Disturbance-oriented flight control system information flow approximate modeling method

Abstract

The invention discloses a disturbance-oriented flight control system information flow approximate modeling method, which constructs an information flow meta-model and takes the information flow meta-model as a modeling norm, and the obtained field model of a flight control system can be expressed by separating from a bottom layer implementation logic, so that the dependence on a specific flight control platform is eliminated, and the modeling efficiency is improved. The portability and the cross-platform deployment capability of the modeling method are improved; symbolized unified representation is carried out on disturbance faced by the flight control system, the limitation that an existing fault-tolerant control method depends on a type specific modeling hypothesis is effectively broken through, and the universality and universality of disturbance modeling are remarkably improved; by establishing a chained propagation analysis mechanism for propagation of disturbance from an input node of a domain model to control output, explicit tracking and quantitative analysis can be performed on a transmission path of the disturbance in the domain model, and interpretable modeling of abnormal behaviors is enhanced. The system is widely applied to the technical field of unmanned aerial vehicles.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned aerial vehicles, and particularly relates to a disturbance-oriented flight control system information flow approximate modeling method. BACKGROUND

[0002] In recent years, with the continuous innovation of technology and the continuous reduction of manufacturing cost, unmanned aerial vehicles (UAV) have been widely used in military and civilian fields. The flight control system, also known as the flight control system, is the core component of the unmanned aerial vehicle, which is responsible for processing data from various sensing components, generating and processing unmanned aerial vehicle telemetry and remote control data, and has an important influence on the stability of the unmanned aerial vehicle flight, the reliability of data transmission and the real-time performance of operation execution. However, the flight control system is subject to various types of disturbances in complex environments, including environmental disturbances caused by natural factors (such as wind speed fluctuations, temperature changes), component failures within the system (such as sensor failures, actuator degradation), and network attacks implemented by malicious attackers (such as data injection, signal spoofing, etc.). The above disturbances may directly or indirectly affect the flight stability, navigation accuracy and task reliability of the unmanned aerial vehicle.

[0003] In order to cope with these disturbances, some unmanned aerial vehicle flight control systems apply fault-tolerant control mechanisms, the basic idea of which is to model specific disturbance types and introduce compensation mechanisms in the control logic. Generally speaking, fault-tolerant controller design can be divided into two categories, namely passive fault-tolerant control and active fault-tolerant control. Passive fault-tolerant control tolerates faults by abstracting disturbance effects as bounded uncertainties, without changing the original control structure, thereby improving the robustness of the system. In contrast, active fault-tolerant control relies on real-time fault detection and diagnosis to reconfigure the controller to respond to disturbances or faults. Although these methods can construct effective model representations in specific disturbance scenarios, they still have the following two key limitations in terms of disturbance characterization and structural abstraction. First, their disturbance modeling relies on idealized assumptions for specific types of disturbances, limiting their generalization of heterogeneous disturbances (especially network attacks). Second, their tight coupling with specific platform control architectures hinders structural generalization and increases the overhead of cross-platform deployment. These limitations highlight the need for a modeling and analysis method that can support generalized disturbance characterization and structural abstraction independent of specific platform control architectures. SUMMARY

[0004] In view of the technical problems of the current disturbance-oriented flight control system modeling technology, such as insufficient generalization of heterogeneous disturbances, high dependence on specific platform control architectures, and the like, the purpose of the present application is to provide a disturbance-oriented flight control system information flow approximate modeling method.

[0005] In one aspect, the embodiment of the present application comprises a disturbance-oriented flight control system information flow approximate modeling method, the disturbance-oriented flight control system information flow approximate modeling method comprising the following steps:

[0006] Taking an information flow meta-model as a modeling paradigm, a domain model of the flight control system is established;

[0007] Symbolically and uniformly representing disturbances faced by the flight control system to obtain heterogeneous disturbance information;

[0008] Embedding the heterogeneous disturbance information into the domain model;

[0009] Performing disturbance propagation analysis on the domain model.

[0010] Further, the information flow meta-model comprises a unit class element, an information class element and a connection class element; the unit class element comprises a functional module, a node and a transfer function; the node is a minimum information processing unit for information flow execution, and the node converts input variables into output variables through the transfer function.

[0011] Further, taking an information flow meta-model as a modeling paradigm, the domain model of the flight control system comprises:

[0012] Dividing the flight control system into functional modules to determine a plurality of functional modules;

[0013] Identifying a key code block from the functional modules, and taking the key code block as a key node;

[0014] Classifying the key node into a measurement node, an execution node, a control node or an aggregation node;

[0015] Obtaining related variables and a transfer function of the key node;

[0016] According to the functional modules, the key node, the related variables and the transfer function, the domain model is composed.

[0017] Further, the flight control system is divided into functional modules to determine a plurality of functional modules, comprising:

[0018] Obtaining a configuration file of the flight control system;

[0019] According to the configuration file, determining a compilation dependency relationship between each module in the flight control system;

[0020] According to the compilation dependency relationship, combining and splitting each module in the flight control system to obtain each functional module.

[0021] Further, the key code block is identified from the functional modules, comprising:

[0022] Tracking and analyzing data dependency relationship of code in the function module by using reverse slicing method, to determine the key code block.

[0023] Further, the tracking and analyzing data dependency relationship of code in the function module by using reverse slicing method, comprising:

[0024] Obtaining the output variable in the function module as a slice target;

[0025] Taking the output variable as a starting point, reverse tracking along the data dependency path until encountering data from upstream or external instructions to terminate;

[0026] Marking the data dependency path that has been reverse tracked;

[0027] Extracting the code block corresponding to the marked data dependency path as the key code block.

[0028] Further, the obtaining the related variables and transfer function of the key node, comprising:

[0029] Obtaining the input variable and the output variable of the key node;

[0030] Determining the transfer function according to the input variable and the output variable.

[0031] Further, the symbolic uniform representation of the disturbance faced by the flight control system to obtain heterogeneous disturbance information, comprising:

[0032] Expressing the disturbance faced by the flight control system as the input variable or transfer function deviation of the disturbed node, and expressing the influence generated by the disturbance as the symbolic deviation term in the output variable of the measurement node, to obtain the heterogeneous disturbance information

[0033] Further, the embedding the heterogeneous disturbance information into the field model, comprising:

[0034] According to the symbolic deviation term in the output variable in the heterogeneous disturbance information, configuring the measurement node in the field model.

[0035] Further, the disturbance propagation analysis of the field model, comprising:

[0036] Obtaining an injection attack;

[0037] Modeling the injection attack as a disturbance to the flight control system, and inputting the field model;

[0038] Obtaining the chain Jacobian matrix between each key node in the field model;

[0039] According to the chain Jacobian matrix, the propagation process of the disturbance between each key node is determined.

