the preset range, the importance of each evaluation index corresponding to the judgment matrix is ​​adjusted until the test result is within the preset range, and then subsequent operations are performed according to the situation when the test result is within the preset range;

[0066] The obtained values ​​of environment complexity and task complexity are divided into intervals corresponding to low, medium and high levels in the range of [0,1] to determine the levels of environment complexity and task complexity;

[0067] Step 3: Subject Setting

[0068] The key to setting subjects is to set reasonable scenarios (environments) and questions (tasks), accurately correspond to the complexity of the environment and tasks, and more realistically reflect the autonomy level of the unmanned system. In the setting of scenarios and questions, the complexity influencing factors are selected according to the level of complexity, the threshold of the influencing factor change is set, and various factors are scientifically matched and integrated, finally forming a more reasonable and scientific evaluation subject.

[0069] Step 301. Preprocessing

[0070] The pre-processing stage mainly includes mission analysis and intelligent empowerment point extraction. Mission analysis refers to the analysis of the structural characteristics, combat performance, main applicable scenarios, and types of tasks that can be performed, the process of performing tasks, and the principles of information and data transmission during the task execution of the unmanned system to be evaluated before setting the evaluation subjects, so as to determine the scenarios and task scope of the subject setting. Intelligent empowerment point extraction is to analyze which tasks of the unmanned system to be evaluated require the help of intelligence, at which link of the task process the "brain power" plays a role, and how it plays a role, etc., in order to determine the content and strategy of the subject setting.

[0071] The analysis of intelligent enabling points is the focus of preprocessing analysis. Unmanned systems are of various types, belong to different fields, and perform different tasks. Different unmanned systems have different intelligent enabling points, and the same unmanned system has different intelligent enabling points when performing different tasks. Intelligent enabling point extraction is decomposed from top to bottom until it is sufficiently detailed.

[0072] Step 302. Scene setup

[0073] According to the environmental complexity classification in step 2, the underlying indicators of environmental complexity are set to obtain the evaluation scenarios of high, medium and low levels of environmental complexity of the selected unmanned system; environmental complexity factors include geographical, meteorological, hydrological, electromagnetic and other environments and their changes. The geographical environment mainly includes cities, jungles, grasslands, oceans, straits, deserts, polar regions, etc., and its specific selection needs to be reasonably selected based on the mission and intelligent empowerment point analysis of the unmanned system being evaluated. The meteorological environment includes wind, rain, thunder, lightning, fog, snow, clouds, light, etc. The inclusion of meteorological environmental elements in the scene and their changes need to be set and adjusted according to the complexity requirements and expert experience. The hydrological and electromagnetic conditions also need to be set and adjusted according to the complexity requirements and expert experience.

[0074] Step 303. Topic Setting

[0075] According to the task complexity classification in step 2, the underlying indicators of task complexity are set to obtain the evaluation questions of high, medium and low levels of task complexity of the selected unmanned system; the task complexity factors include task composition, task requirements, task conditions, task risks and other indicators and their changes. The task composition is mainly set based on the mission and intelligent enabling points of the unmanned system, including identification, positioning, reconnaissance, strike, mapping, etc. The task requirements need to be set based on expert experience and objective requirements, including the execution time, concealment, and degree of intervention of the task. The task conditions are set based on expert experience and objective requirements, including obstacles, navigation, electromagnetic interference, etc. The task risk is mainly based on the perspective of confrontation, considering whether the unmanned system will be interfered with and counterattacked by the other party during execution. It needs to be set according to expert experience and objective requirements. In the process of subject setting, the task risk is difficult to be set and reflected. Through a certain degree of task condition setting, the level of task risk is reflected.

[0076] Step 304: Combining scenes and topics

[0077] By combining the evaluation scenario with the evaluation topic, the autonomy level evaluation subject for a certain type of unmanned system is obtained. The complexity level of the evaluation subject is the higher level among the complexity levels of the evaluation scenario (environment) and the evaluation topic (task). The specific values ​​are shown in Table 2.

