A method for designing air combat maneuvering decision based on fuzzy reasoning

Abstract

The present invention discloses a method for designing air combat maneuvering decision based on fuzzy reasoning, proposes an air combat situation assessment model based on fuzzy reasoning, designs an objective function of maneuvering decision and a unique fuzzy reasoning rule, the output of the fuzzy reasoning engine is a corresponding weight coefficient of the air combat situation that can be adjusted dynamically, and the air combat situation advantage factor and the variable weight together constitute the objective function. In practical applications, it is necessary to combine the maneuvering action library, calculate the corresponding maneuvering action that maximizes the objective function, and output the single-machine flight control instruction corresponding to the action to meet the needs of autonomous maneuvering decision-making. According to this decision-making system, our fighter jets can adjust the speed and flexibility of maneuvering actions in different air combat situations, and achieve the best state relative to the enemy aircraft before launching weapons. At the same time, the optimized maneuvering action library ensures the accuracy of air combat situation updates, which is easy to implement in engineering.

Description

Technical Field

[0001] The invention relates to the technical field of fuzzy control, and in particular to an air combat maneuver decision design method based on fuzzy reasoning. Background Art

[0002] The process of air combat maneuver decision-making refers to simulating the pilot's choice of maneuvering actions in various air combat situations based on mathematical optimization, artificial intelligence and other methods to generate aircraft control instructions. Domestic and foreign scholars have conducted research on air combat decision support systems. At present, the commonly used research methods for air combat maneuver decision-making are divided into expert system method, differential game method, influence diagram method and optimization theory method. The expert system method is to manually summarize the maneuvering situations in different air combat situations into a rule base. The differential game method mainly discretizes the continuous game model, and both parties recursively search for the optimal decision within the decision step in their respective maneuver decision sets. The influence diagram method can reflect the pilot's decision-making process and take into account the uncertainty of the environment and the pilot's preferences. The optimization theory method uses optimization methods such as genetic algorithms, Bayesian reasoning, and statistical optimization to select an optimal value from a limited number of maneuvers as a decision output within a limited decision time. However, due to its complex model and large amount of solution calculation, it is difficult to meet the requirements of real-time decision-making.

[0003] In summary, it is not difficult to find that it is crucial to express situation information reasonably. The decision value of air combat maneuver decision is usually the optimal maneuver control quantity under the current or future air combat situation. Therefore, air combat situation assessment is the basis of fighter maneuver decision and provides necessary information support for fighter air combat maneuver decision. Fuzzy decision-making is a very effective multi-factor decision-making method for comprehensive evaluation of things affected by various factors. The evaluation result is not absolutely positive or negative, but expressed by fuzzy sets. At the same time, fuzzy decision-making can formulate corresponding decision rules based on expert control experience. In the decision-making process, the output parameters can be dynamically adjusted within a specific range to respond to input changes, thereby providing better diversity, dynamics and applicability. Since fuzzy logic has good robustness and computational efficiency, fuzzy logic is used in modeling to construct maneuver decision factor functions. Summary of the invention

[0004] The present invention provides an air combat maneuver decision design method based on fuzzy reasoning in dynamically changing air combat scenarios, which improves the accuracy and rapidity of fighter maneuver decisions in changeable air combat scenarios and increases the air combat advantage maintenance time of our fighter jets over enemy aircraft.

[0005] A method for designing air combat maneuvering decision based on fuzzy reasoning includes the following steps:

[0006] Step 1: By assuming that the performance of our and enemy aircraft is exactly the same and both are point masses, the dynamic model of each aircraft is derived;

[0007] Step 2: Based on the kinematics of both aircraft in step 1, design four maneuver decision factors: angle, distance, speed and altitude advantage factor, expressed as K A ,K R ,K V ,K H ;

[0008] Step 3: Establishment of fuzzy reasoning method;

[0009] 31) Fuzzy logic is used to design the corresponding fuzzy language subsets and membership functions of each fuzzy variable;

[0010] 32) Design a fuzzy algorithm with four maneuver decision factors as input and the output is defuzzified to obtain variable weight ω A ,ω R ,ω V ,ω H :

[0011] Step 4: Setting variable weight maneuver decision function; using four maneuver decision factors K A ,K R ,K V ,K H and variable weight ω A ,ω R ,ω V ,ω H Establish variable weight maneuver decision function;

[0012] Step 5: Optimize the basic maneuver action library of the fighter jet, traverse the action library, select the maneuver action that maximizes the value of the variable-weight maneuver decision function and execute the corresponding instructions to realize the fighter jet's autonomous maneuver decision in close-range air combat scenarios.