[0040] The beneficial effects of the present application are: the disturbance-oriented flight control system information flow approximation modeling method in the embodiment can express the domain model of the flight control system by constructing an information flow meta-model and taking it as a modeling paradigm, thereby breaking away from the dependence on specific flight control platforms and improving the portability and cross-platform deployment capability of the modeling method; by symbolically and uniformly representing the disturbances faced by the flight control system, the limitations of existing fault-tolerant control methods that rely on type-specific modeling assumptions are effectively broken through, and the universality and generality of disturbance modeling are significantly improved; by establishing a chain propagation analysis mechanism for disturbances from the input nodes of the domain model to the control output, the transmission path of the disturbance in the domain model can be explicitly tracked and quantitatively analyzed, and the explainability of abnormal behavior modeling is enhanced. BRIEF DESCRIPTION OF DRAWINGS

[0041] Figure 1 is a schematic diagram of the information flow meta-model in the embodiment;

[0042] Figure 2 is a schematic diagram of the measurement node, execution node, control node and aggregation node contained in the information flow meta-model in the embodiment;

[0043] Figure 3 is a schematic diagram of the steps of the disturbance-oriented flight control system information flow approximation modeling in the embodiment;

[0044] Figure 4 is a schematic diagram of the framework of the disturbance-oriented flight control system information flow approximation modeling method in the embodiment;

[0045] Figure 5 is a schematic diagram of the internal information flow architecture of the domain model established based on the PX4 Autopilot flight control system in the embodiment;

[0046] Figure 6 is a schematic diagram of the performance comparison between PX4 and the domain model in the nominal scenario during the simulation experiment in the embodiment.

[0047] Figure 7 is a schematic diagram of the performance comparison between PX4 and the domain model in the wind interference scenario during the simulation experiment in the embodiment.

[0048] Figure 8 is a schematic diagram of the performance comparison between PX4 and the domain model in the actuator failure scenario during the simulation experiment in the embodiment.

[0049] Figure 9 is a schematic diagram of the performance comparison between PX4 and the domain model in the network attack scenario during the simulation experiment in the embodiment. DETAILED DESCRIPTION

[0050] I. Shortcomings in the field of UAV flight control system modeling and disturbance characterization and solutions

[0051] (1) Lack of generalization in disturbance characterization

[0052] Current fault-tolerant control techniques for UAV flight control systems often rely on targeted disturbance modeling assumptions, such as treating external wind disturbances as Gaussian noise or modeling sensor failures as fixed biases. This type-specific and static disturbance modeling approach is difficult to characterize heterogeneous disturbances in reality, especially highly unstructured or unpredictable cyber attacks. In addition, due to the lack of a unified disturbance modeling framework, this method is difficult to systematically analyze the propagation path of heterogeneous disturbances in the control system, thereby limiting its modeling and evaluation capabilities for system disturbance response mechanisms.

[0053] (2) Lack of generalization in control architecture

[0054] Current fault-tolerant control methods are usually deeply nested in specific platform control architectures, and their control logic is usually highly coupled with platform implementation, making it difficult to migrate across platforms. If a fault-tolerant strategy needs to be deployed to a new UAV platform, it often needs to restructure the control law, adjust the feedback structure, or embed additional modules. This not only increases the engineering cost of deployment, but also affects the practicality and portability of the control system in resource-constrained devices.

[0055] (3) Difficulty in selecting modeling granularity

[0056] UAV flight control systems such as PX4 Autopilot, ArduPilot, etc. usually contain hundreds of adjustable parameters and multiple levels of control logic, with complex internal information flow interactions, making it difficult to model comprehensively through traditional methods. Therefore, determining the appropriate abstraction granularity is crucial. If the abstraction granularity is too coarse, key control dependencies may be missed, affecting modeling accuracy; while too fine abstraction may cause state space explosion, making computation infeasible. Therefore, it is urgent to build a flexible and adjustable approximation modeling method that preserves the core control semantics of the flight control system while enabling controllable abstraction and analysis support for key information flows.

[0057] Information flow analysis provides a new research direction to address the limitations of existing methods in terms of structural generalization and disturbance characterization. Unlike traditional fault-tolerant methods that intervene with low-level control logic, information flow analysis abstracts the system based on the way information propagates and transforms among system components. This higher-level modeling perspective avoids dependence on specific platform control implementations, thus achieving structural decoupling and enhancing modeling portability. In addition to supporting structural abstraction, this modeling perspective also supports the general characterization of heterogeneous disturbances as anomalies in the system information flow, despite their diverse origins and manifestations. For example, component failures can disrupt information flow through data loss, environmental disturbances can interfere with information flow through random noise, and cyber attacks such as sensor spoofing can manipulate or destroy information flow. This unified perspective provides an easy-to-analyze framework for characterizing traditional disturbances and complex adversarial cyber attacks in a single modeling paradigm.

[0058] II. Information Flow Meta-Model

[0059] The information flow meta-model can provide a generalizable abstraction method for system information interaction. It provides a unified representation of heterogeneous disturbances and abstracts internal control logic at an appropriate granularity. In addition, it supports extension to domain-specific models, enabling traceable analysis of the impact of disturbances on control behavior while preserving the control semantics of flight control systems. Therefore, in this embodiment, the information flow meta-model is used as the modeling paradigm for the UAV flight control system.

[0060] The following introduces the relevant knowledge of the information flow meta-model, including element classification, and key elements such as nodes, information flows, and transfer functions in the information flow meta-model.

[0061] (I) Element classification of the information flow meta-model

[0062] The information flow meta-model includes multiple elements. According to the characteristics of the control system, the information flow meta-model classifies system elements into three categories: unit class, information class, and connection class.

[0063] (1) Unit class: includes elements such as tasks (e.g., hovering, path planning), entities (e.g., UAV, unmanned vehicle), modules (e.g., control module, dynamics module), nodes (e.g., measurement node, execution node), and transfer functions (e.g., P-PID control function). These elements define the functional and structural units of the system.

[0064] (2) Information class: includes elements such as instructions (e.g., desired position, desired heading), constants (e.g., gravitational acceleration, UAV mass), variables (e.g., measured velocity, measured angle), constraints (e.g., maximum tilt angle, actuator limits), and disturbances (e.g., component failure, cyber attack). These elements constitute the core data that control and state information passes between nodes.

[0065] (3) Connection: defines the interaction mechanism within the system, including the information flow, association, communication, and dominance between the unit class elements.

