Intelligent flight control coordination method and system based on barycenter recognition

Abstract

The application relates to the technical field of flight control data processing, in particular to an intelligent flight control cooperation method and system based on gravity center identification; the method comprises the following steps: collecting flight state data and a gravity center identification result to generate an abnormal state data set; an abnormal state feature set is constructed, an abnormal explanation judgment input feature is constructed according to the abnormal state feature set, a target abnormal explanation structure identifier is determined; a gravity center judgment data set is constructed, gravity center information decision availability judgment processing is carried out according to a preset judgment basis, a gravity center decision availability state is determined, and a cooperation input constraint set is generated; a flight control cooperation decision input is constructed; flight control cooperation decision input is subjected to cooperation decision processing to generate a flight control cooperation decision data set; the application provides an intelligent flight control cooperation method, in which, under a flight abnormal state, first, an abnormal cause is subjected to structured explanation, then, gravity center information is subjected to conditional availability judgment, and accordingly, gravity center data is restrained to participate in flight control cooperation decision input.

Description

A Smart Flight Control Cooperative Method and System Based on Center of Gravity Recognition Technical Field

[0001] This invention relates to the field of flight control data processing technology, specifically to an intelligent flight control collaborative method and system based on center of gravity identification. Background Technology

[0002] During the flight of a single-person standing multi-lift unit aircraft, the flight control system needs to perform real-time analysis of the flight status and execute collaborative control decisions in an operating environment with highly coupled loads, strong anomalies and frequent state changes. Existing flight control methods usually rely on the identification results of changes in the aircraft's center of gravity as an important basis for anomaly interpretation and flight control coordination.

[0003] During the flight of a single-person standing multi-lift unit aircraft, especially in sudden situations such as takeoff, hovering, obstacle avoidance, or rapid attitude adjustments, the flight control system may over-rely on changes in the center of gravity for interpreting and coordinating abnormal states. This can lead to flight adjustments being performed based on center of gravity shifts even when the abnormality is not primarily caused by changes in the center of gravity. This can result in problems such as inappropriate timing of coordinated intervention, unbalanced adjustment ranges, repeated fluctuations in flight status, and even control failure. Specifically, this includes: First, when the aircraft experiences attitude abnormalities or flight instability, the flight control system often directly interprets the abnormality as a change in the center of gravity and performs coordinated control accordingly. However, in actual operation, abnormalities are often caused by sudden changes in the state of a single lift unit, instantaneous force changes, or external disturbances. Continuing to control based on changes in the center of gravity in such cases can easily lead to deviations in the correction direction, further deteriorating the flight status. Second, even if the flight control system has identified a change in the center of gravity, this change is often merely a passive result or a short-term fluctuation following the occurrence of an abnormality. If the flight control system immediately uses this center of gravity identification result for coordinated control without differentiation, it can easily intervene in flight adjustments at inappropriate times, resulting in excessive coordinated actions, repeated corrections, or even control failure. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent flight control coordination method and system based on center of gravity recognition, in order to solve the problems mentioned in the background art where the flight control system relies too much on changes in the center of gravity for interpreting and coordinating abnormal states. This leads to flight adjustments being performed according to the center of gravity offset even when the abnormality is not dominated by changes in the center of gravity, resulting in problems such as inappropriate timing of coordination intervention, unbalanced adjustment range, repeated fluctuations in flight state, and even control failure.

[0005] To achieve the above objectives, the technical solution of the present invention is: an intelligent flight control cooperative method based on center of gravity recognition, comprising:

[0006] S1. Collect flight status data and center of gravity identification results of the aircraft during flight, perform data alignment and preprocessing, and generate an abnormal status dataset based on the preprocessed flight status data and center of gravity identification results.

[0007] S2. Extract and construct an abnormal state feature set based on the abnormal state dataset, construct an abnormal interpretation judgment input feature based on the abnormal state feature set, perform an abnormal interpretation structure matching judgment with the predefined multi-class abnormal interpretation structure feature pattern, obtain the matching result and determine the target abnormal interpretation structure identifier.

[0008] In S2, the abnormal state feature set is a set of data features used to characterize the abnormal occurrence process, including posture change patterns, center of gravity change features, execution response consistency features, and time correlation features; the abnormal interpretation judgment input features are feature vectors generated by the abnormal state feature set according to a predetermined feature arrangement rule; the predefined multi-class abnormal interpretation structure feature pattern is a predefined interpretation structure pattern, including a center of gravity-dominated interpretation structure feature pattern and a non-center of gravity-dominated interpretation structure feature pattern; the target abnormal interpretation structure identifier is used to indicate the different target abnormal interpretation structures adopted by the current abnormal state, and the target abnormal interpretation structure includes a center of gravity-dominated interpretation structure and a non-center of gravity-dominated interpretation structure;

[0009] S3. Construct a center of gravity determination dataset based on the target interpretation structure identifier and abnormal state dataset, and perform center of gravity information decision availability determination processing on the center of gravity determination dataset according to the preset determination criteria, determine the center of gravity decision availability status of the center of gravity identification data under the current target abnormal interpretation structure, and generate a collaborative input constraint set based on the center of gravity decision availability status.

[0010] In S3, the center of gravity determination dataset is a data set constructed based on the data related to center of gravity identification in the abnormal state dataset and combined with the target abnormal interpretation structure identifier; the preset determination basis is a set of preset determination conditions used to perform center of gravity information decision availability determination; the center of gravity decision availability status is a status identifier used to characterize whether the center of gravity identification data is allowed to undergo subsequent collaborative decision processing; the collaborative input constraint set is a data set generated based on the center of gravity decision availability status to constrain the center of gravity identification data to enter collaborative decision processing;

[0011] S4. Based on the abnormal state dataset, target interpretation structure identifier, and collaborative input constraint set, construct the flight control collaborative decision input; perform collaborative decision processing on the flight control collaborative decision input to generate the flight control collaborative decision dataset;

[0012] In S4, the flight control collaborative decision dataset is a data set used to carry the target anomaly interpretation structure identifier, the available state of the center of gravity decision, and the collaborative input constraint set.

[0013] Preferably, in S2, the abnormal state feature set includes attitude change patterns, center of gravity change features, execution response consistency features, and time correlation features, reflecting the structured joint change relationships of the aircraft in the dimensions of attitude change, center of gravity change, execution response consistency, and time correlation under abnormal flight conditions; the abnormal interpretation judgment input features are the carriers of abnormal interpretation structure matching processing, which are reorganized and generated according to predetermined feature arrangement rules; wherein, the predetermined feature arrangement rules are used to uniformly arrange and format the data of different feature dimensions in the abnormal state feature set.