[0013] Furthermore, in step 1, in the inertial coordinate system, the three-degree-of-freedom kinematic equation of each aircraft is:

[0014]

[0015] In the formula, [x, y, z, V, θ, ψ] is the state vector of the model, where x, y, z represent the position of the center of mass of the aircraft in the inertial system, V represents the total speed of the aircraft along the body axis, θ represents the pitch angle of the aircraft in the inertial system, and ψ represents the yaw angle of the aircraft in the inertial system;

[0016] The kinetic model is specifically:

[0017]

[0018] In the formula, [n x ,n z ,γ] is the control vector of the aircraft; g represents the acceleration of gravity; tangential overload n x Along the direction of flight speed, forward is positive; normal overload n z Perpendicular to the flight velocity vector, vertically upward with the direction of the aircraft's head, it is positive; γ is the track roll angle, and it is positive when the aircraft tilts to the right.

[0019] Furthermore, the calculation method of each maneuver decision factor in step 2 is:

[0020] 21) Angle advantage factor K A ∈[0,1]; K A The calculation method is:

[0021]

[0022] in, Indicates the azimuth of the enemy aircraft; Indicates the azimuth of our fighter plane;

[0023] 22) Distance advantage factor K R ∈[0,1]; K R The calculation method is

[0024]

[0025] Among them, R is the distance between the enemy fighter and our fighter; Rm r The maximum range of missiles carried by our fighter planes; Rm b The maximum range of missiles carried by enemy aircraft; Rr r is the maximum detection distance of our fighter radar; Rr b is the maximum detection distance of the enemy fighter radar;

[0026] 23) Speed ​​advantage factor K V ∈[0,1];

[0027]

[0028] Among them, V r is the airspeed of our fighter, V b is the airspeed of the enemy aircraft;

[0029] 24) Height advantage factor K H ∈[0,1]:

[0030] K H The calculation method is:

[0031]

[0032] Where: ΔZ=Z r -Z b The height of our aircraft minus the height of the enemy aircraft.

[0033] Furthermore, in step 21: when the azimuth of our fighter plane Azimuth of enemy aircraft When K A =0, at this time the enemy fighter is right behind our fighter, our fighter is being followed by the enemy fighter, and the angle advantage value takes the minimum value of 0;

[0034] The azimuth of our fighter plane Azimuth of enemy aircraft When K A =0.5, at this time our fighter and the enemy fighter are flying parallel to each other, and the angle advantage factors of both sides are the same, which is 0.5;

[0035] The azimuth of our fighter plane Azimuth of enemy aircraft When K A =1, at this time our fighter is right behind the enemy fighter, our attack is very advantageous, and the angle advantage factor is the maximum value 1.

[0036] Furthermore, in step 22: the distance advantage factor is related to the missile range and radar detection range of the enemy and our fighters, and needs to be classified and discussed: there are three equal situations in formula (4):

[0037] Both the enemy and our fighters are within the missile attack range of the other side, both are beyond the radar detection range of the other side, and both are beyond the missile attack range of both sides, but within the radar detection range of the other side, that is, they can only sense their presence but cannot attack. All three situations are K. R =0.5;

[0038] If Party A's fighter is within the maximum range of Party B's airborne missiles, but Party B's fighter is beyond the maximum range of Party A's airborne missiles, Party A's radar can only detect Party B but cannot attack it. Party B is now in a great advantage in air combat, and the distance advantage factor is 0.5 < K R <0.7, Party A is at a disadvantage in air combat, and the distance advantage factor is 0.3 <K R <0.5.