[0066] According to the classification, the membership relationship between elements is established, and finally the reference meta-model for information flow analysis, i.e., the information flow meta-model, is constructed, as shown in Figure 1

[0067] (II) Key elements in the information flow meta-model

[0068] (1) Node classification

[0069] In the information flow meta-model, a node is defined as the smallest information processing unit, which converts input variables into output variables through a transfer function. These nodes constitute the basic unit blocks of information flow execution, and the interconnection and combination of node-level transformations together define the control logic of the system. As shown in Figure 2 , according to the role in the control system, the nodes in the information flow meta-model can be divided into four functional types: measurement nodes, execution nodes, control nodes, and aggregation nodes.

[0070] Among them, the measurement node is shown as the orange square in the upper left corner of Figure 2 , which can sense physical quantities and convert them into digital information as the source of information flow. The execution node is shown as the green square in the upper right corner of Figure 2 , which can receive control commands and issue instructions to change the operating state of the actuator. The control node is shown as the blue square in the lower left corner of Figure 2 , which can convert input into control output through a pre-defined control algorithm. The aggregation node is shown as the red square in the lower right corner of Figure 2 , which can integrate multiple input streams or distribute data to multiple outputs, thereby realizing information fusion and parallel propagation within the system.

[0071] (2) Information flow

[0072] Referring to Figure 2 , the dependency relationship between nodes can be represented by the direction of information flow and modeled by an association matrix:

[0073]

[0074] where T is the association matrix, N m is the number of nodes. If the input of node j depends on the output of node k, then t jk =1; otherwise t jk =0.

[0075] In the flight control system, the control signal is based on the estimated measurement state x m ​The generated control signal of the kth node is denoted as uk k The control output vector u can be defined as:

[0076]

[0077] If the structure and function of the control system are known, the information flow from the estimated measurement x m to the control output u can be modeled as the following nonlinear mapping relationship:

[0078] x m (N)→u(N)=Φ(x m (N)) (3)

[0079] where Φ represents the internal information conversion process determined by the control logic.

[0080] (3) Node transfer function

[0081] The transfer function is a basic representation of the dynamic characteristics of the node, which is used to describe how each node processes the input variables into the corresponding output. For the ith node, let the input variables be and the output variables be The transfer function of this node is defined as:

[0082]

[0083] However, when the node is affected by disturbances, its input or transfer function may change, and the disturbed output can be represented as:

[0084]

[0085] where Generally represents the input deviation caused by environmental disturbance, and δF i Generally represents the change in transfer function caused by component failure. Network attacks may interfere with input and transfer function alone or simultaneously. For example, data injection attacks usually cause while code injection attacks produce δF i Therefore, considering that network attacks may introduce any form of disturbance with uncertainty, discontinuity or nonlinearity, no structural assumptions are imposed on and δF i

[0086] III. Disturbance-oriented information flow approximation modeling method for flight control system

[0087] ​To solve the problem of insufficient generalization ability in disturbance characterization and structure abstraction of traditional fault-tolerant control methods, in the embodiment, based on the information flow meta-model, a disturbance-oriented flight control system information flow approximation modeling method is provided. This method abstracts the control logic of the flight control system into composable information flow, realizes structural generalization and supports unified representation of heterogeneous disturbances. Based on the instantiation of the domain model based on this method, the propagation path of the disturbance in the control logic and its influence on the system behavior can be traced and analyzed, so as to approximate the dynamic response of the real flight system under the disturbance scenario.

[0088] Referring to Figure 3 , the disturbance-oriented flight control system information flow approximation modeling method comprises the following steps:

[0089] S1. Establish a domain model of the flight control system based on the information flow meta-model as the modeling paradigm;

[0090] S2. Symbolically represent the disturbances faced by the flight control system to obtain heterogeneous disturbance information;

[0091] S3. Embed the heterogeneous disturbance information into the domain model;

[0092] S4. Perform disturbance propagation analysis on the domain model.

[0093] In the embodiment, Figure 3 The framework of the disturbance-oriented flight control system information flow approximation modeling method shown in Figure 4 The framework mainly consists of three parts: first, in step S1, the information flow meta-model (meta-model for short) is used as an abstract modeling paradigm to define three key elements including unit class, information class and connection class; then, based on the modeling paradigm of the meta-model, the specific domain model of the flight control system is constructed through steps such as functional module division, key node identification and variable and transfer function mapping; then, step S2 is executed, according to the definition of the information flow meta-model, the disturbances (including network attacks, component failures and environmental disturbances, etc.) faced by the flight control system are uniformly represented as measurable information flow deviations of the system, i.e. heterogeneous disturbance information is obtained, and finally steps S3-S4 are executed to analyze how the disturbance propagates along the information flow and changes the control output in the domain model.

[0094] The information flow domain model of the flight control system to be established in step S1 (hereinafter referred to as the domain model) is equivalent to a specific instantiation of the information flow meta-model. The domain model extends the meta-model to enable fine-grained representation of control logic and information dependency relationships in actual platforms.

[0095] Since the system elements in the meta-model are divided into three categories: unit class, information class and connection class. In the construction of the specific domain model for the UAV flight control system, the invention focuses on the key elements that can accurately reflect the control logic and information dependency. These key elements include modules, nodes and transfer functions in the unit class, and variables in the information class. Therefore, the construction of the domain model in step S1 mainly includes three stages: functional module division, key node identification and variable-transfer function mapping.

[0096] Based on the above principles, in this embodiment, when performing step S1, that is, taking the information flow meta-model as the modeling paradigm, the step of establishing the domain model of the flight control system, the following steps can be performed:

[0097] S101. Dividing the functional modules of the flight control system to determine a plurality of functional modules;

[0098] S102. Identifying key code blocks from the functional modules, and taking the key code blocks as key nodes;

[0099] S103. Obtaining related variables and transfer functions of the key nodes;

[0100] S104. Composing the domain model according to the functional modules, key nodes, related variables and transfer functions.

[0101] Among them, step S101 belongs to the functional module division stage, step S102 belongs to the key node identification stage, and step S103 belongs to the variable-transfer function mapping stage.

[0102] (1) Step S101: Functional module division

[0103] In the UAV flight control system, each module is responsible for performing a specific control function, and the information flow between modules determines the control behavior of the system. Therefore, effective module division constitutes the basis of domain model construction. Reasonable division can reduce unnecessary inter-module dependencies and improve information traceability.

[0104] To reveal the information flow relationship between modules, in this embodiment, the system construction configuration file CMakeLists.txt is analyzed, and the compilation dependency relationship of the custom modules in the system architecture is extracted. Modules that frequently exchange data or are tightly coupled in calculation are divided into the same functional module, while parts that are functionally independent and have limited interaction are divided into different modules. For example, in the PX4 default architecture, mc_pos_control and mc_att_control are implemented as two independent modules, but there is a close information coupling relationship between them, that is, the expected acceleration output by the former is used as the input of the latter. Therefore, in order to reduce the modeling complexity, the two can be considered as one module in the domain model.