[0014] Preferably, in S2, the multiple types of anomaly interpretation structure feature patterns are also used to distinguish and model the dominant relationships between different feature dimensions in the anomaly state feature set. The center-of-gravity dominant interpretation structure feature patterns and non-center-of-gravity dominant interpretation structure feature patterns correspond to different anomaly state feature association relationships. Specifically, the center-of-gravity dominant interpretation structure feature pattern is an interpretation structure feature pattern that uses center-of-gravity change features as the main representation feature of the anomaly state, and its corresponding anomaly state feature association relationship shows a consistent dominant relationship between the center-of-gravity change feature and the posture change pattern. The non-center-of-gravity dominant interpretation structure feature pattern is an interpretation structure feature pattern that uses non-center-of-gravity change features as the main representation feature of the anomaly state, and its corresponding anomaly state feature association relationship shows that at least one of the posture change pattern, the execution response consistency feature, and the time association feature has a non-dominant relationship with the center-of-gravity change feature. The specific logic for distinguishing between the center-of-gravity dominant interpretation structure feature pattern and the non-center-of-gravity dominant interpretation structure feature pattern is as follows: based on the strength of the association relationship between each feature dimension in the anomaly state feature set, the degree of dominance of the center-of-gravity change feature in the anomaly state is determined, and the interpretation structure feature pattern type corresponding to the anomaly state is determined based on the degree of dominance.

[0015] Preferably, in step S2, the anomaly explanation structure matching determination refers to a data processing process that compares the anomaly explanation determination input features with the corresponding relationships of the multiple anomaly explanation structure feature patterns, used to determine the matching relationship between the anomaly state and different anomaly explanation structure feature patterns; the specific method of the anomaly explanation structure matching determination is as follows: based on the anomaly explanation determination input features, calculate the association matching value between the anomaly explanation determination input features and each anomaly explanation structure feature pattern respectively, and form a matching result for each anomaly explanation structure feature pattern according to the association matching value; the matching result is a data set used to characterize the degree of matching between the anomaly explanation determination input features and each anomaly explanation structure feature pattern; the target anomaly explanation structure identifier is the type identifier of the anomaly explanation structure feature pattern whose corresponding association matching value in the matching result satisfies the preset determination rule.

[0016] Preferably, in step S3, the preset judgment basis is a set of pre-set judgment conditions for performing the judgment of the availability of center of gravity information decision. The set of pre-set judgment conditions includes judgment conditions for judging the consistency between the center of gravity change features and the target anomaly interpretation structure identifier, judgment conditions for judging the persistence of the center of gravity change features in the abnormal state, and judgment conditions for judging the causal role of the center of gravity change features in the process of anomaly occurrence.

[0017] Preferably, in step S3, the center of gravity information decision availability determination process is performed under the constraint of the target anomaly interpretation structure corresponding to the target anomaly interpretation structure identifier. Specifically, it involves determining the set of judgment conditions corresponding to the target anomaly interpretation structure identifier, and performing the center of gravity information decision availability determination process on the center of gravity determination dataset under the constraint of the set of judgment conditions to obtain the center of gravity decision availability status.

[0018] Preferably, in step S3, the available state of the center of gravity decision includes allowed use, restricted use, and prohibited use; the available state of the center of gravity decision is used to characterize the degree to which the center of gravity identification data is allowed to enter the collaborative decision processing under the current target anomaly interpretation structure; and a corresponding collaborative input constraint set is generated based on the available state of the center of gravity decision to constrain the center of gravity identification data to enter the subsequent collaborative decision processing; the collaborative input constraint set includes constraint information for allowing the center of gravity identification data to directly participate in the collaborative decision processing, constraint information for limiting the scope or weight of participation of the center of gravity identification data, and constraint information for prohibiting the center of gravity identification data from entering the collaborative decision processing.

[0019] Preferably, the available state of the center of gravity decision is determined based on the comprehensive judgment result of each judgment condition in the pre-set judgment condition set, wherein the judgment condition for determining consistency, the judgment condition for determining persistence, and the judgment condition for determining causal role jointly participate in the center of gravity information decision availability judgment processing to generate the corresponding available state of the center of gravity decision.

[0020] Preferably, in step S4, the flight control collaborative decision input is constructed under the constraints of the collaborative input constraint set. Specifically, based on the collaborative input constraint set, the participation mode of the center of gravity identification data in the flight control collaborative decision input is constrained to construct the flight control collaborative decision input.

[0021] On the other hand, the present invention provides an intelligent flight control cooperative system based on center of gravity recognition, including a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement the aforementioned intelligent flight control cooperative method based on center of gravity recognition.

[0022] Compared with the prior art, the above-mentioned technical solution of the present invention has the following beneficial technical effects:

[0023] 1. In this invention, the collaborative processing method based on the combination of anomaly interpretation structure determination and center of gravity information decision availability determination can avoid the flight control system from simply attributing all anomalies to center of gravity changes and directly intervening in collaborative control when the flight anomaly is in a state of abnormality. This prevents unnecessary or directional deviation collaborative adjustments from being introduced due to the incorrect use of center of gravity information when the center of gravity change is not the main cause of the anomaly or is only a secondary result.

[0024] 2. In this invention, by introducing a set of collaborative input constraints during the collaborative decision-making input construction stage, the structured constraints on whether and how the center of gravity identification data participates in collaborative decision-making are realized, thereby transforming the center of gravity information from default participation to conditional participation. This enables flight control collaborative decision-making to selectively use center of gravity information based on different anomaly interpretation structures, thereby improving the stability and rationality of collaborative decision-making processing under abnormal conditions. Attached Figure Description

[0025] Figure 1 is a flowchart of an embodiment of the present invention. Detailed Implementation

[0026] Example 1, as shown in Figure 1, illustrates an intelligent flight control cooperative method based on center of gravity recognition proposed in this invention. The specific implementation steps are as follows:

[0027] S1. Collect flight status data and center of gravity identification results of the aircraft during flight, perform data alignment and preprocessing, and generate an abnormal status dataset based on the preprocessed flight status data and center of gravity identification results.

[0028] S2. Extract and construct an abnormal state feature set based on the abnormal state dataset, construct an abnormal interpretation judgment input feature based on the abnormal state feature set, perform an abnormal interpretation structure matching judgment with the predefined multi-class abnormal interpretation structure feature pattern, obtain the matching result and determine the target abnormal interpretation structure identifier.

[0029] In S2, the abnormal state feature set is a set of data features used to characterize the abnormal occurrence process, including posture change patterns, center of gravity change features, execution response consistency features, and time correlation features; the abnormal interpretation judgment input features are feature vectors generated by the abnormal state feature set according to a predetermined feature arrangement rule; the predefined multi-class abnormal interpretation structure feature pattern is a predefined interpretation structure pattern, including a center of gravity-dominated interpretation structure feature pattern and a non-center of gravity-dominated interpretation structure feature pattern; the target abnormal interpretation structure identifier is used to indicate the different target abnormal interpretation structures adopted by the current abnormal state, and the target abnormal interpretation structure includes a center of gravity-dominated interpretation structure and a non-center of gravity-dominated interpretation structure;

[0030] S3. Construct a center of gravity determination dataset based on the target interpretation structure identifier and abnormal state dataset, and perform center of gravity information decision availability determination processing on the center of gravity determination dataset according to the preset determination criteria, determine the center of gravity decision availability status of the center of gravity identification data under the current target abnormal interpretation structure, and generate a collaborative input constraint set based on the center of gravity decision availability status.

