A design method for aircraft autonomous maneuvering action library

Abstract

The present application belongs to the technical field of aircraft automatic flight control system design, and particularly relates to a method for designing an aircraft autonomous maneuvering action library. The method includes step S1, determining a variety of dangerous scenarios for different aircraft models; step S2, determining autonomous evasive maneuvers for the dangerous scenarios; step S3, integrating all autonomous evasive maneuvers to form an action library, wherein the autonomous evasive maneuvers in the action library are divided into basic control actions and complex tactical actions; step S4, quantitatively describing the control input of the autonomous evasive maneuvers in the action library to form a corresponding relationship table between autonomous evasive maneuvers and control instructions; step S5, deciding on autonomous evasive maneuvers based on the dangerous situation, and quantitatively describing and adding the decided autonomous evasive maneuvers that are not in the action library to the action library. The present application can continuously improve the action library and enrich the autonomous maneuvers in the action library.

Description

Technical Field

[0001] The present application belongs to the technical field of aircraft automatic flight control system design, and in particular relates to a method for designing an aircraft autonomous maneuvering action library. Background Art

[0002] As the battlefield environment becomes increasingly complex, aircraft face more and more threats. For some threat scenarios that require quick response and precise control, pilots may experience delayed response, insufficient control accuracy, or even misoperation due to emotions such as nervousness when performing maneuvers, which may lead to danger. Therefore, it is very necessary to study autonomous aircraft maneuvers and establish an autonomous maneuver action library.

[0003] Commonly used maneuver action libraries are mainly divided into two categories: one is the "typical tactical action library" which contains various classic air combat tactical flight actions; the other is the "basic control action library" which contains commonly used air combat control methods.

[0004] There are two commonly used typical tactical action libraries:

[0005] The first type includes 16 maneuvers: 1) straight flight; 2) steady circle; 3) turn the nose toward the target; 4) descent speed increase; 5) dive; 6) pull-up; 7) half somersault; 8) half roll; 9) combat turn; 10) tracking and aiming; 11) sharp evasive turn; 12) sharp circle; 13) serpentine maneuver; 14) barrel roll; 15) high-speed Yo-Yo; 16) low-speed Yo-Yo.

[0006] The second type includes 25 maneuvers: 1) straight flight; 2) steady circle; 3) front tracking; 4) dive acceleration; 5) oblique pull-up; 6) sudden pull-up; 7) pure tracking; 8) half somersault; 9) climb half roll; 10) high-G roll up; 11) high-G roll down; 12) evasive turn; 13) sudden evasive turn; 14) acceleration turn; 15) serpentine maneuver; 16) barrel roll; 17) sudden descent; 18) steep outside turn; 19) turn; 20) sharp climbing turn; 21) sudden descending turn; 22) gliding turn; 23) half roll turn; 24) high-speed Yo-Yo; 25) low-speed Yo-Yo.

[0007] The Basic Maneuver Library primarily refers to a library developed by NASA researchers based on the most common maneuvers in air combat. It primarily includes the following seven maneuvers: 1) Maximum Acceleration; 2) Maximum Deceleration; 3) Maximum G-Load Climb; 4) Maximum G-Load Dive; 5) Maximum G-Load Left Turn; 6) Maximum G-Load Right Turn; and 7) Steady Flight (all control variables remain unchanged). Any maneuver library designed in this manner is considered a Basic Maneuver Library.

