Normative dependency logic driven flight test mission profile generation method and device

Abstract

The application provides a flight test task outline generation method and device based on normative dependency logic driving, which comprises the following steps: analyzing development general requirements and related normative files, and identifying normative entity information therein; constructing logical dependency relationships among entities to form a normative dependency logic graph; dividing normative entity information of different categories into corresponding pre-constructed resource libraries; establishing mapping relationships among different resource libraries; generating a logical sub-query expression from the normative dependency logic graph; searching for normative entity information pieces meeting the logical dependency conditions in the resource libraries to obtain candidate pieces and perform sorting; generating a preliminary flight test task outline according to the logical conditions, performing dependency chain integrity checking, and outputting a final structured and standardized flight test task outline; and the whole process of automatically generating a structured flight test task outline from development general requirements and standard norms can be realized.

Description

Technical Field

[0001] This invention belongs to the field of artificial intelligence and flight test mission planning technology, specifically relating to a method and apparatus for generating flight test mission outlines based on normative dependency logic. Background Technology

[0002] During the development of new aircraft models, the general development requirements and related normative documents typically contain numerous clauses, performance indicators, flight test subjects, and test points, and exhibit complex conditional dependencies. For example, some clauses stipulate that "if condition X is met, then conditions Y and Z must be met," and these logical dependency chains run throughout the entire normative document. Since normative documents usually cover multiple aircraft models, the indicator requirements for different models include both explicit numerical values ​​and ambiguous or unclear clauses. Therefore, during the mission planning process, it is necessary to digitize clauses without explicit indicators by combining historical data and existing resource libraries to generate operable parameter values.

[0003] The existing resource database contains capability indicators, standards and specifications, test flight subjects, test sites, potential risk items, and related flight parameters. However, there are complex mapping relationships between the various resource databases. For example, test flight subjects need to be associated with capability indicators, and test sites need to correspond to test flight subjects. At the same time, indicator items may be associated with potential risks or regulatory clauses. In this situation, ordinary text retrieval or semantic similarity-based retrieval enhancement systems often cannot accurately identify the logical dependencies between clauses, nor can they follow normative constraints when generating mission outlines. This can lead to incomplete coverage of the generated test flight mission outlines, an increase in the number of repeated test flights, and even an inability to guarantee the compliance and execution efficiency of the missions, thus posing potential risks to the reliability and safety of the test flight plan. Summary of the Invention

[0004] To address the problems of incomplete coverage, increased repetitive test flights, and even the inability to guarantee mission compliance and execution efficiency in existing technologies, which pose potential risks to the reliability and safety of flight test plans, this invention provides a method and apparatus for generating flight test mission outlines based on normative dependency logic. This method enables the automatic generation of structured flight test mission outlines from overall development requirements and standards. The technical solution is as follows: Firstly, a method for generating flight test mission outlines based on normative dependency logic is provided, the method comprising the following steps: Step 1: Analyze the overall development requirements and related normative documents to identify the normative entity information within them; Step 2: Abstract the identified clauses and indicator requirements into normative entities, and construct the logical dependencies between entities to form a normative dependency logic diagram; Step 3: Based on the type of normative entity, classify the information of different categories of normative entities into the corresponding pre-built resource repositories; Step 4: Establish mapping relationships between different resource repositories to ensure that various normative entity information entities and their dependencies can be accurately referenced in subsequent task processing; Step 5: Based on the target device model and task requirements, generate logical subquery expressions from the normative dependency logic graph to retrieve candidate resource fragments; Step 6: Retrieve canonical entity information fragments that meet the logical dependency conditions in the resource library to obtain candidate fragments, and sort the candidate fragments; Step 7: Based on the sorted candidate segments, generate a preliminary flight test mission outline according to logical conditions, including flight test subjects, test points, test methods and risk analysis; Furthermore, the method also includes: Step 8: Perform dependency chain integrity verification on the generated preliminary task outline, and perform logical subquery iterative retrieval and generation optimization on the parts that do not meet the specification conditions until the task outline meets all logical dependency constraints; Step 9: Output the final structured and standardized flight test mission outline, mission basis, and annotations mapping to the original development general requirements and specification clauses.

[0005] The final structured and standardized flight test mission outline includes flight test subjects, flight test sites, test methods, and risk analysis.

[0006] The normative entity information includes clauses, indicator requirements, test flight subjects, test points, and logical constraints; The resource library includes a capability index library, a standard and specification library, a flight test subject library, a test point library, a risk library, and a parameter library.