[0031] In S3, the center of gravity determination dataset is a data set constructed based on the data related to center of gravity identification in the abnormal state dataset and combined with the target abnormal interpretation structure identifier; the preset determination basis is a set of preset determination conditions used to perform center of gravity information decision availability determination; the center of gravity decision availability status is a status identifier used to characterize whether the center of gravity identification data is allowed to undergo subsequent collaborative decision processing; the collaborative input constraint set is a data set generated based on the center of gravity decision availability status to constrain the center of gravity identification data to enter collaborative decision processing;

[0032] S4. Based on the abnormal state dataset, target interpretation structure identifier, and collaborative input constraint set, construct the flight control collaborative decision input; perform collaborative decision processing on the flight control collaborative decision input to generate the flight control collaborative decision dataset;

[0033] In S4, the flight control collaborative decision dataset is a data set used to carry the target anomaly interpretation structure identifier, the available state of the center of gravity decision, and the collaborative input constraint set.

[0034] In this embodiment S1, during the flight of the aircraft, flight status data and center of gravity identification results are collected. The flight status data characterizes the overall operational state of the aircraft during flight and includes at least attitude-related status data, execution response-related status data, and corresponding time stamp information. The center of gravity identification result is the center of gravity position or center of gravity change-related data output by the aircraft's existing center of gravity identification module, reflecting the changes in the aircraft's center of gravity state during flight. To ensure that flight status data and center of gravity identification results from different sources can jointly describe the same flight process, data alignment processing is performed on the collected data. This data alignment processing includes at least temporal alignment and data granularity alignment, enabling different data to form a comparable data sequence within the same temporal reference frame. This ensures that various types of data can be jointly analyzed based on a unified flight status context during subsequent processing. After data alignment, preprocessing operations are performed on the flight status data and center of gravity identification results. These preprocessing operations are used to standardize the format of the original data, handle outliers, handle missing data, and remove noisy data, so that the processed data meets the input requirements for subsequent feature extraction and judgment processing. Based on the flight status data and center of gravity identification results after data alignment and preprocessing, an abnormal state dataset is constructed. The abnormal state dataset is a collection of data used to describe the aircraft during the abnormal process. It is organized according to the time interval of the abnormality, and is used to provide a unified data foundation for subsequent abnormal state feature extraction, abnormal interpretation structure determination, and center of gravity information decision availability determination.

[0035] In this embodiment S2, the abnormal state feature set includes attitude change patterns, center of gravity change features, execution response consistency features, and time correlation features, reflecting the structured joint change relationship of the aircraft in the dimensions of attitude change, center of gravity change, execution response consistency, and time correlation under abnormal flight conditions; the abnormal interpretation judgment input features are the carrier of abnormal interpretation structure matching processing, which are reorganized and generated according to a predetermined feature arrangement rule; wherein, the predetermined feature arrangement rule is used to uniformly arrange and format the data of different feature dimensions in the abnormal state feature set.

[0036] In this embodiment S2, based on the abnormal state dataset, feature extraction processing is performed on the multi-source state data of the aircraft under abnormal flight conditions to construct an abnormal state feature set for characterizing the abnormal occurrence process. The abnormal state feature set is used to structurally express the changes of the abnormal state in different feature dimensions, providing a unified feature basis for the matching and judgment of the subsequent abnormal interpretation structure. The construction of the abnormal state feature set is based on the changing relationships of various flight state data in the abnormal state dataset, specifically including performing feature extraction and feature processing on attitude-related data, center of gravity-related data, execution response-related data, and time series data respectively. The feature extraction processing may include, but is not limited to, data processing methods such as trend extraction, change amplitude representation, time series correlation analysis, and state consistency analysis, thereby converting the original abnormal state data into a feature representation that can reflect the structural characteristics of the abnormal state.

[0037] In this embodiment S2, the attitude change mode is used to characterize the changes in attitude-related parameters of the aircraft during the occurrence of an abnormal state. It reflects the trend characteristics, combination characteristics, or staged change relationships of the aircraft's attitude changes during the abnormal state. The attitude change mode is not limited to specific attitude angle values ​​or attitude control parameters, but is used to describe the overall change behavior of the attitude during the abnormal process. The center of gravity change feature is used to characterize the changes in the position or distribution of the center of gravity of the aircraft during the occurrence of an abnormal state. It reflects the trend, magnitude, or duration of the change of the center of gravity during the abnormal process. The center of gravity change feature is used to characterize the change behavior of the center of gravity state before, during, and after the occurrence of the abnormality, but does not presuppose that the change of the center of gravity necessarily constitutes the direct cause of the abnormality.

[0038] In this embodiment S2, the execution response consistency feature is used to characterize the correspondence between control commands and execution responses of the aircraft in an abnormal state. It reflects the degree of consistency between the control commands output by the flight control system and the actual execution state of the aircraft. When the aircraft is in an abnormal state, by analyzing the relationship between changes in control commands and changes in execution feedback, it can be determined whether the abnormal state is related to the deviation of the execution layer response, thereby providing a reference for the determination of the abnormal interpretation structure. The time correlation feature is used to characterize the correlation between various features in the abnormal state feature set in the time dimension. It reflects the order of appearance, duration, and time correspondence between different features in the abnormal occurrence process. The time correlation feature is used to support the analysis of the sequential relationship and evolution process of various features in the abnormal state, providing structural information in the time dimension for the subsequent matching and determination of the abnormal interpretation structure.

[0039] In this embodiment S2, after constructing the abnormal state feature set, the abnormal state feature set is used as the basis for generating the abnormal interpretation judgment input features. The abnormal interpretation judgment input features are not a simple feature set, but a unified feature carrier used to carry the overall structural information of the abnormal state. It is used to support the matching and judgment processing between the abnormal interpretation structure feature pattern and the abnormal state. The abnormal interpretation judgment input features are generated by recombining data of different feature dimensions in the abnormal state feature set according to a predetermined feature arrangement rule. The predetermined feature arrangement rule is used to uniformly constrain the arrangement order and data format of posture change pattern, center of gravity change feature, execution response consistency feature, and time correlation feature in the input features, thereby avoiding ambiguity in the matching and judgment process of different feature dimensions and ensuring that the abnormal interpretation structure matching and judgment can be based on a consistent feature expression.