[0008] Both types of maneuver libraries mentioned above have shortcomings. For the basic tactical maneuver library, in actual application, the process of selecting and executing a maneuver may cause the battlefield situation to change due to the enemy's reaction, making it impossible to complete a complete tactical maneuver in most cases. A common solution is to set several intermediate exits during the maneuver, but this approach still cannot completely solve this problem. For the basic control action library, although basic control actions can be judged step by step, this is not practical because these seven maneuvers are extreme control actions, and these seven basic actions cannot constitute all tactical actions. Summary of the Invention

[0009] To address at least one of the above technical issues, this application proposes a method for designing an autonomous aircraft maneuvering action library to address the issue of autonomous aircraft maneuvering decision-making and control. The method primarily includes:

[0010] Step S1: determining multiple dangerous scenarios for different aircraft models;

[0011] Step S2: determining an autonomous evasive maneuver for the dangerous scenario;

[0012] Step S3: Integrate all autonomous evasive maneuvers to form an action library. The autonomous evasive maneuvers in the action library are divided into basic control actions and complex tactical actions. The basic control actions are actions that can be completed with a single control input, and the complex tactical actions are formed by connecting multiple basic control actions.

[0013] Step S4, quantitatively describing the control input of the autonomous evasive maneuver in the action library to form a correspondence table between the autonomous evasive maneuver and the control command;

[0014] Step S5: Determine an autonomous evasive maneuver based on the dangerous situation, and convert it into a control amount of the aircraft's rudder based on the set specific control input parameters. For the autonomous evasive maneuver that is not in the action library, quantitatively describe it and add it to the action library.

[0015] Preferably, step S1 further comprises:

[0016] Step S11: extracting threat sources from the maneuvering characteristics and mission environments of different aircraft models. The threat sources are those that require precise control or rapid response, and the control burden exceeds a set value. Once the control error occurs, the consequences will exceed the set dangerous consequences.

[0017] Step S12: Determine a specific dangerous scenario based on the threat source and the aircraft's own maneuverability.

[0018] Preferably, in step S11, the aircraft types include at least fighters and transport aircraft, and the threat sources include at least threat sources generated by enemy attacks, threat sources generated by subjective movement of the aircraft, and threat sources generated by the natural environment.

[0019] Preferably, in step S2, determining the autonomous evasive maneuver for the dangerous scenario includes:

[0020] Based on the aircraft's maneuverability, the maneuver that allows the aircraft to escape from danger most quickly is selected as the autonomous evasive maneuver.

[0021] Preferably, if there are multiple maneuvers to choose from, a maneuver with low energy consumption, small overload change, and small maneuverability requirement is selected as the autonomous avoidance maneuver.

[0022] Preferably, in step S4, the quantitative description of the control input includes:

[0023] For basic control actions, longitudinal overload, normal overload and roll angle are used as inputs for quantitative description;

[0024] For complex tactical actions, multiple basic manipulation actions and the connection conditions between the basic manipulation actions are used as input for quantitative description.

[0025] Preferably, in step S5, determining an autonomous evasive maneuver based on the dangerous situation includes:

[0026] Step S51: Applying the RPD typical method of natural decision theory, making a real-time decision on an autonomous evasive maneuver to represent the amount of recovery from the dangerous environment, and decomposing the autonomous evasive maneuver into a longitudinal maneuver and a lateral maneuver;

[0027] Step S52: Generate a control instruction using the corresponding relationship table, including axial overload, normal overload, and roll angle;

[0028] Step S53: Taking the determined axial overload, normal overload and roll angle as targets, a negative feedback controller is designed to track the targets.

[0029] This application can continuously improve the action library and enrich the autonomous maneuvering actions in the action library. The action library formed by this method contains multiple simple maneuvering actions to complete the replicated maneuvering actions, which can respond to situation changes during the flight mission in a timely manner and improve the efficiency of action execution. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 This is a flow chart of a preferred embodiment of the method for designing an autonomous maneuvering action library for aircraft of the present application.

[0031] Figure 2This is a typical method flow chart of the RPD model of a preferred embodiment of the present application. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solutions and advantages of the implementation of this application clearer, the technical solutions in the implementation of this application will be described in more detail below in conjunction with the drawings in the implementation of this application. In the drawings, the same or similar numbers throughout represent the same or similar elements or elements with the same or similar functions. The described implementation is a part of the implementation of this application, not all of the implementations. The implementation described below with reference to the drawings is exemplary and is intended to be used to explain this application, and should not be understood as a limitation on this application. Based on the implementation in this application, all other implementations obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. The implementation of this application is described in detail below in conjunction with the drawings.