[0007] In step 6, candidate fragments are sorted according to constraint compliance scores. The constraint compliance scores for candidate resource fragments include node constraint matching degree, dependency chain integrity score, and semantic relevance score. The score results are used to comprehensively sort candidate fragments.

[0008] The process of generating the logical subquery expression in step 5 includes: The nodes in the normative dependency logic diagram are sorted according to logical priority. The nodes are used to indicate the compliance status of the clauses, the quantitative requirements of the indicators, and the prerequisites for the execution of the items. Generate logical subquery expressions directly from clauses with explicit indicator requirements; For clauses without explicit indicator requirements, digital indicators are generated by matching standards, specifications, and existing flight test data in the resource library, and logical subquery expressions are formed.

[0009] In step 7, a preliminary flight test mission outline is generated based on logical conditions, including: If the logical constraints on the candidate fragments are not met, the logical subquery will be re-executed to retrieve and generate supplementary fragments. Dynamically adjust the generated prompt words or logical subquery expressions to meet the requirements of normative logical dependencies; During the generation process, the test plans of different test flight subjects are cross-checked to avoid redundant test flights. Test points of different test flight subjects are intelligently merged and sorted to optimize the number of test flights and avoid repeated execution.

[0010] Among them, the test flight subjects, test points and indicators in the resource library have hierarchical relationships. When generating the preliminary test flight mission outline, subjects and test points that meet the dependency logic and optimize the test flight order are selected first to improve mission execution efficiency and reduce the number of repeated test flights.

[0011] In a second aspect, a flight test mission outline generation apparatus based on normative dependency logic is provided, which performs the method described in any of the first aspects, the apparatus comprising: The document parsing module is used to parse the overall development requirements and related normative documents, identify normative entity information, construct logical dependencies between entities, and form a normative dependency logic diagram. The logical subquery generation module is used to divide the normative entity information of different categories into corresponding pre-built resource libraries, establish mapping relationships between different resource libraries, and generate logical subquery expressions from the normative dependency logic graph according to the target machine model and task requirements. The adaptive retrieval module is used to retrieve canonical entity information fragments that meet logical dependency conditions from the resource repository, obtain candidate fragments, and sort them. The generation module is used to generate a preliminary flight test mission outline based on the sorted candidate fragments and triggered by logical conditions.

[0012] The device further includes: The verification module is used to perform dependency chain integrity verification on the generated preliminary task outline, and perform logical subquery iterative retrieval and generation optimization on parts that do not meet the specification conditions until the task outline meets all logical dependency constraints. The output module is used to output the final structured and standardized flight test mission outline, mission basis, and annotations mapped to the original development general requirements and specification clauses.

[0013] The logical subquery generation module, adaptive retrieval module, and generation module support an iterative closed loop to ensure that the task outline meets all normative dependency constraints, and can dynamically adjust the logical subquery expression and generated prompt words for unmet dependency constraints.

[0014] The beneficial effects of this invention are at least as follows: To address the challenges of incomplete outline coverage and compliance risks caused by complex logical dependencies between clauses and the lack of clear indicators, this invention employs a standardized dependency logic diagram and a closed-loop dependency chain integrity verification system to ensure the complete fulfillment of preconditions and logical constraints, achieving structured coverage and compliance assurance for the outline. Furthermore, to address the issues of flight redundancy and resource waste caused by the lack of overall coordination for cross-subject test points in traditional planning, this invention effectively avoids the dispersion and overlap of equivalent test points through precise logical sub-queries, candidate segment constraint compliance scoring and ranking, and intelligent merging and sorting of cross-subject test points. This significantly improves task execution efficiency, thereby effectively enhancing the reliability and safety of flight test plans. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. The drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart of the method for generating a flight test mission outline according to the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] The features and illustrative embodiments of various aspects of the present invention will now be described in detail. Numerous specific details are set forth in the following detailed description to provide a thorough understanding of the invention. However, it will be apparent to those skilled in the art that the invention may be practiced without requiring some of these specific details. The following description of embodiments is merely intended to provide a better understanding of the invention by illustrating examples of the invention. The invention is by no means limited to any specific setups and methods set forth below, but covers any improvements, substitutions, and modifications to structures, methods, and devices without departing from the spirit of the invention. Well-known structures and techniques are not shown in the drawings and the following description to avoid unnecessarily obscuring the invention.