[0040] In this embodiment S2, the multiple types of anomaly interpretation structure feature patterns are also used to distinguish and model the dominant relationships between different feature dimensions in the anomaly state feature set. The centroid-dominant interpretation structure feature pattern and the non-centroid-dominant interpretation structure feature pattern correspond to different anomaly state feature association relationships. Specifically, the centroid-dominant interpretation structure feature pattern uses centroid change features as the main representation feature of the anomaly state, and its corresponding anomaly state feature association relationship shows a consistent dominant relationship between centroid change features and posture change patterns. The non-centroid-dominant interpretation structure feature pattern uses non-centroid change features as the main representation feature of the anomaly state, and its corresponding anomaly state feature association relationship shows that at least one of posture change patterns, execution response consistency features, and time association features has a non-dominant relationship with centroid change features. The specific logic for distinguishing between the centroid-dominant and non-centroid-dominant interpretation structure feature patterns is as follows: based on the strength of the association relationship between each feature dimension in the anomaly state feature set, the degree of dominance of the centroid change feature in the anomaly state is determined, and the interpretation structure feature pattern type corresponding to the anomaly state is determined based on the degree of dominance.

[0041] In this embodiment S2, after the construction of the abnormal state feature set is completed, in order to avoid interpreting the flight abnormal state based on only a single feature, this embodiment introduces an abnormal interpretation structure feature pattern to perform structured modeling of the relationship between different feature dimensions in the abnormal state feature set; the abnormal interpretation structure feature pattern is used to describe the dominant relationship structure between various features in the abnormal state, thereby providing different interpretation paths for the abnormal state, so that the subsequent abnormal interpretation process can be selected based on the inherent feature correlation relationship of the abnormal state.

[0042] In this embodiment S2, multiple anomaly explanation structure feature patterns are predefined to form an anomaly explanation structure pattern set. This set stores explanation structure templates for different types of anomaly states. Each anomaly explanation structure feature pattern describes the typical dominant relationship structure between each feature dimension in the anomaly state feature set. It can be saved in structured data form, including but not limited to feature dimension identifiers, feature relationship type identifiers, and relationship constraint description information, so that the anomaly explanation structure feature pattern can serve as a reference object for subsequent anomaly explanation structure matching and determination. The anomaly state feature association relationship describes the association between different feature dimensions in the anomaly state feature set during the anomaly occurrence process. The association relationship is not an abstract correlation description, but exists in the form of an expressible relationship object. The association relationship object may include feature dimension identifiers participating in the association, association direction identifiers, and association degree description information, which are used to characterize the mutual influence relationship between feature dimensions in the anomaly state, thereby providing a structured relationship basis for determining the dominant relationship.

[0043] In this embodiment S2, the degree of dominance is used to characterize the relative dominance of the center of gravity change feature in the multidimensional feature association relationship in the abnormal state feature set. It is a judgment result obtained based on the association strength analysis. The degree of dominance is used to reflect the importance level of the center of gravity change feature in the interpretation of the abnormal state. It can be represented in the form of a level identifier, interval identifier, or category identifier to support the determination of the abnormal interpretation structure feature pattern type, without being limited to a specific numerical form. The association strength is used to measure the closeness of the relationship between different feature dimensions in the abnormal state feature set. It is formed based on the change features of the abnormal state data within the time period of the abnormality. The association strength can be formed based on the consistency of the feature change trend, the synchronization of the change sequence, or the stability of the response relationship. It is used to reflect the closeness of the association between different feature dimensions in the abnormal state and to provide a quantitative or ordinal quantitative basis for the subsequent determination of the degree of dominance.

[0044] In this embodiment S2, multiple types of anomaly interpretation structure feature patterns constitute an interpretation structure pattern set in a predefined manner. Each anomaly interpretation structure feature pattern is used to describe the dominant relationship type between different feature dimensions in the anomaly state feature set. The anomaly interpretation structure feature pattern includes at least a center-dominant interpretation structure feature pattern and a non-center-dominant interpretation structure feature pattern. Different interpretation structure feature patterns correspond to different anomaly state feature association structures, which are used to reflect the difference in the degree of dominance of the center-dominant change feature in the multidimensional feature relationship in the anomaly state.

[0045] In this embodiment S2, the center-of-gravity dominant explanatory structural feature pattern is used to describe the situation where the center-of-gravity change feature is the main characterizing feature of the abnormal state during the occurrence of the abnormal state. Under this explanatory structural feature pattern, the center-of-gravity change feature and the posture change pattern have a consistent dominant relationship, that is, the trend, direction or time evolution of the center-of-gravity change and posture change in the abnormal state are highly consistent, so that the center-of-gravity change feature constitutes the main reference axis in the explanation of the abnormal state, thereby the abnormal state can be explained as an abnormal structure dominated by the center-of-gravity change. The non-center-of-gravity dominant explanatory structural feature pattern is used to describe the situation where the main characterizing feature of the abnormal state is not dominated by the center-of-gravity change feature during the occurrence of the abnormal state. Under this explanatory structural feature pattern, at least one of the posture change pattern, the execution response consistency feature and the time correlation feature has a non-dominant relationship with the center-of-gravity change feature, that is, the center-of-gravity change feature does not form the main explanatory basis in the abnormal state, the key evidence of the abnormal state comes from other feature dimensions, and the center-of-gravity change feature only shows an accompanying change or a resultant change, so that the explanation of the abnormal state no longer takes the center-of-gravity change as the core explanatory axis.

[0046] In this embodiment S2, to distinguish between the center-of-gravity dominant explanatory structure feature patterns and the non-center-of-gravity dominant explanatory structure feature patterns, this embodiment models the correlation between each feature dimension based on the abnormal state feature set, forming a correlation set to describe the degree of correlation between features. The correlation set is used to characterize the correlation strength between the center-of-gravity change feature and the posture change pattern, the execution response consistency feature, and the time correlation feature in the abnormal state, thereby providing basic data support for subsequent dominance relationship determination. The correlation strength is used to characterize the closeness of the relationship between different feature dimensions in the abnormal state feature set, which can be formed based on the change relationship, synchronization relationship, or response relationship of the abnormal state data in the time series. By comparing and analyzing the correlation strength between the center-of-gravity change feature and other feature dimensions, the dominance of the center-of-gravity change feature in the abnormal state can be obtained, which is used to reflect the relative importance of the center-of-gravity change in the multidimensional abnormal feature relationship structure.

[0047] In this embodiment S2, after obtaining the dominance of the center of gravity change feature, the abnormal state is mapped to a center of gravity-dominant explanatory structure feature pattern or a non-center of gravity-dominant explanatory structure feature pattern according to the preset explanatory structure differentiation rule. The explanatory structure differentiation rule is used to correspond the dominance of the center of gravity change feature with the type of abnormal explanatory structure feature pattern, thereby determining the type of abnormal explanatory structure to be adopted for the current abnormal state, and providing a structured explanatory basis for subsequent abnormal explanatory structure matching and determination of the availability of center of gravity information decision.