[0033] This application provides a method for designing an autonomous aircraft maneuvering action library, such as Figure 1 As shown, it mainly includes:

[0034] Step S1: Determine multiple dangerous scenarios for different aircraft models.

[0035] In some optional embodiments, step S1 further includes:

[0036] Step S11: Threat sources are extracted from the maneuvering characteristics and mission environments of different aircraft types. Threat sources are defined as manual operations that require precise control or rapid response, and whose operational burden exceeds a set value. If a misoperation occurs, the resulting consequences may exceed the set dangerous threshold. Aircraft types include at least fighter jets and transport aircraft. Threat sources include at least those generated by enemy attacks, those generated by the aircraft's subjective movements, and those generated by the natural environment.

[0037] In this embodiment, the operating burden exceeding the set value in the set value can be quantified according to the operating actions completed per unit time, and the set dangerous consequences can be quantified according to the casualties, economic losses, etc. caused by the dangerous consequences. In this embodiment, it is generally considered that the operating burden is very heavy when manual operations such as precise operation / rapid response are required, and once the operation is wrong, it will cause a threat source of serious dangerous consequences, as shown in Table 1, as the autonomous maneuvering requirements.

[0038] Table 1 Analysis of maneuverability characteristics and threats faced by different aircraft models

[0039]

[0040] Step S12: Determine a specific dangerous scenario based on the threat source and the aircraft's own maneuverability.

[0041] In this embodiment, specific dangerous scenarios are extracted based on the threat sources in the mission environment faced by each aircraft type and the aircraft's own maneuverability, as shown in Table 2 below.

[0042] Table 2 Typical dangerous scene extraction

[0043]

[0044] Step S2: Determine an autonomous evasive maneuver for the dangerous scenario.

[0045] In some optional embodiments, determining the autonomous evasive maneuver for the dangerous scenario in step S2 includes: selecting, based on the aircraft's maneuverability, the maneuver that will most quickly enable the aircraft to escape danger as the autonomous evasive maneuver; and, if multiple maneuvers are available, selecting the maneuver that minimizes energy consumption, overload variation, and maneuverability requirements as the autonomous evasive maneuver, as shown in Table 3 below.

[0046] Table 3 Selection of maneuvers in typical dangerous scenarios

[0047]

[0048] Step S3: All autonomous evasive maneuvers are integrated to form an action library. The autonomous evasive maneuvers in the action library are divided into basic control actions and complex tactical actions. The basic control actions refer to actions that can be completed given a single control input, and the complex tactical actions are formed by connecting multiple basic control actions.

[0049] In this embodiment, based on the maneuver selection results of each typical dangerous scenario and combined with existing research results, all maneuvers are summarized into a library to form an autonomous maneuver library including basic control actions and complex tactical actions.

[0050] Among them, the maneuvering action library contains 20 maneuvers, including maneuvers in the vertical plane, maneuvers in the horizontal plane and maneuvers in space.

[0051] Maneuvers in the vertical plane: 1) Acceleration and deceleration in level flight (constant forward flight, accelerated forward flight, decelerated forward flight); 2) Climb; 3) Dive; 4) Somersault;

[0052] Maneuvers in the horizontal plane: 1) Turning (left, right); 2) Circling;

[0053] Maneuvers in space: 1) Climb left / right; 2) Dive left / right; 3) Roll; 4) Oblique somersault; 5) Combat turn; 6) Combat half roll; 7) Half roll inverted; 8) High-speed Yo-Yo; 9) Low-speed Yo-Yo.

[0054] Step S4: quantitatively describe the control input of the autonomous evasive maneuver in the action library to form a correspondence table between the autonomous evasive maneuver and the control instruction.