[0039] Furthermore, the step 31 is specifically as follows:

[0040] Accurate amount K A The fuzzy language subset is:

[0041]

[0042] Blur Amount The membership function of is:

[0043]

[0044] Accurate amount K R The fuzzy language subset is:

[0045]

[0046] Blur Amount The membership function of is:

[0047]

[0048] Accurate amount K V The fuzzy language subset is:

[0049]

[0050] Blur Amount The membership function of is:

[0051]

[0052] Accurate amount K H The fuzzy language subset is:

[0053]

[0054] Blur Amount The membership function of is:

[0055]

[0056] Furthermore, the fuzzy algorithm in step 32 is specifically as follows: according to four input variables and four output variables, the fuzzy rules include 32 types of air combat situation classifications, which can reflect the air combat situation in more detail and convert the abstract air combat situation into a specific weight factor in the advantage function;

[0057] When our side is in an advantageous situation, we should shorten the distance between the two sides, put the enemy in an inescapable zone, maintain the angle advantage, and expand the height advantage and speed advantage. Therefore, in an advantageous situation, the distance factor has the largest weight, followed by the angle factor, and finally the height factor and speed factor.

[0058] When we are at a disadvantage, we should increase the distance between us, increase the lead angle of the enemy aircraft, and escape from the attack area of ​​the enemy aircraft; therefore, the distance factor and angle factor under disadvantage are the main factors;

[0059] When the two sides are in a balance of power, our aircraft should increase the angle to occupy the advantageous position as soon as possible, and then increase the height factor and speed to obtain a greater energy advantage;

[0060] When both sides are at a disadvantage, our aircraft should increase the angle factor, altitude factor and speed factor to gain greater energy advantage and battlefield initiative.

[0061] Furthermore, the centroid method is used to solve the fuzziness in step 32:

[0062]

[0063] Where μ(z) is the fuzzy output variable The corresponding output memberships, a and b are the four maneuver decision factors K A ,K R ,K V ,K H The corresponding lower and upper limits of the range.

[0064] Furthermore, the variable weight maneuver decision function established in step 4 using the four maneuver decision factors and variable weights is P=ω A ·K A +ω R ·K R +ω V ·K V +ω H ·K H ; This function can adaptively adjust the weights of maneuver decision factors according to the results of situation assessment to ensure the fighter's dominant position.

[0065] The beneficial effects of the present invention are:

[0066] Aiming at the autonomous maneuvering decision-making problem of fighter jets in 1V1 air combat, the present invention proposes a decision-making method based on fuzzy reasoning. By modeling the maneuvering decision-making process of fighter jets, evaluating the air combat situation from four aspects, namely angle, distance, speed and altitude, and dynamically adjusting the corresponding weights of various situation factors using the fuzzy reasoning method, the autonomous maneuvering decision-making of fighter jets in close-range air combat scenarios is realized. In addition, the maneuvering action library is optimized, the fighter jets' responsiveness in air combat is improved, and maneuvering decisions are made more effective. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0068] Figure 1 The air combat maneuver decision logic diagram of the present invention;

[0069] Figure 2 This is a schematic diagram of the air combat situation of the present invention;

[0070] Figure 3 is the fuzzy rule table of the present invention;

[0071] Figure 4 It is a three-dimensional trajectory diagram of air combat in the embodiment;

[0072] Figure 5 Maneuver decision result timing diagram;

[0073] Figure 6 Weight change curve. DETAILED DESCRIPTION

[0074] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention is further described in detail below in conjunction with specific implementation methods.

[0075] The embodiment of the present invention provides an air combat maneuver decision design method based on fuzzy reasoning, the structural diagram is as follows: Figure 1 As shown, it mainly involves the design of multiple modules in the decision-making system, including the calculation of air combat advantage factors, the establishment of fuzzy reasoning methods and the setting of variable weight maneuver decision functions; the other related signal quantity closed-loop systems mainly include fighter flight control systems, maneuver action library determination, enemy aircraft position prediction, etc.

[0076] In this paper, fighter jets are fixed-wing aircraft with autopilots. In a one-on-one dogfight, two fighter jets were simulated simultaneously, assuming that the performance of both aircraft was exactly the same.