[0105] (2) Step S102: Critical Node Identification

[0106] After the completion of the function module partitioning, the next step is to identify the critical nodes, i.e. the code blocks that have significant impact on the system stability, task execution or information integrity during the system operation. To identify the critical code blocks in the system, in this embodiment, the backward slicing method in program static analysis is adopted for analysis through data dependency tracking. The core idea of the backward slicing method is to take a certain node (e.g. an actuator) output variable (e.g. actuator_controls) as the starting point, perform backward tracking along the data dependency path, identify the upstream computing logic, and terminate until the data or external instruction input from the upstream of this node (e.g. the upstream of the actuator - the sensor) is encountered. All data dependency paths related to the target variable will be marked, and the corresponding code blocks will be extracted for analysis. According to their defined functional roles in the meta-model paradigm, these code blocks will be classified as aggregation nodes, control nodes, measurement nodes or execution nodes.

[0107] In this embodiment, the algorithm included in the flight control system is taken as an example to illustrate the critical node identification in step S102.

[0108] Algorithm 1 implements the backward slicing process based on the target variable tv and the flight control source code SC, for identifying the code blocks that affect the calculation of the target variable. These code blocks are critical code blocks, referred to as critical nodes in the domain model.

[0109] For algorithm 1, first, the variable analysis list W, the visited variable set V and the output set RC are initialized (lines 2-4). Then, the variables v are taken from W one by one, and if v has not been visited, it is marked and further analyzed (lines 5-9). The assignment statement for the variable v is found through the FindAssignment(v, SC) function (line 10), and the source variable and the code block where it is located are extracted (lines 11-12). If the source variable comes from the sensor topic or the navigator, the tracking is terminated; otherwise, the source variable is re-added to W for further analysis (lines 13-15). Finally, the algorithm returns the set RC containing all critical nodes (line 22).

[0110]

[0111]

[0112] (3) Step S103: Variable - Transfer Function Mapping

[0113] In the reverse slicing process, the input and output variables of the key nodes and their corresponding transfer functions are mapped into the domain model synchronously to define their information dependency. As shown in Algorithm 1, starting from line 16, each identified key node (i.e., key code block) is traversed. The assignment statements within each node are analyzed to extract the output variable on the left side and the input variable on the right side, which are stored in the variable set Var (lines 17-19). Subsequently, the internal computation logic of each node is parsed to derive the corresponding transfer function based on the input-output relationship, which is stored in the transfer function set TF (lines 20-21). Finally, the variable set Var and the transfer function set TF are returned as the algorithm output (line 22).

[0114] Taking the control node in the PX4 system (a specific flight control system) as an example, the input variables of the yaw control node include att_sp.yaw_body, yaw_sp_move_rate, yaw_body, and yaw_move_rate, and the output variable is actuator_controls[2]. The transfer function corresponding to this node can be extracted as a cascade P-PID equation. In addition to input and output variables, each key node can also be associated with other types of information elements, such as constants, constraints, and instructions. These elements are also integrated into the domain model to complete its information dependency representation.

[0115] (4) Specific examples of the domain model established in steps S101-S104

[0116] In this embodiment, based on the PX4 Autopilot, a specific flight control system, a specific instantiation of the domain model established by executing steps S101-S104 on this flight control system is demonstrated. Specifically, according to the classification principles of the meta-model and the aforementioned domain model construction methods such as steps S101-S104, the functional modules, key nodes, related variables, and transfer functions are described in detail.

[0117] (4.1) Functional modules and key nodes

[0118] Figure 5 The internal information flow architecture of the instantiated domain model is demonstrated, which mainly consists of four modules: navigation module, control module, dynamics module, and measurement module. These modules are connected in a closed-loop structure: first, the navigation module converts task instructions into control set values; then, the control module calculates actuator instructions based on the set values and sensor feedback; the dynamics module receives these instructions and executes them through the motor; finally, the measurement module collects sensor data to provide feedback for closed-loop control. The above information flow sequence constitutes the basic operation logic of the system.

[0119] Based on the inverse slicing method used in step S102, the system output variable actuator_controls is chosen as the slicing target in this embodiment.

[0120] Figure 5 The identified key nodes are shown and classified into four categories according to the meta-model. The aggregation nodes include the navigator, MAVLink receiver, and Extended Kalman Filter (EKF), which are used to integrate task inputs, communicate data, and fuse multi-sensor measurements to generate state estimates, respectively. The control nodes cover the six-degree-of-freedom P-PID control nodes and the control allocation node. The execution nodes include the motor execution node and six motion nodes related to translational and rotational dynamics. The measurement nodes include the gyroscope, accelerometer, magnetometer, GPS, and barometer, which provide raw inertial and environmental data for state estimation.

[0121] (4.2) Node variables and transfer functions

[0122] The domain model is finally formed by specifying the transformation behavior (i.e., transfer functions) of the key nodes. These functions describe how the input variables are processed to generate the output variables, embodying the basic control logic of the system. In this embodiment, four representative nodes are analyzed: the P-PID control node, control allocation node, motor execution node, and motion node.

[0123] Taking the height control as an example, the P-PID control node adopts a cascade control structure: the outer loop input is the desired height z d (t) and the measured height z(t), and the inner loop input is the desired vertical velocity v zd (t) and the measured vertical velocity v z (t), and the output variable is the total thrust command u1(t). The outer loop uses a proportional control function to convert the height error into the desired vertical velocity:

[0124] v zd (t) = K P,z · e z (t) = K P,z · (z d (t) - z(t)) (6)

[0125] where K P,z is the proportional gain, and the calculated v zd (t) is used as the set value for the inner loop velocity control. The inner loop applies a PID control function to generate the thrust command:

[0126]

[0127] where, the PID gains K P, K I , K D respectively determine the proportional, integral and derivative response characteristics. The control allocation node maps the overall control command to individual actuator commands. The input variables are four control quantities: total thrust ui and three directional moment commands u2 (roll), u3 (pitch), u4 (yaw), and the output variables are the four motor speeds

[0128] For the symmetric cross-shaped quadrotor, the allocation relationship is as follows:

[0129]

[0130] where K T is the thrust coefficient, K d is the drag coefficient, and l is the arm length. The above transformation ensures that the control commands are properly allocated to the motors according to the vehicle geometry.