[0048] In this embodiment S2, the anomaly explanation structure matching determination refers to the data processing process of comparing the anomaly explanation determination input features with the corresponding relationships of the multiple anomaly explanation structure feature patterns, in order to determine the matching relationship between the anomaly state and different anomaly explanation structure feature patterns. The specific method of the anomaly explanation structure matching determination is as follows: based on the anomaly explanation determination input features, the association matching value between the anomaly explanation determination input features and each anomaly explanation structure feature pattern is calculated respectively, and a matching result is formed for each anomaly explanation structure feature pattern according to the association matching value; the matching result is a data set used to characterize the degree of matching between the anomaly explanation determination input features and each anomaly explanation structure feature pattern; the target anomaly explanation structure identifier is the type identifier of the anomaly explanation structure feature pattern whose corresponding association matching value in the matching result satisfies the preset determination rule.

[0049] In this embodiment S2, after generating the anomaly interpretation judgment input features and completing the predefinition of multiple anomaly interpretation structure feature patterns, anomaly interpretation structure matching judgment is performed to determine the interpretation structure type corresponding to the current anomaly state. The anomaly interpretation structure matching judgment is a correspondence comparison process oriented towards data processing. Its input includes at least the anomaly interpretation judgment input features and multiple anomaly interpretation structure feature patterns, and its output includes at least the matching result and the target anomaly interpretation structure identifier. The anomaly interpretation judgment input features are used to carry the structured expression of the anomaly state feature set in each feature dimension, and the multiple anomaly interpretation structure feature patterns are used to provide reference relationship structures for different interpretation structure types, so that the anomaly state can be judged as an anomaly interpretation structure type that matches a certain interpretation structure feature pattern in the candidate interpretation structure type space.

[0050] In this embodiment S2, the correspondence comparison used in the anomaly interpretation structure matching determination is not based on the similarity comparison of a single feature value, but rather on the consistency of the correspondence between the feature dimension relationship structure expressed by the anomaly interpretation determination input feature and the dominant relationship structure expressed by each anomaly interpretation structure feature pattern. Therefore, this embodiment performs association matching value calculation processing for each anomaly interpretation structure feature pattern during the matching determination process. The association matching value is used to characterize the degree of matching between the anomaly interpretation determination input feature and the corresponding anomaly interpretation structure feature pattern. It can be formed based on the degree of conformity between the relationship expression of each feature dimension in the anomaly interpretation determination input feature and the predefined relationship constraints in the anomaly interpretation structure feature pattern, thus... Each anomaly explanation structure feature pattern corresponds to an associated matching value, thereby supporting the formation of comparable matching descriptions for multiple types of anomaly explanation structure feature patterns. After completing the calculation of the associated matching value, each anomaly explanation structure feature pattern and its corresponding associated matching value are organized to form a matching result. The matching result is a data set used to characterize the degree of matching between the anomaly explanation judgment input features and each anomaly explanation structure feature pattern. The matching result includes at least the type identifier of the anomaly explanation structure feature pattern and the associated matching value corresponding to the type identifier, so that the matching judgment process can carry the overall matching description of the candidate explanation structure type space in the form of a set result, and provide a unified data foundation for the subsequent determination of the target anomaly explanation structure identifier.

[0051] In this embodiment S2, the target anomaly explanation structure identifier is determined by the matching results according to preset judgment rules. The preset judgment rules are a set of pre-defined rules used to filter and determine the target anomaly explanation structure type from the matching results. They are used to compare, filter, or constrain the associated matching values ​​of each anomaly explanation structure feature pattern in the matching results to determine the anomaly explanation structure feature pattern type that meets the explanation structure selection conditions, and output the type as the target anomaly explanation structure identifier. The target anomaly explanation structure identifier, as structured identifier data of the anomaly explanation structure type, is used to indicate the explanation structure path that should be adopted in the current anomaly state, and serves as the input basis for explanation structure constraints in the subsequent determination of the availability of center of gravity information, so that the subsequent processing of the center of gravity identification data can be constrained and executed under the determined anomaly explanation structure type.

[0052] In this embodiment S2, the target anomaly interpretation structure includes a center-of-gravity dominant interpretation structure and a non-center-of-gravity dominant interpretation structure. The target anomaly interpretation structure is a structured interpretation type used to characterize the interpretation path of the current anomaly state, and it includes at least a center-of-gravity dominant interpretation structure and a non-center-of-gravity dominant interpretation structure. The center-of-gravity dominant interpretation structure is used to represent an anomaly state that is mainly dominated by the center-of-gravity change feature. Under this interpretation structure, the center-of-gravity change feature serves as the main interpretation basis for the anomaly state and participates in the subsequent anomaly handling process. The non-center-of-gravity dominant interpretation structure is used to represent an anomaly state that is not dominated by the center-of-gravity change feature. Under this interpretation structure, the main interpretation basis for the anomaly state comes from at least one of the posture change pattern, execution response consistency feature, or time correlation feature. The center-of-gravity change feature exists only as an accompanying or resultant feature in the anomaly interpretation. The anomaly state is interpreted as a center-of-gravity dominant interpretation structure or a non-center-of-gravity dominant interpretation structure.

[0053] In this embodiment S3, the preset judgment basis is a set of pre-set judgment conditions for performing the judgment of the availability of center of gravity information decision. The set of pre-set judgment conditions includes judgment conditions for judging the consistency between the center of gravity change feature and the target anomaly interpretation structure identifier, judgment conditions for judging the persistence of the center of gravity change feature in the abnormal state, and judgment conditions for judging the causal role of the center of gravity change feature in the anomaly occurrence process.

[0054] In this embodiment S3, after determining the anomaly interpretation structure type, to avoid the center of gravity identification data being indiscriminately used for collaborative decision-making under anomaly conditions, this embodiment performs a decision availability determination on the center of gravity information. The center of gravity information decision availability determination is based on a pre-set determination criterion, which exists in the form of a set of determination conditions. This set is used to make a constraint judgment on the rationality and applicability of the center of gravity change features under the current anomaly condition from multiple dimensions, thereby providing a determination basis for the construction of subsequent collaborative decision inputs. The preset determination criterion is a pre-configured set of determination conditions, which is used to limit the usage conditions of the center of gravity change features under different anomaly interpretation structure types. The set of determination conditions can be stored in the form of rule entries, condition templates, or configuration item sets. Each determination condition includes at least the determination object, the applicable anomaly time period, and the corresponding determination output form, so that the center of gravity information decision availability determination can be performed within a clear data boundary and time range.