[0055] In some optional embodiments, the quantitative description of the control input includes:

[0056] For basic control actions, longitudinal overload, normal overload and roll angle are used as inputs for quantitative description;

[0057] For complex tactical actions, multiple basic manipulation actions and the connection conditions between the basic manipulation actions are used as input for quantitative description.

[0058] In this embodiment, all maneuvers are divided into basic and complex maneuvers. Complex maneuvers can be composed of a combination of basic maneuvers. Therefore, the control inputs for basic maneuvers are primarily provided, while complex maneuvers simply provide the connection conditions between different basic maneuvers. Basic maneuvers are described in the track coordinate system, using the following parameters: axial overload nx, normal overload ny, and roll angle γ, as shown in Tables 4 and 5.

[0059] Table 4 Basic maneuver control input

[0060] Serial number Maneuvering method <![CDATA[n x ]]> <![CDATA[n y ]]> γ 1 Constant forward flight 0 1 0 2 Accelerate forward <![CDATA[C 1max ]]> 1 0 3 Slow down and fly forward <![CDATA[-C 1max ]]> 1 0 4 Turn right (circle right) 0 <![CDATA[C2]]> <![CDATA[C3]]> 5 Turn left (circle left) 0 <![CDATA[C2]]> <![CDATA[-C3]]> <![CDATA[6]]> <![CDATA[ 俯冲 ]]> <![CDATA[±C1]]> <![CDATA[C2(<1)]]> <![CDATA[0]]> <![CDATA[7]]> <![CDATA[ 跃升 ]]> <![CDATA[C1]]> <![CDATA[C2(>1)]]> <![CDATA[0]]> <![CDATA[8]]> <![CDATA[ 右俯冲 ]]> <![CDATA[±C1]]> <![CDATA[C2(<1)]]> <![CDATA[C3]]> <![CDATA[9]]> <![CDATA[ 右跃升 ]]> <![CDATA[C1]]> <![CDATA[C2(>1)]]> <![CDATA[C3]]> <![CDATA[ 10 ]]> <![CDATA[ 左俯冲 ]]> <![CDATA[±C1]]> <![CDATA[C2(<1)]]> <![CDATA[ - C3]]> <![CDATA[ 11 ]]> <![CDATA[ 左跃升 ]]> <![CDATA[C1]]> <![CDATA[C2(>1)]]> <![CDATA[ - C3]]> <![CDATA[ 12 ]]> <![CDATA[ 筋斗(斜筋斗)前半段 ]]> <![CDATA[0]]> <![CDATA[C2(=1)→C2(<1)]]> <![CDATA[C3]]> <![CDATA[ 13 ]]> <![CDATA[ 筋斗(斜筋斗)后半段 ]]> <![CDATA[0]]> <![CDATA[C2(<1)→C2(=1)]]> <![CDATA[C3]]> 14 roll 0 1 <![CDATA[[0,C3]]]>

[0061] Table 5 Complex maneuver control input

[0062]

[0063]

[0064] Step S5: Determine an autonomous evasive maneuver based on the dangerous situation, and convert it into a control amount of the aircraft's rudder based on the set specific control input parameters. For the autonomous evasive maneuver that is not in the action library, quantitatively describe it and add it to the action library.

[0065] In some optional implementations, in step S5, determining an autonomous evasive maneuver based on the dangerous situation includes:

[0066] Step S51: Applying the RPD typical method of natural decision theory, making a real-time decision on an autonomous evasive maneuver to represent the amount of recovery from the dangerous environment, and decomposing the autonomous evasive maneuver into a longitudinal maneuver and a lateral maneuver;

[0067] Step S52: Generate a control instruction using the corresponding relationship table, including axial overload, normal overload, and roll angle;

[0068] Step S53: Taking the determined axial overload, normal overload and roll angle as targets, a negative feedback controller is designed to track the targets.

[0069] Typical RPD methods of natural decision theory are as follows: Figure 2 As shown, it mainly includes:

[0070] S11: Experience with situations in changing contexts;

[0071] S12: Determine whether the situation is typical. If so, proceed to S13; otherwise, proceed to S17.