[0078] Table 2 Evaluation subject complexity value table

[0079]

[0080] The low, medium and high levels of complexity of the assessment subjects correspond to the three levels of A1, A2 and A3 in the autonomy level of the unmanned system, completing the mapping of the assessment subjects to the autonomy level of the unmanned system. In the subject setting, how to use expert experience and objective requirements will affect the objectivity of the assessment subjects to a certain extent. It is necessary to continuously adjust and optimize the assessment subjects through feedback from the use of the assessment subjects to more accurately and truly reflect the autonomy level of the unmanned system.

[0081] Step 305. Reasonableness test

[0082] The setting of questions and scenarios must follow their own rules and principles, and there must be no mutual exclusion or contradiction. The scenarios and questions are also set and combined by various environmental complexity factors and task complexity factors, and the rationality of the combination also needs to be tested. If the combination is reasonable, it can be determined as an assessment subject. If there are conflicts and contradictions, it should be modified and improved until the entire subject is reasonably perfected.

[0083] The rationality test mainly focuses on the consistency of upper and lower indicators and the mutual exclusivity of indicators at the same level. The consistency of upper and lower factors refers to whether there are inconsistencies in the process of decomposition and assignment from the first-level indicators to the bottom-level indicators; the mutual exclusivity of indicators at the same level refers to whether there are contradictions among the indicators at the same level.

Claims

1. A method for generating subjects for autonomous level assessment of unmanned systems based on complexity analysis, characterized in that: The steps include: Step 1: Determine the level of autonomy Autonomy refers to the level of independence that an unmanned system demonstrates when performing tasks. According to the intelligence level assessment criteria, three levels of autonomy of unmanned systems are determined: Machine Assistance (A1): The unmanned system is able to complete the set combat mission within the known fixed mission and environment range; Machine Assistance (A1) level unmanned systems have a low level of autonomy, and tasks beyond the scope are controlled or intervened by humans; Human-machine collaboration (A2): The unmanned system can complete the set combat mission under known changing tasks or environments; the unmanned system of human-machine collaboration (A2) level has medium-level autonomy, and tasks beyond the scope are controlled or intervened by humans; Machine-based (A3): Unmanned systems are able to complete set combat missions in unknown and changing missions or environments. Unmanned systems at the machine-based (A3) level have a high level of autonomy and can complete missions without human involvement. Step 2: Complexity level classification According to the standards for the classification of the levels of autonomy of unmanned systems, changes in the two dimensions of environmental complexity and task complexity are used as the basic elements of the subject setting; The environmental complexity is divided into three levels: low, medium, and high, corresponding to the three types of environments: known fixed, known changing, and unknown changing; the task complexity is also divided into three levels: low, medium, and high, corresponding to the three types of tasks: known fixed, known changing, and unknown changing; The next level of sub-indicators of environmental complexity factors include geographical, meteorological, hydrological, and electromagnetic environments and changes in indicators; The next level of sub-indicators of the task complexity factor include task composition, task requirements, task conditions, task risks, and indicator changes; The levels of environmental complexity and task complexity are mapped by experts based on the data evaluation and scoring of the next-level sub-indicators of each factor; the next-level sub-indicators of environmental complexity factors and task complexity factors are used as bottom-level indicators or further decomposed into more hierarchical indicators; According to the obtained next-level indicator data of environmental complexity and task complexity, the evaluation results of environmental complexity and task complexity are obtained through hierarchical analysis method; the details are as follows: Determine the judgment matrices corresponding to the indicators at each level according to the comparison results of the importance of the indicators at the same level belonging to the same upper level; perform consistency test on the judgment matrix according to the maximum characteristic root and random consistency index of each judgment matrix to determine the test result; When the inspection result is within a preset range, determining a weight coefficient corresponding to the judgment matrix according to the judgment matrix; The lowest level indicators are quantified into a score value between 0 and 1 according to the change quantity, and the weight coefficients of the indicators at each level are calculated with the corresponding indicator score values ​​to obtain the score values ​​of the previous level indicators, and the values ​​of the environment complexity and task complexity are obtained level by level; If the test result is not within the preset range, the importance of each evaluation index corresponding to the judgment matrix is ​​adjusted until the test result is within the preset range, and then subsequent operations are performed according to the situation when the test result is within the preset range; The obtained values ​​of environment complexity and task complexity are divided into intervals corresponding to low, medium and high levels in the range of [0,1] to determine the levels of environment complexity and task complexity; Step 3: Subject Setting Step 301. Preprocessing The pre-processing stage mainly includes mission analysis and intelligent enabling point extraction; Mission analysis refers to analyzing the structural characteristics, combat performance, main applicable scenarios, and types of tasks that can be performed, the process of performing tasks, and the principles of information and data transmission during the process of performing tasks of the unmanned system to be evaluated before setting the evaluation subjects, so as to determine the scenarios and task scopes of the selected unmanned system subject settings; The intelligent enabling point extraction is to analyze the intelligent algorithms or intelligent processing methods required for the tasks to be performed by the unmanned system to be evaluated, and determine the content and strategy of the subject setting of the selected unmanned system; Step 302. Scene setup According to the environmental complexity classification in step 2, the underlying indicators of environmental complexity are set to obtain evaluation scenarios of high, medium and low levels of environmental complexity for the selected unmanned system; Step 303. Topic Setting According to the task complexity classification in step 2, the underlying indicators of task complexity are set to obtain evaluation questions of high, medium and low levels of task complexity of the selected unmanned system; Step 304: Combining scenes and topics By combining the evaluation scenario with the evaluation topic, an evaluation subject of the autonomy level of the selected unmanned system is obtained; the complexity level of the evaluation subject is a higher level among the environmental complexity level and the task complexity level; The low, medium and high complexity levels of the assessment subjects correspond to the A1, A2 and A3 levels of the unmanned system autonomy levels, completing the mapping of assessment subjects to the unmanned system autonomy levels.