[0055] In this embodiment S3, the consistency determination condition is used to judge the consistency between the center of gravity change feature and the target anomaly interpretation structure identifier. This consistency does not refer to the formal consistency between the center of gravity change feature and the anomaly interpretation structure type label, but rather whether the change relationship of the center of gravity change feature in the anomaly state conforms to the dominant relationship structure required by the target anomaly interpretation structure. The consistency determination takes the relationship structure between the center of gravity change feature and the posture change pattern, execution response consistency feature, and time-related feature in the anomaly state feature set as the determination object. By analyzing the change trend, change direction, or time evolution relationship between each feature during the anomaly occurrence process, it is determined whether the center of gravity change feature meets the constraint requirements of the current anomaly interpretation structure type on the dominant relationship. The persistence determination condition is used to judge the persistence of the center of gravity change feature during the anomaly occurrence process. This persistence determination is based on the anomaly time corresponding to the anomaly occurrence process. The process involves executing continuous sub-time windows within or between time intervals to distinguish between short-term fluctuating center-of-gravity changes and continuous offset center-of-gravity changes. When a center-of-gravity change feature exhibits only short-term fluctuations within an abnormal time period and fails to form a stable change feature, this center-of-gravity change feature should be considered as lacking a stable basis for judgment in subsequent collaborative decision-making, thus limiting its usability in center-of-gravity information decision-making. The judgment criteria for determining causal roles are used to determine the causal role type of the center-of-gravity change feature during the abnormal occurrence process. The causal role type describes the attribute of the center-of-gravity change feature's role in the abnormal state. The causal role determination is based on the time correlation features and execution response consistency features of the abnormal state feature set. By analyzing the sequential and corresponding relationships between the center-of-gravity change feature and the posture change pattern or execution response anomaly, the center-of-gravity change feature is determined as a causal feature, a result feature, or an accompanying feature, thereby avoiding misjudging abnormal result changes as abnormal causal features.

[0056] In this embodiment S3, the determination conditions for consistency, persistence, and causal roles together constitute the basis for determining the availability of center of gravity information for decision-making. Each determination condition makes a constraint judgment on the characteristics of center of gravity change from three different dimensions: interpretive structural constraints, time stability, and causal relationship. The determination results are used as the input basis for determining the availability of center of gravity decision-making in the future, so that whether the center of gravity identification data is allowed to enter the collaborative decision-making process can be completed under the constraints of structured and interpretable determination conditions.

[0057] In this embodiment S3, the center of gravity information decision availability determination processing is performed under the constraints of the target anomaly interpretation structure corresponding to the target anomaly interpretation structure identifier. Specifically, it involves determining the set of determination conditions corresponding to the target anomaly interpretation structure identifier, and performing center of gravity information decision availability determination processing on the center of gravity determination dataset under the constraints of the set of determination conditions to obtain the center of gravity decision availability status.

[0058] In this embodiment S3, the determination of the availability of center of gravity information is not performed under a fixed rule for the center of gravity change features, but rather under the constraint of the target anomaly interpretation structure corresponding to the target anomaly interpretation structure identifier. The target anomaly interpretation structure serves as the determination context, used to limit the interpretation path of the center of gravity change features under the current anomaly state, enabling the availability determination of center of gravity information to be differentiated based on the anomaly interpretation structure type, thereby avoiding the use of the same determination logic under different anomaly structure scenarios. Based on the target anomaly interpretation structure identifier, a set of determination conditions corresponding to it is determined. This set of determination conditions is not a fixed combination of conditions, but rather selected or combined from preset determination criteria according to the target anomaly interpretation structure type. Different target anomaly interpretation structures correspond to different determination condition configurations. This leads to differences in the judgment constraints applicable to center-of-gravity change characteristics under center-of-gravity-dominant and non-center-of-gravity-dominant explanatory structures, thus introducing the type of abnormal explanatory structure into the condition selection process for determining the decision availability of center-of-gravity information. When the target abnormal explanatory structure identifier corresponds to a center-of-gravity-dominant explanatory structure, judgment conditions related to the consistency of center-of-gravity change characteristics are prioritized and included in the current judgment condition set to verify whether the center-of-gravity change characteristics conform to the dominant relationship structure required by the explanatory structure. When the target abnormal explanatory structure identifier corresponds to a non-center-of-gravity-dominant explanatory structure, judgment conditions related to the persistence and causal role of center-of-gravity change characteristics are prioritized and included in the current judgment condition set to identify whether the center-of-gravity change characteristics are merely accompanying or consequential changes, thereby avoiding mistakenly using center-of-gravity change characteristics as the primary decision-making basis in non-center-of-gravity-dominant abnormal states.

[0059] In this embodiment S3, after determining the currently valid set of judgment conditions, the center of gravity judgment dataset is subjected to center of gravity information decision availability judgment processing under the constraints of the set of judgment conditions. The center of gravity judgment dataset, as the direct input object for judgment, contains center of gravity change features and context data related to the abnormal state, so that each judgment condition can make a constrained judgment on the center of gravity change features within a clear data range, rather than directly affecting the original value of the center of gravity identification result. By performing center of gravity information decision availability judgment processing under the constraints of the target abnormal interpretation structure, the same center of gravity change feature can obtain different judgment results in different abnormal interpretation structure contexts. Thus, whether the center of gravity change feature is allowed to enter the subsequent collaborative decision processing no longer depends only on its numerical change, but on its structural role in the current abnormal interpretation structure, thereby providing a judgment basis based on the abnormal interpretation structure for determining the center of gravity decision availability state.

[0060] In this embodiment S3, the available state of the center of gravity decision includes allowed use, restricted use, and prohibited use; the available state of the center of gravity decision is used to characterize the degree to which the center of gravity identification data is allowed to enter the collaborative decision processing under the current target anomaly interpretation structure; and a corresponding collaborative input constraint set is generated based on the available state of the center of gravity decision to constrain the center of gravity identification data to enter the subsequent collaborative decision processing; the collaborative input constraint set includes constraint information for allowing the center of gravity identification data to directly participate in the collaborative decision processing, constraint information for limiting the scope or weight of participation of the center of gravity identification data, and constraint information for prohibiting the center of gravity identification data from entering the collaborative decision processing.

[0061] In this embodiment S3, after completing the determination of the availability of center of gravity information for decision-making, the determination result is structured into a center of gravity decision availability state. The center of gravity decision availability state is used to characterize the degree to which the center of gravity identification data is allowed to enter the collaborative decision-making process under the current target anomaly interpretation structure. The center of gravity decision availability state includes three state types: allowed use, restricted use, and prohibited use. Allowed use means that the center of gravity identification data meets the constraints of the current anomaly interpretation structure and can be directly used as part of the collaborative decision input to participate in subsequent processing. Restricted use means that the center of gravity identification data can only participate in the collaborative decision-making process under limited conditions under the current anomaly interpretation structure. Prohibited use means that the center of gravity identification data is not allowed to enter the subsequent collaborative decision-making process under the current anomaly interpretation structure.