[0131] The motor execution node and the motion node in the dynamics module describe how the actuators generate actual motion. The input variable for each motor node is the motor speed The output variable is the corresponding thrust f i , which is given by:

[0132]

[0133] The total thrust drives the vehicle to produce translational motion (lateral, longitudinal, vertical). The translational motion node receives the thrust f i , the current attitude angles φ, θ, ψ, and the linear velocities v x , v y , v z as inputs, and the output variable is the linear acceleration The calculation formula is as follows:

[0134]

[0135] where m is the mass of the vehicle and g is the gravitational constant. The above function describes how the thrust is converted into the inertial system acceleration considering the attitude and aerodynamic drag.

[0136] The rotational motion node describes the dynamics behavior in the roll, pitch, yaw directions, with the thrust f i and angular velocity p, q, r as inputs, and the angular acceleration is calculated using the Euler dynamics formula as follows:

[0137]

[0138] where I xx , I yy , I zz, , represent the moment of inertia of the aircraft in three directions, and s is the distance from the motor to the aircraft's center of mass. These functions describe how the aircraft generates angular acceleration based on the thrust difference between the motors.

[0139] The domain model constructed using the aforementioned transfer function clearly characterizes the entire process from control commands to actuator signals, and then to changes in physical state. This formal structure provides a structural basis for subsequent analysis of disturbance propagation.

[0140] To enhance the modeling and generalization capabilities of flight control systems when facing heterogeneous disturbances, this embodiment proposes a symbolic disturbance modeling and analysis method. Specifically, in step S2, which involves symbolically and uniformly representing the disturbances faced by the flight control system to obtain heterogeneous disturbance information, the heterogeneous disturbances are first uniformly represented. Then, step S3 is executed to embed the heterogeneous disturbances into the flight control domain model and analyze the propagation path and impact mechanism of the disturbances in the control information flow.

[0141] (5) Steps S2-S3: Unified representation of heterogeneous perturbations and embedding them into the domain model

[0142] In the meta-model, each key information processing unit within the control system is abstracted as a node, whose definition includes input variables. Transfer function F i And by formula The calculated output variable is shown in formula (4). Assuming the system consists of L serial nodes, then in an ideal scenario, the measured output x m (t) is derived from the expected value x d (t) and the output x from the previous time step m (t-1) The combined result after transforming all nodes in sequence:

[0143]

[0144] According to formula (5), heterogeneous disturbances in the system may affect input variables or internal transfer functions. When network attacks, component failures, and environmental interference coexist, node N... i The disturbed output can be expressed as:

[0145]

[0146] To analyze the resulting bias, a first-order Taylor expansion of the formula is performed:

[0147]

[0148] in, denotes the local Jacobian matrix. The above equation quantitatively describes the influence of input perturbation and transfer function variation on the output of a single node. This influence is then propagated downstream along the information flow. According to the chain rule, the influence of a perturbation on the output of node N i downstream to the measurement node N L , the deviation caused by the perturbation at the measurement output is:

[0149]

[0150] Substituting equation (14) into equation (15), we have:

[0151]

[0152] where, denotes the cumulative sensitivity of the final output x to the input perturbation m , and describes the propagated influence of the function variation δF i .

[0153] When multiple nodes are simultaneously affected by a perturbation, the total deviation ε of the measurement output can be represented by the superposition effect of all perturbed nodes as:

[0154]

[0155] where, denotes the high-order residual term neglected in the first-order approximation.

[0156] Equation (17) uniformly characterizes heterogeneous perturbations, and its core idea is to represent the influence of a perturbation as a symbolic deviation term ε in the measurement output. This deviation term serves as a canonical form for embedding heterogeneous perturbations into a specific information flow domain model, enabling these perturbations to be modeled and calculated in a formalized manner within the model.

[0157] Therefore, in step S2, the perturbations faced by the flight control system are represented as input variables or transfer function deviations of the perturbed nodes, and the influence of the perturbations is represented as a symbolic deviation term in the output variable of the measurement node, obtaining heterogeneous perturbation information; in step S3, the output variable of the measurement node in the domain model is configured using the symbolic deviation term, thereby embedding the heterogeneous perturbation information into the domain model.

[0158] (6) Step S4: Perturbation propagation analysis in the domain model

[0159] The disturbance unified representation method proposed in step S2 makes any disturbance source be symbolically mapped as a measurement bias term, thereby realizing the closed-loop traceability from the disturbance source to the control input. Compared with the traditional fault-tolerant control method, which mainly faces component faults and environmental disturbances, the method further integrates network attacks into a unified and computable modeling structure, significantly expanding its scope of application.

[0160] To verify the computational feasibility and scalability of the representation method, in this embodiment, network attacks as a representative scenario are taken as an example of disturbances that the UAV flight control system may face, to illustrate how to instantiate the accelerometer injection attack in the PX4 system, a specific flight control system, in the domain model, and approximately calculate its impact on the downstream control output. In this scenario, the attacker injects fake data into the accelerometer measurement stream. The attack is modeled as an input bias at the acceleration measurement node N1, which is mathematically expressed as:

[0161]

[0162] where δa(t) represents the tampered acceleration vector. The disturbance is then propagated to the EKF node N2, whose output is the estimated state vector including position, velocity, attitude angle and angular velocity:

[0163]

[0164] According to formula (19), the propagation process of the disturbance is represented as:

[0165]

[0166] where J 1:2 represents the chain Jacobian matrix from the accelerometer node N1 to the output of the EKF node N2. Since the transfer function of the node does not change, δF1=0. The chain Jacobian can be further decomposed as:

[0167]

[0168] where F1 represents the transfer function of the accelerometer, which can be approximately represented as a unit mapping, and its derivative is a unit matrix. F2 is the transfer function of the EKF node, and the partial derivative of the acceleration input gives the Jacobian matrix Therefore, the final disturbance term can be expressed as:

[0169] ε(t) = J EKF · δa(t) (22)

[0170] The following is an example of the height control loop of PX4, evaluating the impact of the disturbance on the control output. The control loop adopts the structure of cascading outer loop position controller and inner loop velocity controller, and the controller relies on z(t) and v z (t) in the state estimation, corresponding to the 3rd and 6th elements of x m (t), respectively. After being affected by the disturbance, the state estimation value becomes:

[0171]

[0172] The outer loop controller is used to calculate the expected vertical velocity, and its transfer function (see equation (6)) is updated as:

[0173]

[0174] The inner loop controller is used to calculate the thrust command, and its transfer function (see equation (7)) is updated as:

[0175]

[0176] Thus, the deviation of the thrust command output affected by the disturbance is:

[0177]

[0178] Substituting the symbolic expression of the disturbance term (see equation (22)), we get:

[0179]

[0180] The final control deviation term is expressed as:

[0181]

[0182] where, and are the 3rd and 6th rows of J EKF , respectively.