[0077] The numerical values ​​of this embodiment are as follows: our fighter is red and the enemy fighter is blue. Assume that in the simulation experiment, the two sides have the same perception of battlefield information, weapon performance and aircraft maneuverability. The speed of the red fighter is 35m0s, and the minimum turning radius is 350m; the initial state vector is [-600, -300, 1000, 350, 0, -π / 3]. The speed of the blue fighter is 250m / s, and the minimum turning radius is 500m; the initial state vector is [-3800, 4000, 1000, 250, 0, -2π / 3].

[0078] Step 1: By assuming that the performance of our and enemy aircraft is exactly the same, both are point masses, the dynamic model of each aircraft is derived; after determining the aircraft control instructions, it is necessary to obtain the perception of battlefield situation information and target information. In this article, the air combat scenario is a 1V1 close combat, our side versus the enemy, and the air combat scenario is described as Figure 2 ;

[0079] In step 1, in the inertial coordinate system, the three-degree-of-freedom kinematic equation of each aircraft is:

[0080]

[0081] In the formula, [x, y, z, V, θ, ψ] is the state vector of the model, where x, y, z represent the position of the center of mass of the aircraft in the inertial system, V represents the total speed of the aircraft along the body axis, θ represents the pitch angle of the aircraft in the inertial system, and ψ represents the yaw angle of the aircraft in the inertial system;

[0082] The kinetic model is specifically:

[0083]

[0084] In the formula, [n x ,n z ,γ] is the control vector of the aircraft; g represents the acceleration of gravity; tangential overload n x Along the direction of flight speed, forward is positive; normal overload n z Perpendicular to the flight velocity vector, vertically upward with the direction of the aircraft's head, it is positive; γ is the track roll angle, and it is positive when the aircraft tilts to the right.

[0085] Step 2: Based on the above battlefield situation information, four maneuver decision factors are designed based on the kinematics of the two aircraft in step 1: angle, distance, speed and altitude advantage factor, expressed as K A ,K R ,K V ,K H ;

[0086] The calculation method of each maneuver decision factor is:

[0087] 21) Angle advantage factor K A ∈[0,1]; K A The calculation method is:

[0088]

[0089] in, Indicates the azimuth of the enemy aircraft; Indicates the azimuth of our fighter plane;

[0090] The azimuth of our fighter plane Azimuth of enemy aircraft When K A =0, at this time the enemy fighter is right behind our fighter, our fighter is being followed by the enemy fighter, and the angle advantage value takes the minimum value of 0;

[0091] The azimuth of our fighter plane Azimuth of enemy aircraft When K A =0.5, at this time our fighter and the enemy fighter are flying parallel to each other, and the angle advantage factors of both sides are the same, which is 0.5;

[0092] The azimuth of our fighter plane Azimuth of enemy aircraft When K A =1, at this time our fighter is right behind the enemy fighter, our attack is very advantageous, and the angle advantage factor is the maximum value 1.

[0093] 22) Distance advantage factor K R ∈[0,1]; K R The calculation method is

[0094]

[0095] Among them, R is the distance between the enemy fighter and our fighter; Rm r The maximum range of missiles carried by our fighter planes; Rm b The maximum range of missiles carried by enemy aircraft; Rr r is the maximum detection distance of our fighter radar; Rr b is the maximum detection distance of the enemy fighter radar;

[0096] The distance advantage factor is related to the missile range and radar detection distance of the enemy and our fighters, and needs to be discussed in a classified manner: There are three equal situations in formula (4):

[0097] Both the enemy and our fighters are within the missile attack range of the other side, both are beyond the radar detection range of the other side, and both are beyond the missile attack range of both sides, but within the radar detection range of the other side, that is, they can only sense their presence but cannot attack. All three situations are K. R =0.5;

[0098] If Party A's fighter is within the maximum range of Party B's airborne missiles, but Party B's fighter is beyond the maximum range of Party A's airborne missiles, Party A's radar can only detect Party B but cannot attack it. Party B is now in a great advantage in air combat, and the distance advantage factor is 0.5 < K R <0.7, Party A is at a disadvantage in air combat, and the distance advantage factor is 0.3 <K R <0.5;

[0099] It can be seen from this that the airborne radar detection distance and the airborne missile attack distance are the main factors affecting the distance advantage. The better the fighter's radar detection and missile strike performance, the greater the distance advantage factor of our fighter.