[0062] In this embodiment S3, based on the determined center of gravity decision availability state, a corresponding collaborative input constraint set is generated. The collaborative input constraint set is a set of constraint information used to restrict center of gravity identification data from entering the collaborative decision processing mode, and its target is the subsequent flight control collaborative decision input construction stage. When the center of gravity decision availability state is "allowed to use", the collaborative input constraint set includes constraint information for allowing center of gravity identification data to directly participate in collaborative decision processing. When the center of gravity decision availability state is "restricted to use", the collaborative input constraint set includes constraint information for restricting the participation scope, participation weight, or participation conditions of center of gravity identification data. When the center of gravity decision availability state is "prohibited to use", the collaborative input constraint set includes constraint information for prohibiting center of gravity identification data from entering collaborative decision processing, so that the collaborative decision input construction process can be executed under structured constraint conditions.

[0063] In this embodiment S4, the available state of the center of gravity decision is determined based on the comprehensive judgment result of each judgment condition in the pre-set judgment condition set. The judgment condition for determining consistency, the judgment condition for determining persistence, and the judgment condition for determining causal role jointly participate in the center of gravity information decision availability judgment processing to generate the corresponding available state of the center of gravity decision.

[0064] In this embodiment S3, the available state of the center of gravity decision is not determined independently by a single judgment condition, but is formed based on the comprehensive judgment result of each judgment condition in a pre-set set of judgment conditions. The comprehensive judgment is used to jointly process the judgment results generated by the judgment conditions for determining consistency, the judgment conditions for determining persistence, and the judgment conditions for determining causal role, so that the center of gravity information decision availability judgment can simultaneously reflect the anomaly explanation structure constraints, time stability, and causal role judgment results, thereby avoiding one-sided conclusions caused by a single judgment condition.

[0065] In this embodiment S3, during the comprehensive judgment process, all judgment conditions jointly participate in the determination of the availability of center of gravity information for decision-making. By coordinating and converging the results of each judgment condition under the constraints of the current target anomaly interpretation structure, a unique center of gravity decision availability state is generated. The center of gravity decision availability state serves as the final output of the determination of the availability of center of gravity information for decision-making, providing a unified basis for the generation of subsequent collaborative input constraint sets, so that whether and how the center of gravity identification data participates in collaborative decision-making can be stably determined under the comprehensive constraints of multi-dimensional judgment conditions.

[0066] In this embodiment S4, the flight control collaborative decision input is constructed under the constraints of the collaborative input constraint set. Specifically, based on the collaborative input constraint set, the participation mode of the center of gravity identification data in the flight control collaborative decision input is constrained to construct the flight control collaborative decision input.

[0067] In this embodiment S4, after determining the anomaly interpretation structure type and generating the collaborative input constraint set, the flight control collaborative decision input construction process is executed. The flight control collaborative decision input is a set of data inputs for subsequent collaborative decision processing. Its construction process simultaneously receives the anomaly state dataset, the target anomaly interpretation structure identifier, and the collaborative input constraint set as inputs. The anomaly state dataset is used to provide context information of the current anomaly state, the target anomaly interpretation structure identifier is used to indicate the interpretation path of the anomaly state, and the collaborative input constraint set is used to structurally constrain the participation mode of various types of data in the collaborative decision input.

[0068] In this embodiment S4, the flight control collaborative decision input is not an unconditional aggregation of the abnormal state dataset, but is constructed under the constraints of the collaborative input constraint set. Specifically, the collaborative input constraint set acts on the construction stage of the flight control collaborative decision input, constraining the participation of the center of gravity identification data in the collaborative decision input. Whether the center of gravity identification data is included, in what way it is included, and to what extent it is included in the collaborative decision input are all limited by the collaborative input constraint set, thereby avoiding the direct introduction of center of gravity information that lacks decision-making usability into the collaborative decision processing under abnormal states. When the collaborative input constraint set indicates that the center of gravity identification data is in a usable state, the center of gravity identification data is used as part of the collaborative decision input in subsequent collaborative decision processing; when the collaborative input constraint set indicates that the center of gravity identification data is in a restricted state... When center-of-gravity identification data participates in the construction of collaborative decision inputs, it is subject to constraints regarding the scope, degree, or conditions of participation. When the collaborative input constraint set indicates that center-of-gravity identification data is in a prohibited state, it is not included in the flight control collaborative decision inputs, allowing collaborative decision processing to be performed without relying on the center-of-gravity identification data. After completing the construction of the flight control collaborative decision inputs constrained by the collaborative input constraint set, collaborative decision processing is performed on the flight control collaborative decision inputs to generate a flight control collaborative decision dataset. The collaborative decision processing is used to perform collaborative processing at the decision level among multiple types of input information, but does not limit the specific control algorithm or execution method, thereby enabling the flight control collaborative decision dataset to be formed under the premise of following the anomaly interpretation structure constraints and input governance rules, providing a compliant data decision foundation for subsequent flight control-related processing.

[0069] Example 2: The present invention proposes an intelligent flight control cooperative system based on center of gravity identification, which is applied to the intelligent flight control cooperative method based on center of gravity identification proposed in Example 1. It includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement the intelligent flight control cooperative method based on center of gravity identification in Example 1.

[0070] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.

Claims

1. A collaborative intelligent flight control method based on center of gravity recognition, characterized in that, Includes the following steps: S1. Collect flight status data and center of gravity identification results of the aircraft during flight, perform data alignment and preprocessing, and generate an abnormal status dataset based on the preprocessed flight status data and center of gravity identification results. S2. Based on the abnormal state dataset, extract and construct an abnormal state feature set. Construct an abnormal interpretation judgment input feature set based on the abnormal state feature set. Perform an abnormal interpretation structure matching judgment between the abnormal interpretation judgment input feature set and predefined multi-class abnormal interpretation structure feature patterns to obtain the matching result and determine the target abnormal interpretation structure identifier. In S2, the abnormal state feature set is a set of data features used to characterize the abnormal occurrence process, including posture change pattern, center of gravity change feature, execution response consistency feature, and time correlation feature. The input features for anomaly interpretation determination are feature vectors generated from the anomaly state feature set according to a predetermined feature arrangement rule; the predefined multi-class anomaly interpretation structure feature patterns are predefined interpretation structure patterns, including the centroid-dominated interpretation structure feature pattern and the non-centroid-dominated interpretation structure feature pattern. The target anomaly interpretation structure identifier is used to indicate different target anomaly interpretation structures adopted in the current anomaly state. The target anomaly interpretation structure includes a center-of-gravity dominant interpretation structure and a non-center-of-gravity dominant interpretation structure. S3: Construct a center-of-gravity determination dataset based on the target interpretation structure identifier and the anomaly state dataset, and perform center-of-gravity information decision availability determination processing on the center-of-gravity determination dataset according to preset determination criteria to determine the center-of-gravity decision availability status of the center-of-gravity identification data under the current target anomaly interpretation structure, and generate a collaborative input constraint set based on the center-of-gravity decision availability status. In S3, the center-of-gravity determination dataset is a data set constructed based on the data related to center-of-gravity identification in the anomaly state dataset and combined with the target anomaly interpretation structure identifier. The preset judgment criteria are a set of pre-defined judgment conditions used to determine the availability of center of gravity information decisions; the center of gravity decision availability status is a status identifier used to characterize whether the center of gravity identification data is allowed to undergo subsequent collaborative decision processing; the collaborative input constraint set is a set of data generated based on the center of gravity decision availability status to constrain the center of gravity identification data from entering the collaborative decision processing; S4, based on the abnormal state dataset, target interpretation structure identifier, and collaborative input constraint set, construct the flight control collaborative decision input; The flight control collaborative decision input is processed to generate a flight control collaborative decision dataset; in S4, the flight control collaborative decision dataset is a data set used to carry the target anomaly interpretation structure identifier, the available state of the center of gravity decision, and the collaborative input constraint set.