[0183] The above derivation formally describes the propagation process of sensor injection attacks along the information flow path, and derives the analytical expression of its impact on the control output. This shows that the domain model obtained by steps S1-S3 is not only suitable for environmental disturbances and component failures, but also can be used for symbolic modeling and traceable analysis of disturbances that unmanned aerial vehicle flight control systems may be subjected to, such as network attacks, thereby effectively realizing more generalizable disturbance representation.

[0184] IV. Simulation Experiment

[0185] In this embodiment, a series of simulation experiments are designed and carried out to compare the mainstream open source flight control system PX4Autopilot with the flight control system information flow field model, so as to verify the effectiveness of the modeling method proposed in the application.

[0186] (1) Experimental setup

[0187] The flight control system used in this experiment is integrated by PX4 Autopilot, Gazebo simulator and Robot Operating System (ROS), and runs on a virtual machine configured with 64-bit Ubuntu, which is equipped with a quad-core 2.6GHz processor and 8GB of memory. The unmanned aerial vehicle model used in PX4 simulation is Iris quadrotor, and its detailed parameters are shown in Table 1.

[0188] Table 1 Parameters of Iris quadrotor

[0189]

[0190]

[0191] By using SVF 3.0 tool, cross-process static backward slicing analysis is performed on the intermediate representation of LLVM 13.0 to build the field model. The instantiated field model is implemented based on Python 3.9, and the configured PID gain parameters are listed in Table 2. All simulation experiments are completed on a desktop computer equipped with an Intel Core i7-10700 (2.90GHz) processor and 16GB of memory.

[0192] Table 2 PID gain parameters of field model

[0193]

[0194] (2) Effect verification

[0195] In this embodiment, four typical representative scenarios are selected to evaluate the effectiveness and generalization ability of the proposed field model. In all experiments, the unmanned aerial vehicle performs a unified task: taking off from the initial position (0, 0, 0) m, flying to the target position (3, 3, 3) m, and achieving stable hovering at this point.

[0196] To achieve the comparability evaluation under different systems, the PX4 and the domain model respectively introduce disturbances in their own adaptive way. In the PX4, the actuator failure, wind disturbance and sensor attack are directly realized by modifying the simulation environment or the dynamic model. In the domain model, the disturbance is modeled in the form of symbolic deviation in the measurement information and injected into the input variable of the measurement node. The related disturbance modeling method and propagation mechanism are described in the embodiments of steps S2-S4. The symbolic modeling method has good computational feasibility and propagation traceability, and can support the approximate modeling of the influence of real disturbances.

[0197] The evaluation covers four scenarios: a nominal scenario without disturbance, a wind disturbance scenario, an actuator failure scenario, and a sensor attack scenario. The control performance of the PX4 and the domain model is compared and evaluated by the following three indicators: overshoot, which refers to the maximum deviation of the system response from the set value, used to measure the aggressiveness of the response; settling time, which refers to the time required for the system output to enter and remain within the set value ± 2% interval; root mean square error, which refers to the average deviation between the actual trajectory and the desired trajectory of the system, used to reflect the overall tracking accuracy.

[0198] (1) Nominal scenario without disturbance

[0199] In the nominal scenario without disturbance, this embodiment performs step response analysis on the systems based on PX4 and the domain model to verify the approximation ability of the domain model to the real flight control behavior. Figure 6 The response curves of both in the six control channels are shown, and the corresponding quantitative performance indicators are shown in Table 3.

[0200] From Figure 6 It can be seen that the response trend of the domain model in each axis is basically consistent with that of the PX4. In terms of position control, the domain model and the PX4 show similar rising trend and steady-state value in X, Y and Z axes. Although there is a certain amplitude of overshoot and response delay in X and Y directions, the root mean square error is almost the same as that of the PX4. In terms of attitude control, the domain model can accurately reproduce the heading stability and dynamic convergence characteristics of roll and pitch of the PX4, and the difference in root mean square error is negligible. From the above results, it can be seen that although the domain model has slight deviation in some dynamic indicators, the overall control effect is highly consistent with the PX4, verifying its good structural inheritance and dynamic response consistency in the nominal scenario.

[0201] Table 3 Comparison of control indicators in the nominal scenario

[0202]

[0203] (2) Wind disturbance scenario

[0204] This scenario simulates the influence of environmental wind disturbance on system control performance. Specifically, a constant horizontal gust along the positive Y-axis direction is applied at t = 10 s and persists during the subsequent flight to simulate the wind disturbance that may occur in reality. Figure 7 The PX4 and field model response curves for each axis under wind disturbance conditions are shown, and Table 4 summarizes the corresponding performance indicators.

[0205] As can be seen from Table 4, the response performance of the field model in all axes is similar to that of PX4, especially in Z-axis height control and yaw angle adjustment, where there is almost no difference between the two, with root mean square error and overshoot close to zero. On the main axis of Y-axis wind disturbance, although the field model has slightly higher overshoot and root mean square error than PX4, the overall trend and control stability are comparable to PX4. The control performance of the two is almost identical in the X-axis, which is the lateral non-main disturbance direction. The above results verify that the constructed field model can approximate the control response of PX4 under wind disturbance conditions.

[0206] Table 4 Comparison of control indicators under wind disturbance scenarios

[0207]

[0208] (3) Actuator fault scenario

[0209] To evaluate the field model's ability to approximate the control response of PX4 under actuator fault scenarios, a mild fault is injected into the thrust command for 30 seconds during hovering. Figure 8 The PX4 and field model response curves for each axis under this fault condition are shown, and Table 5 lists the corresponding control performance indicators.

[0210] From Figure 8 It can be seen that although the fault causes a certain deviation in the system trajectory, the response trend of the field model in all axes is similar to that of PX4, with synchronous changes in the dynamic process. In the X-axis and Y-axis directions, the overshoot and root mean square error of the field model and PX4 are close, with similar settling times, showing consistent closed-loop control characteristics. In the Z-axis direction, due to the direct impact of the fault on the thrust channel, the response of the field model shows more obvious dynamic deviation, but overall it can still approximate the system behavior of PX4 and maintain stable control. In the yaw angle channel, the control accuracy of the two is extremely small. Overall, the field model can effectively reproduce the influence of actuator faults on the dynamic performance of the control system and has good fault modeling capability.

[0211] Table 5 Comparison of control indicators under actuator fault scenarios

[0212]

[0213]

[0214] (4) Sensor attack scenario

[0215] To verify the approximation capability of the domain model to the real control behavior under sensor attack, an attack signal lasting 30 seconds was injected into the accelerometer observation during the steady hovering phase. Figure 9 The response changes of PX4 and the domain model in six axes are shown, and Table 6 summarizes the corresponding control performance indicators.