[0100] 23) Speed ​​advantage factor KV ∈[0,1];

[0101]

[0102] Among them, V r is the airspeed of our fighter, V b is the airspeed of the enemy aircraft;

[0103] If the enemy's airspeed is far less than ours (defined in this article as the enemy's airspeed is less than 0.6 times ours), it means that it is difficult for the enemy to overtake ours by non-maneuvering flight alone, so our speed advantage factor is extremely large, which is 1; if the enemy's airspeed is far greater than ours (defined in this article as the enemy's airspeed is greater than 1.5 times ours), it means that the enemy can easily overtake ours by non-maneuvering flight alone, so our speed advantage factor is extremely small, which is 0.1; if the enemy's airspeed is not much different from ours (defined in this article as the enemy's airspeed is between 0.6 and 1.5 times ours), then our speed advantage factor is 0.1<K V <1.0.

[0104] 24) Height advantage factor K H ∈[0,1]:

[0105] K H The calculation method is:

[0106]

[0107] Where: ΔZ=Z r -Z b The height of our aircraft minus the height of the enemy aircraft.

[0108] When our fighter plane is much higher than the enemy's (this article defines that our fighter plane is 5km or more higher than the enemy's), it means that we have formed a height advantage over the enemy's, so our fighter plane's height advantage factor is extremely large, which is 1; if the enemy's height is much higher than ours (this article defines that the enemy's height is 5km or more higher than ours), the enemy has a height advantage over ours, so our fighter plane's height advantage factor is extremely small, which is 0; if the height difference between the two fighter planes is not large (this article defines that our fighter plane's height minus the enemy's height is between [-5km, 5km)), then our fighter plane's height advantage factor is 0<K H <1.0.

[0109] Step 3: Establishment of fuzzy reasoning method;

[0110] 31) Fuzzy logic is used to design the corresponding fuzzy language subsets and membership functions of each fuzzy variable;

[0111] Accurate amount K A The fuzzy language subset is:

[0112]

[0113] Blur Amount The membership function of is:

[0114]

[0115] Accurate amount K R The fuzzy language subset is:

[0116]

[0117] Blur Amount The membership function of is:

[0118]

[0119] Accurate amount K V The fuzzy language subset is:

[0120]

[0121] Blur Amount The membership function of is:

[0122]

[0123] Accurate amount K H The fuzzy language subset is:

[0124]

[0125] Blur Amount The membership function of is:

[0126]

[0127] 32) Design a fuzzy algorithm ( Figure 3 ), the algorithm input is four maneuver decision factors; the output is defuzzified to obtain variable weight ω A ,ω R ,ω V ,ω H ; The weight change curve is as follows Figure 6 shown.

[0128] The fuzzy algorithm is specifically as follows: based on four input variables and four output variables, the fuzzy rules include 32 types of air combat situation classification, which can reflect the air combat situation in more detail and transform the abstract air combat situation into a specific weight factor in the advantage function;

[0129] When our side is in an advantageous situation, we should shorten the distance between the two sides, put the enemy in an inescapable zone, maintain the angle advantage, and expand the height advantage and speed advantage. Therefore, in an advantageous situation, the distance factor has the largest weight, followed by the angle factor, and finally the height factor and speed factor.

[0130] When we are at a disadvantage, we should increase the distance between us, increase the lead angle of the enemy aircraft, and escape from the attack area of ​​the enemy aircraft; therefore, the distance factor and angle factor under disadvantage are the main factors;

[0131] When the two sides are in a balance of power, our aircraft should increase the angle to occupy the advantageous position as soon as possible, and then increase the height factor and speed to obtain a greater energy advantage;

[0132] When both sides are at a disadvantage, our aircraft should increase the angle factor, altitude factor and speed factor to gain greater energy advantage and battlefield initiative. Figure 4 The first 16 fuzzy rules are shown.

[0133] Defuzzification using the centroid method:

[0134]

[0135] Where μ(z) is the fuzzy output variable The corresponding output memberships, a and b are the four maneuver decision factors K A ,K R ,K V ,K H The corresponding lower and upper limits of the range.