[0072] S13: The four factors of the RPD model: expectations, relevant cues, desirable goals, and actions. If the expectations match the pilot's expectations, proceed to S14; otherwise, proceed to S17.

[0073] S14: Assessing mental simulations;

[0074] S15: Is the mental simulation valid? If so, go to S16; otherwise, go to S12.

[0075] S16: Implement the action plan;

[0076] S17: Plot construction, feature matching, obtaining more effective information, and entering S11.

[0077] It should also be noted that in step S5, in a new dangerous scenario, if the existing maneuver does not meet the recovery requirements, a complex maneuver will be added to complete the recovery from the dangerous environment, and the maneuver that is not in the library will be self-learned and updated.

[0078] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A method for designing an aircraft autonomous maneuvering action library, characterized in that: include: Step S1: determining multiple dangerous scenarios for different aircraft models; Step S2: determining an autonomous evasive maneuver for the dangerous scenario; Step S3: Integrate all autonomous evasive maneuvers to form an action library. The autonomous evasive maneuvers in the action library are divided into basic control actions and complex tactical actions. The basic control actions are actions that can be completed with a single control input, and the complex tactical actions are formed by connecting multiple basic control actions. Step S4, quantitatively describing the control input of the autonomous evasive maneuver in the action library to form a correspondence table between the autonomous evasive maneuver and the control command; Step S5: Determine an autonomous evasive maneuver based on the dangerous situation, and convert it into a control amount of the aircraft's rudder based on the set specific control input parameters. For the autonomous evasive maneuver that is not in the action library, quantitatively describe it and add it to the action library.

2. The method for designing an autonomous aircraft maneuvering library according to claim 1, wherein: Step S1 further comprises: Step S11: extracting threat sources from the maneuvering characteristics and mission environments of different aircraft models. The threat sources are those that require precise control or rapid response, and the control burden exceeds a set value. Once the control error occurs, the consequences will exceed the set dangerous consequences. Step S12: Determine a specific dangerous scenario based on the threat source and the aircraft's own maneuverability.

3. The method for designing an aircraft autonomous maneuvering action library according to claim 2, wherein: In step S11, the aircraft types include at least fighters and transport aircraft, and the threat sources include at least threat sources generated by enemy attacks, threat sources generated by subjective movement of the aircraft, and threat sources generated by the natural environment.

4. The method for designing an autonomous aircraft maneuvering library according to claim 1, wherein: In step S2, determining an autonomous evasive maneuver for the dangerous scenario includes: Based on the aircraft's maneuverability, the maneuver that allows the aircraft to escape from danger most quickly is selected as the autonomous evasive maneuver.

5. The method for designing an aircraft autonomous maneuvering action library according to claim 4, wherein: If there are multiple maneuvers available, the maneuver with low energy consumption, small overload change, and small maneuverability requirement is selected as the autonomous avoidance maneuver.

6. The method for designing an aircraft autonomous maneuvering action library according to claim 1, wherein: In step S4, the quantitative description of the control input includes: For basic control actions, longitudinal overload, normal overload and roll angle are used as inputs for quantitative description; For complex tactical actions, multiple basic manipulation actions and the connection conditions between the basic manipulation actions are used as input for quantitative description.

7. The method for designing an aircraft autonomous maneuvering action library according to claim 1, wherein: In step S5, the decision to perform an evasive maneuver based on the dangerous situation includes: Step S51: Applying the RPD typical method of natural decision theory, making a real-time decision on an autonomous evasive maneuver to represent the amount of recovery from the dangerous environment, and decomposing the autonomous evasive maneuver into a longitudinal maneuver and a lateral maneuver; Step S52: Generate a control instruction using the corresponding relationship table, including axial overload, normal overload, and roll angle; Step S53: Taking the determined axial overload, normal overload and roll angle as targets, a negative feedback controller is designed to track the targets.

Images