2. The method for generating subjects for unmanned system autonomy level assessment based on complexity analysis according to claim 1 is characterized in that: After step 304, proceed to step 305. Rationality check: Check the logical relationship between the underlying indicators of environmental complexity and task complexity of question settings and scenario settings; if there is a logical conflict, reset the underlying indicators; if there is no logical conflict, complete the subject generation.

3. The method for generating subjects for unmanned system autonomy level assessment based on complexity analysis according to claim 2 is characterized in that: The rationality test includes the consistency of upper and lower indicators and the mutual exclusivity of indicators at the same level; the consistency of upper and lower factors is decomposed step by step from the first-level indicators to the bottom-level indicators, and whether there are inconsistencies in the decomposition and assignment process; the mutual exclusivity of indicators at the same level refers to whether there are any contradictions between the indicators at the same level.

4. The method for generating subjects for unmanned system autonomy level assessment based on complexity analysis according to claim 1, characterized in that: The intelligent enabling point extraction is performed by decomposing the unmanned system functions layer by layer until all the minimum functional units of the unmanned system that apply intelligent technology are decomposed, and the intelligent algorithm or intelligent processing method is determined according to the minimum functional units of each unmanned system.

5. The method for generating subjects for unmanned system autonomy level assessment based on complexity analysis according to claim 1, characterized in that: The geographical environment includes cities, jungles, grasslands, oceans, straits, deserts, and polar regions; the meteorological environment includes wind, rain, thunder, lightning, fog, snow, clouds, and light; the hydrological environment includes hydrological type, water level, flow rate, flow, and water temperature; the electromagnetic environment includes electromagnetic type, electromagnetic intensity, and electromagnetic spectrum.

6. The method for generating subjects for evaluating the autonomy level of unmanned systems based on complexity analysis according to claim 1 is characterized in that: The mission structure includes identification, positioning, reconnaissance, strike, and mapping; the mission requirements include the mission execution time, concealment, and degree of intervention; the mission conditions include obstacles, navigation, and electromagnetic interference; the mission risks include whether the unmanned system will be interfered with and counterattacked by the opponent during execution.

7. The method for generating subjects for evaluating the autonomy level of unmanned systems based on complexity analysis according to claim 6 is characterized in that: The task structure is set based on the mission and intelligent enabling points of the unmanned system; the task conditions and task requirements are set based on expert experience.

8. The method for generating subjects for unmanned system autonomy level assessment based on complexity analysis according to claim 1 is characterized in that: In the step three, according to the unmanned system type, the scene settings and topic settings corresponding to the typical unmanned system type are set as the basic database.

9. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

10. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method according to any one of claims 1 to 8 is implemented.