[0216] From the experimental results, it can be seen that under the influence of the attack, both systems maintain stable control in each axis, and the trajectory response remains consistent. Among them, in the X and Y translation axes, the differences between the domain model and PX4 in overshoot and root mean square error are small, and the response characteristics are highly close. In Z-axis and yaw angle control, the domain model shows slightly enlarged response fluctuations, but the overall still maintains good dynamic consistency with PX4. The above results show that the domain model can effectively simulate the disturbance effect of sensor attack on observation information and achieve dynamic approximation to the control behavior of PX4, verifying the approximation modeling capability of the symbolic disturbance modeling and propagation method proposed in this research.

[0217] Table 6 Comparison of control indicators under network attack scenario

[0218]

[0219] (5) Efficiency verification

[0220] To evaluate the performance overhead of the proposed model, two implementation schemes are compared in this embodiment, namely PX4 and the domain model of the present application. The evaluation is carried out in a unified simulation environment, and three efficiency evaluation indicators are used: code lines, average execution time and CPU usage. Table 7 summarizes the corresponding performance results.

[0221] Since PX4 includes multiple components such as main control logic, simulation environment, task scheduling framework and peripheral driver, in order to ensure fair comparison, we only count the code lines of the main control logic in the src directory corresponding to the domain model, which is about 404,000 lines. In contrast, the domain model only contains 2,797 lines of code, with a refined structure and simple implementation. In terms of runtime performance, the average execution time of the domain model per control cycle is 0.6 milliseconds, which is better than the 0.7 milliseconds of PX4. The average CPU usage of PX4 is 46.5%, while that of the domain model is only 36.38%. In terms of the three evaluation indicators, the domain model performs more lightweight in terms of resource consumption, and is suitable for deployment on resource-constrained platforms.

[0222] Table 7 Comparison of experimental overhead

[0223]

[0224] V. CONCLUSION

[0225] The disturbance-oriented flight control system information flow approximate modeling method in this embodiment has the following characteristics:

[0226] (1) The information flow element model paradigm used in step S1 for flight control system modeling abstracts flight control logic into an information flow network, defines a system structure composed of unit classes (such as modules, nodes), information classes (such as instructions, variables, disturbances), and connection classes (such as information flow, association), and supports composable modeling of control logic.

[0227] (2) Step S1 implements a flight control system information flow domain model construction method: based on the information flow element model, the domain model is constructed through steps such as functional module division, key node identification, and variable-transfer function mapping. The key node identification uses the reverse slicing algorithm to track data dependency along the control path, extracts control-related key variables and their corresponding transfer functions, and is used to construct a structured information flow representation.

[0228] (3) Step S2 implements a unified representation method for heterogeneous disturbances: by defining input variable deviation (δξ) and transfer function deviation (δF), component faults, environmental disturbances, and network attacks are uniformly abstracted as node-level information flow offset terms, realizing formal modeling and representation of heterogeneous disturbances.

[0229] (4) Step S4 implements a symbolic disturbance propagation analysis method: using the Jacobian chain rule, the disturbance propagation process is derived from the disturbance source node along the system information flow dependency path, the influence of the disturbance on the measurement results and the control output is formally represented, and the traceability modeling and quantitative analysis of disturbance propagation in flight control are realized.

[0230] (5) By executing steps S1-S4, a lightweight flight control system information flow domain model is realized, which has the characteristics of simple structure and controllable modeling granularity. Without relying on real flight control code, it can realize fast evaluation of system response under disturbance conditions, and is suitable for control system safety verification, adversarial evaluation, and pre-deployment simulation optimization tasks.

[0231] The disturbance-oriented flight control system information flow approximate modeling method in this embodiment has the following technical effects:

[0232] (1) Improve the generalization ability of disturbance representation: by uniformly abstracting environmental disturbances, component faults, and network attacks as symbolic deviations in information flow, the limitations of existing fault-tolerant control methods relying on type-specific modeling assumptions are effectively overcome, and the generality and universality of disturbance modeling are significantly improved.

[0233] (2) Platform decoupled structural abstraction: By constructing the information flow meta-model and taking it as the modeling paradigm, the system control structure can be expressed independently of the underlying implementation logic, thus breaking the dependence on specific flight control platforms (such as PX4, ArduPilot), and improving the portability and cross-platform deployment capability of the modeling method.

[0234] (3) Traceable modeling of disturbance propagation: By establishing a chain propagation analysis mechanism for disturbances from input nodes to control outputs, the transmission path of disturbances in the system can be explicitly tracked and quantitatively analyzed, enhancing the explainability modeling of abnormal behavior.

[0235] (4) Light-weight implementation with controllable granularity: By appropriately abstracting the control logic and information dependency at a suitable granularity, the state space dimension explosion problem caused by traditional fine-grained modeling is avoided, while the original control logic is preserved, effectively reducing the computational complexity, making the model lightweight, and suitable for resource-constrained embedded systems.

[0236] (5) Engineering practicability of approximate modeling: Under multiple typical disturbance scenarios (such as wind disturbance, fault, attack), the domain model shows high consistency with the real flight control system (e.g., PX4) in dynamic response, tracking performance, etc., verifying that the proposed method has good engineering application value while maintaining modeling accuracy.

[0237] The disturbance-oriented flight control system information flow approximate modeling method in the embodiment can be executed by a computer. The computer capable of executing the disturbance-oriented flight control system information flow approximate modeling method has all the technical effects of the disturbance-oriented flight control system information flow approximate modeling method.

[0238] It should be noted that, unless otherwise specified, when a feature is referred to as being "fixed" or "connected" to another feature, it can be directly fixed or connected to the other feature, or indirectly fixed or connected to the other feature. In addition, the up, down, left, right and other descriptions used in the disclosure are only relative to the relative position relationship of the components of the disclosure in the drawings. In the disclosure, the singular forms "a", "an" and "the" are also intended to include the plural forms, unless the context clearly indicates otherwise. In addition, unless otherwise defined, all technical and scientific terms used in the embodiments have the same meaning as understood by those skilled in the art. The terms used in the embodiments are only used to describe the specific embodiments, and are not intended to limit the present application. The term "and / or" used in the embodiments includes any combination of one or more related listed items.

[0239] It should be understood that, although the terms first, second, third, etc. can be used herein to describe various elements, components, regions, layers and / or sections, these elements should not be limited to these terms. These terms are only used to distinguish one element from another. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of the present disclosure. The use of any and all examples, or exemplary language (e.g., "such as", "for instance", etc.) provided herein, is intended merely to better illuminate the present embodiments and does not pose a limitation on the scope of the present disclosure unless otherwise claimed.