[0019] See Figure 1This invention provides a method for generating flight test mission outlines based on normative dependency logic, the specific steps of which are as follows: Step 1: Analyze the overall development requirements and related normative documents, and identify the normative entity information therein. The normative entity information includes clauses, indicator requirements, test flight subjects, test points, and logical constraints. Step 2: Abstract the identified clauses and indicator requirements into normative entities, and construct the logical dependencies between entities to form a normative dependency logic diagram; Step 3: Based on the type of normative entity, classify the information of different categories of normative entities into the corresponding pre-built resource libraries. The resource libraries include capability index library, standard specification library, flight test subject library, test point library, risk library, and parameter library. Step 4: Establish mapping relationships between different resource repositories to ensure that various normative entity information entities and their dependencies can be accurately referenced in subsequent task processing; Step 5: Based on the target device model and task requirements, generate logical subquery expressions from the normative dependency logic graph to guide the retrieval of candidate resource fragments; Specifically, nodes in the normative dependency logic graph are sorted according to logical priority, and clauses with explicit indicator requirements are directly generated into logical subquery expressions; for clauses without explicit indicator requirements, digital indicators are generated by matching standards, specifications and existing flight test data in the resource library and forming logical subquery expressions.

[0020] Step 6: Retrieve normative entity information fragments that meet the logical dependency conditions in the resource library to obtain candidate fragments. Then, give a comprehensive score based on node constraint matching degree, dependency chain integrity score, and semantic relevance score, and sort the candidate fragments according to constraint compliance score. Step 7: Based on the sorted candidate segments, generate a preliminary flight test mission outline according to logical conditions, including flight test subjects, test points, test methods, and risk analysis. For candidate segments whose logical constraints are not met, regeneration is triggered, dynamically adjusting generation prompts or logical subquery expressions to meet normative logical dependency requirements. During the generation process, cross-checking of test plans between flight test subjects is performed to avoid redundant flight tests. Simultaneously, test points between different flight test subjects are intelligently merged and sorted to optimize the number of flight tests and avoid repeated execution.

[0021] Step 8: Perform dependency chain integrity verification on the generated preliminary task outline, and perform logical subquery iterative retrieval and generation optimization on the parts that do not meet the specification conditions until the task outline meets all logical dependency constraints; Step 9: Output the final structured and standardized flight test mission outline, mission basis, and annotations mapping to the original development requirements and specifications. The flight test mission outline includes flight test subjects, flight test points, test methods, and risk analysis.

[0022] This embodiment uses the generation of a stall speed flight test outline for a certain aircraft model as an example to illustrate the method for generating a flight test mission outline based on normative dependency logic: 1) Import the general requirements and airworthiness specifications for the development of this type of aircraft, parse out the compliance targets and items such as stall speed, construct a three-level entity of "requirements-items-test points", and generate a directed acyclic normative dependency logic graph containing logical dependencies; 2) The parsed entities and indicators are stored in resource libraries such as the standard specification library, flight test subject library, risk library, and parameter library, and an inter-library mapping relationship is established; 3) For the "stall speed test flight" mission objective, a multi-hop logical sub-query chain is generated, and candidate fragments such as stall test flight clauses, test points and historical parameters are retrieved from the resource library in topological order; for the case where there is no clear indicator item for "stall deceleration rate" in the overall development requirements, semantic features are extracted to retrieve the 1kn / s interval of similar aircraft models in the historical library, and combined with the aircraft model parameter scaling correction, the recommended digital indicator of 1kn / s is output after comparing with the mandatory clauses of the specification. 4) The retrieved candidate segments are sorted by weighted score based on node constraint matching degree, dependency chain integrity, and semantic relevance; 5) Input the high-quality segments into the generation module, and generate a draft outline containing stall test flight subjects, test points, methods and risk analysis according to logical conditions; 6) Perform dependency chain integrity verification on the initial draft outline, and backtrack along the logic graph to verify the completeness of supporting clauses; backtracking reveals that the prerequisites for "stall test flight" are missing "engine in-flight start test" and "high angle of attack characteristic test flight", locate the defect node and automatically trigger supplementary subqueries, iterate the retrieval and generation optimization until the logic loop is closed; 7) Intelligent merging and sorting of test points across subjects: test points with the same flap configuration in stall test flights and takeoff configuration test flights are merged into the same sorting, and the envelope points are sorted according to the approximation principle. Based on the risk database, high-risk stall subjects are placed in front, and equivalent duplicate points are eliminated. 8) Output a structured stall test flight mission outline containing mappings of regulatory clauses, generate and verify a closed loop to ensure the outline is fully compliant.