2. The intelligent flight control cooperative method based on center of gravity recognition according to claim 1, characterized in that: In S2, the abnormal state feature set includes attitude change patterns, center of gravity change features, execution response consistency features, and time correlation features, reflecting the structured joint change relationships of the aircraft in the dimensions of attitude change, center of gravity change, execution response consistency, and time correlation under abnormal flight conditions; the abnormal interpretation judgment input features are the carrier of abnormal interpretation structure matching processing, which are reorganized and generated according to a predetermined feature arrangement rule; wherein, the predetermined feature arrangement rule is used to uniformly arrange and format the data of different feature dimensions in the abnormal state feature set.

3. The intelligent flight control cooperative method based on center of gravity recognition according to claim 2, characterized in that: In S2, the multiple types of anomaly interpretation structure feature patterns are also used to distinguish and model the dominant relationships between different feature dimensions in the anomaly state feature set. These include a center-of-gravity dominant interpretation structure feature pattern and a non-center-of-gravity dominant interpretation structure feature pattern, each corresponding to different anomaly state feature association relationships. Specifically, the center-of-gravity dominant interpretation structure feature pattern uses center-of-gravity change features as the main representation feature of the anomaly state, and its corresponding anomaly state feature association relationship shows a consistent dominant relationship between the center-of-gravity change feature and the posture change pattern. The non-center-of-gravity dominant interpretation structure feature pattern uses non-center-of-gravity change features as the main representation feature of the anomaly state, and its corresponding anomaly state feature association relationship shows that at least one of the posture change pattern, execution response consistency features, and time association features has a non-dominant relationship with the center-of-gravity change feature. The specific logic for distinguishing between the center-of-gravity dominant interpretation structure feature pattern and the non-center-of-gravity dominant interpretation structure feature pattern is as follows: based on the strength of the association relationship between each feature dimension in the anomaly state feature set, the degree of dominance of the center-of-gravity change feature in the anomaly state is determined, and the interpretation structure feature pattern type corresponding to the anomaly state is determined based on the degree of dominance.

4. The intelligent flight control cooperative method based on center of gravity recognition according to claim 3, characterized in that: In step S2, the anomaly explanation structure matching determination refers to the data processing process of comparing the anomaly explanation determination input features with the corresponding relationships of the multiple anomaly explanation structure feature patterns, used to determine the matching relationship between the anomaly state and different anomaly explanation structure feature patterns; the specific method of the anomaly explanation structure matching determination is as follows: based on the anomaly explanation determination input features, the association matching value between the anomaly explanation determination input features and each anomaly explanation structure feature pattern is calculated respectively, and a matching result is formed for each anomaly explanation structure feature pattern based on the association matching value; the matching result is a data set used to characterize the degree of matching between the anomaly explanation determination input features and each anomaly explanation structure feature pattern; The target anomaly interpretation structure identifier is the type identifier of the anomaly interpretation structure feature pattern that corresponds to the associated matching value in the matching result and satisfies the preset judgment rule.

5. The intelligent flight control cooperative method based on center of gravity recognition according to claim 4, characterized in that: In S3, the preset judgment basis is a set of pre-set judgment conditions for performing the judgment of the availability of center of gravity information decision. The set of pre-set judgment conditions includes judgment conditions for judging the consistency between the center of gravity change features and the target anomaly interpretation structure identifier, judgment conditions for judging the persistence of the center of gravity change features in the abnormal state, and judgment conditions for judging the causal role of the center of gravity change features in the process of anomaly occurrence.

6. The intelligent flight control cooperative method based on center of gravity recognition according to claim 5, characterized in that: In S3, the center of gravity information decision availability determination processing is performed under the constraints of the target anomaly interpretation structure corresponding to the target anomaly interpretation structure identifier. Specifically, it involves determining the set of judgment conditions corresponding to the target anomaly interpretation structure identifier, and performing center of gravity information decision availability determination processing on the center of gravity determination dataset under the constraints of the set of judgment conditions to obtain the center of gravity decision availability status.

7. The intelligent flight control cooperative method based on center of gravity recognition according to claim 6, characterized in that: In step S3, the available state of the center of gravity decision includes allowed use, restricted use, and prohibited use. The available state of the center of gravity decision is used to characterize the degree to which the center of gravity identification data is allowed to enter the collaborative decision processing under the current target anomaly interpretation structure. Based on the available state of the center of gravity decision, a corresponding collaborative input constraint set is generated to constrain the center of gravity identification data to enter the subsequent collaborative decision processing. The collaborative input constraint set includes constraint information for allowing the center of gravity identification data to directly participate in the collaborative decision processing, constraint information for limiting the scope or weight of participation of the center of gravity identification data, and constraint information for prohibiting the center of gravity identification data from entering the collaborative decision processing.

8. The intelligent flight control cooperative method based on center of gravity recognition according to claim 7, characterized in that: The available state of the center of gravity decision is determined based on the comprehensive judgment result of each judgment condition in the pre-set judgment condition set. The judgment condition for determining consistency, the judgment condition for determining persistence, and the judgment condition for determining causal role jointly participate in the center of gravity information decision availability judgment processing to generate the corresponding available state of the center of gravity decision.

9. The intelligent flight control cooperative method based on center of gravity recognition according to claim 8, characterized in that: In S4, the flight control collaborative decision input is constructed under the constraints of the collaborative input constraint set. Specifically, based on the collaborative input constraint set, the participation mode of the center of gravity identification data in the flight control collaborative decision input is constrained to construct the flight control collaborative decision input.

10. A smart flight control cooperative system based on center of gravity recognition, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: The processor executes a computer program to implement the intelligent flight control cooperative method based on center of gravity recognition as described in any one of claims 1-9.

Images