[0240] It will be appreciated that embodiments of the present application can be realized by computer hardware, a combination of hardware and software, or by computer instructions stored on a non-transitory computer-readable storage medium. The methods can be implemented in a computer program, using standard programming techniques, including the configuration of non-transitory computer-readable storage media with a computer program, wherein the storage medium so configured with the computer program instructs the computer to operate in a specific and predefined manner according to the method described in the specific embodiments and the accompanying drawings. Each program can be implemented in a high-level procedural or object-oriented programming language to communicate with a computer system. However, the program can be implemented in assembly or machine language, if desired. In any case, the language can be a compiled or interpreted language. Furthermore, the program can be able to run on a specially programmed integrated circuit for this purpose.

[0241] Furthermore, the operations of the processes described in the present embodiments can be performed in any suitable order unless otherwise indicated herein or otherwise clearly contradicted by context. The processes described in the present embodiments (or variations and / or combinations thereof) can be implemented under the control of one or more computer systems configured with executable instructions (e.g., computer programs, one or more computer programs, or one or more applications), and can be implemented as code (e.g., executable instructions, one or more computer programs, or one or more applications) executing collectively on one or more processors, by hardware, or combinations thereof. A computer program includes a plurality of instructions executable by one or more processors.

[0242] Further, the methods can be implemented in any type of computing platform operably connected to the appropriate, including but not limited to a personal computer, mini-computer, mainframe, workstation, network or distributed computing environment, separate or integrated computer platforms, or in communication with charged particle tools or other imaging devices, and the like. Aspects of the present application can be implemented in machine readable code stored on a non-transitory storage medium or device, whether removable or integrated into a computing platform, such as a hard disk, optical read and / or write storage media, RAM, ROM, and the like, such that it can be read by a programmable computer to configure and operate the computer to perform the processes described herein when the storage medium or device is read by the computer. In addition, the machine readable code, or portions thereof, can be transmitted over wired or wireless networks. The present embodiments of the application include these and other different types of non-transitory computer readable storage media when such media include instructions or programs that implement the above steps in conjunction with a microprocessor or other data processor. The present application also includes the computer itself when programmed in accordance with the methods and techniques of the present application.

[0243] The computer program can be applied to input data to perform the functions of the present embodiments, thereby transforming the input data to generate output data that is stored to non-volatile memory. The output information can also be applied to one or more output devices, such as a display. In preferred embodiments of the present application, the transformed data represents a physical and tangible object, including a particular visual depiction of the physical and tangible object produced on a display.

[0244] The above merely preferred embodiments of the present application and are not intended to limit the present application thereto. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the scope of the present application. The technical scheme and / or embodiments of the present application can have various modifications and changes within the scope of the present application.

Claims

1. A disturbance-oriented flight control system information flow approximate modeling method, characterized in that, The disturbance-oriented flight control system information flow approximate modeling method comprises: Taking an information flow meta-model as a modeling paradigm, a domain model of a flight control system is established; A disturbance information of a disturbance faced by the flight control system is symbolically and uniformly represented to obtain heterogeneous disturbance information; The heterogeneous disturbance information is embedded into the domain model; A disturbance propagation analysis is performed on the domain model.

2. The perturbation-oriented flight control system information flow approximate modeling method according to claim 1, characterized in that, The information flow meta-model comprises a unit class element, an information class element and a connection class element; the unit class element comprises a module, a node and a transfer function; the node is a minimum information processing unit for an information flow execution, and the node converts an input variable into an output variable through the transfer function.

3. The perturbation-oriented flight control system information flow approximate modeling method according to claim 1, characterized in that, The taking an information flow meta-model as a modeling paradigm, a domain model of a flight control system is established, comprising: A functional module division is performed on the flight control system to determine a plurality of functional modules; A key code block is identified from the functional modules, and the key code block is taken as a key node; The key node is classified into a measurement node, an execution node, a control node or an aggregation node; Related variables and a transfer function of the key node are obtained; According to the functional modules, the key node, the related variables and the transfer function, the domain model is composed.

4. The perturbation-oriented flight control system information flow approximate modeling method according to claim 3, characterized in that, The functional module division is performed on the flight control system to determine a plurality of functional modules, comprising: A configuration file of the flight control system is obtained; According to the configuration file, a compilation dependency relationship between each module in the flight control system is determined; According to the compilation dependency relationship, each module in the flight control system is combined and split to obtain each functional module.

5. The perturbation-oriented flight control system information flow approximate modeling method according to claim 3, characterized in that, The key code block is identified from the functional modules, comprising: A reverse slicing method is adopted to perform a tracking analysis on a data dependency relationship of a code in the functional module to determine the key code block.

6. The perturbation-oriented flight control system information flow approximate modeling method according to claim 5, characterized in that, The tracking analysis on the data dependency relationship of the functional module by adopting the reverse slicing method, comprising: An output variable in the functional module is taken as a slicing target; The reverse tracking is performed along a data dependency path from the output variable as a starting point until a data from an upstream or an external instruction is encountered to terminate; A data dependency path subjected to the reverse tracking is marked; A code block corresponding to the marked data dependency path is extracted as the key code block.

7. The perturbation-oriented flight control system information flow approximate modeling method according to claim 3, characterized in that, The related variables and the transfer function of the key node are obtained, comprising: Input variables and output variables of the key node are obtained; According to the input variables and the output variables, the transfer function is determined.

8. The perturbation-oriented flight control system information flow approximate modeling method according to claim 2, characterized in that, The disturbance faced by the flight control system is symbolically and uniformly represented to obtain the heterogeneous disturbance information, comprising: The disturbance faced by the flight control system is represented as an input variable or a transfer function deviation of a disturbed node, and an influence generated by the disturbance is represented as a symbolic deviation term in an output variable of a measurement node to obtain the heterogeneous disturbance information.

9. The perturbation-oriented flight control system information flow approximate modeling method according to claim 8, characterized in that, The heterogeneous disturbance information is embedded into the domain model, comprising: According to a symbolic deviation term in an output variable in the heterogeneous disturbance information, the measurement node in the domain model is configured.

10. The perturbation-oriented flight control system information flow approximate modeling method according to claim 8, characterized in that, The disturbance propagation analysis is performed on the domain model, comprising: An injection attack is obtained; modeling the injection attack as a disturbance to the flight control system, inputting the domain model; obtaining a chain Jacobian matrix between each of the key nodes in the domain model; determining a propagation process of the disturbance between each of the key nodes according to the chain Jacobian matrix.

Images