Claims

1. A method for generating flight test mission outlines based on normative dependency logic, characterized in that, The method includes: Step 1: Analyze the overall development requirements and related normative documents to identify the normative entity information within them; Step 2: Abstract the identified clauses and indicator requirements into normative entities, and construct the logical dependencies between entities to form a normative dependency logic diagram; Step 3: Based on the type of normative entity, classify the information of different categories of normative entities into the corresponding pre-built resource repositories; Step 4: Establish mapping relationships between different resource repositories; Step 5: Generate logical subquery expressions from the normalized dependency logic graph based on the target device model and task requirements; Step 6: Retrieve canonical entity information fragments that meet the logical dependency conditions in the resource library, obtain candidate fragments, and sort them; Step 7: Based on the sorted candidate segments, generate a preliminary test flight mission outline according to logical conditions.

2. The method according to claim 1, characterized in that, The method further includes: Step 8: Perform dependency chain integrity verification on the generated preliminary task outline, and perform logical subquery iterative retrieval and generation optimization on the parts that do not meet the specification conditions until the task outline meets all logical dependency constraints; Step 9: Output the final structured and standardized flight test mission outline, mission basis, and annotations mapping to the original development general requirements and specification clauses. The final structured and standardized flight test mission outline includes flight test subjects, flight test points, test methods, and risk analysis.

3. The method according to claim 2, characterized in that, The normative entity information includes clauses, indicator requirements, test flight subjects, test sites, and logical constraints; The resource library includes a capability index library, a standard and specification library, a flight test subject library, a test point library, a risk library, and a parameter library.

4. The method according to claim 3, characterized in that, In step 6, the candidate fragments are sorted according to the constraint compliance score. The constraint compliance score of the candidate resource fragments includes the node constraint matching degree, the dependency chain integrity score, and the semantic relevance score. The score results are used to comprehensively sort the candidate fragments.

5. The method according to claim 3, characterized in that, The process of generating the logical subquery expression in step 5 includes: The nodes in the normative dependency logic diagram are sorted according to logical priority. The nodes are used to indicate the compliance status of the clauses, the quantitative requirements of the indicators, and the prerequisites for the execution of the items. Generate logical subquery expressions directly from clauses with explicit indicator requirements; For clauses without explicit indicator requirements, digital indicators are generated by matching standards, specifications, and existing flight test data in the resource library, and logical subquery expressions are formed.

6. The method according to claim 3, characterized in that, In step 7, a preliminary flight test mission outline is generated based on logical conditions, including: If the logical constraints on the candidate fragments are not met, the logical subquery will be re-executed to retrieve and generate supplementary fragments. Dynamically adjust the generated prompt words or logical subquery expressions to meet the requirements of normative logical dependencies; During the generation process, the test plans of different test flight subjects are cross-checked to avoid redundant test flights; the test points of different test flight subjects are intelligently merged and sorted to optimize the number of test flights and avoid repeated execution.

7. The method according to claim 3, characterized in that, The test flight subjects, test points, and indicators in the resource library have hierarchical relationships. When generating the preliminary test flight mission outline, subjects and test points that satisfy the dependency logic and optimize the test flight sequence are selected first.

8. A flight test mission outline generation device based on normative dependency logic, characterized in that, The apparatus for performing the method as described in any one of claims 1 to 7 includes: The document parsing module is used to parse the overall development requirements and related normative documents, identify normative entity information, construct logical dependencies between entities, and form a normative dependency logic diagram. The logical subquery generation module is used to divide the normative entity information of different categories into corresponding pre-built resource libraries, establish mapping relationships between different resource libraries, and generate logical subquery expressions from the normative dependency logic graph according to the target machine model and task requirements. The adaptive retrieval module is used to retrieve canonical entity information fragments that meet logical dependency conditions from the resource library, obtain candidate fragments, and sort them. The generation module is used to generate a preliminary flight test mission outline based on the sorted candidate fragments and triggered by logical conditions.

9. The apparatus according to claim 8, characterized in that, The device further includes: The verification module is used to perform dependency chain integrity verification on the generated preliminary task outline, and perform logical subquery iterative retrieval and generation optimization on parts that do not meet the specification conditions until the task outline meets all logical dependency constraints. The output module is used to output the final structured and standardized flight test mission outline, mission basis, and annotations mapped to the original development general requirements and specification clauses; The logical subquery generation module, adaptive retrieval module, and generation module support an iterative closed loop to ensure that the task outline meets all normative dependency constraints, and can dynamically adjust the logical subquery expression and generated prompt words for unmet dependency constraints.

Images