[0136] Step 4: Setting variable weight maneuver decision function; using four maneuver decision factors K A ,K R ,K V ,K H and variable weight ω A ,ω R ,ω V ,ω H Establish a variable weight maneuver decision function; the variable weight maneuver decision function is P = ω A ·K A +ω R ·K R +ω V ·K V +ω H ·K H; This function can adaptively adjust the weights of maneuver decision factors according to the results of situation assessment to ensure the fighter's dominant position.

[0137] Step 5: Optimize the basic maneuver action library of the fighter jet, traverse the action library, select the maneuver action that maximizes the value of the variable-weight maneuver decision function and execute the corresponding instructions to realize the fighter jet's autonomous maneuver decision in close-range air combat scenarios.

[0138] The optimized maneuver action library contains maneuver action names and corresponding control instruction vectors [n x ,n z ,γ] is:

[0139] Motor No. Maneuver name <![CDATA[n x ]]> <![CDATA[n z ]]> γ 1 Level flight 1 1 0 2 accelerate 1.4 1 0 3 slow down 0.7 1 0 4 Pull up 1.4 7 0 5 Dive 1.4 7 π 6 Turn left 1.4 7 -π / 4 7 Turn right 1.4 7 π / 4

[0140] 1-7 in the table correspond to seven basic maneuvers: uniform level flight, maximum overload acceleration, maximum overload deceleration, climb, maximum overload dive, left turn and right turn.

[0141] The experimental results are as follows Figure 4 As shown. The confrontation curve of the enemy and our fighters in the first 120 seconds in Case 2 is given, where ☆ is the starting position of our fighter, the red solid line is the flight trajectory of our fighter, * is the starting position of the enemy fighter, and the blue solid line is the flight trajectory of the enemy fighter. It can be seen from the confrontation curve that our fighter's initial altitude is lower than the enemy's, and the speed is also lower than the enemy's. The overall air combat is at a disadvantage, but the final altitude is better than the enemy's, making it difficult for the enemy to attack us, and the enemy is locked within the attack range of our airborne missiles.

[0142] Select the timing diagram from the maneuver ( Figure 5 ) It can be seen that the enemy fighter first performed a dive maneuver (maneuver 5) in an attempt to use the altitude advantage to attack immediately. After the attack failed, it alternately performed a left turn maneuver (maneuver 6) and a level flight maneuver (maneuver 1). The combined effect was a level flight left turn, and the intention should be to wait for the next attack opportunity. However, our fighter, through situation analysis, immediately climbed by pulling up maneuver (maneuver 4) at the beginning, trying to eliminate the altitude difference with the enemy aircraft to prevent the enemy aircraft from suppressing its altitude. After that, it took turns to perform a left turn maneuver (maneuver 6) and a right turn maneuver (maneuver 7), and successfully avoided the attack of the enemy fighter by constantly turning and changing direction. The pull-up maneuver was also effective at this time. Our aircraft successfully counterattacked and surpassed the enemy aircraft in altitude, turning the altitude disadvantage into an advantage.

[0143] Aiming at the autonomous maneuvering decision-making problem of fighter jets in 1V1 air combat, the present invention proposes a decision-making method based on fuzzy reasoning. By modeling the maneuvering decision-making process of fighter jets, evaluating the air combat situation from four aspects, namely angle, distance, speed and altitude, and dynamically adjusting the corresponding weights of various situation factors using fuzzy reasoning methods, the autonomous maneuvering decision-making of fighter jets in close-range air combat scenarios is realized. In addition, the maneuvering action library is optimized to improve the fighter jets' responsiveness in air combat and make maneuvering decisions more effective; and an enemy aircraft position prediction module is introduced to more comprehensively describe the air combat situation, guide our fighter jets to make more targeted decisions, and enhance the robustness of the maneuvering decision results.

[0144] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for designing air combat maneuvering decisions based on fuzzy reasoning ， Its characteristics include the following steps: Step 1: By assuming that the performance of our and enemy aircraft is exactly the same and both are point masses, the dynamic model of each aircraft is derived; Step 2: Based on the kinematics of both aircraft in step 1, design four maneuver decision factors: angle, distance, speed and altitude advantage factor, expressed as K A ,K R ,K V ,K H ; Step 3: Establishment of fuzzy reasoning method; 31) Fuzzy logic is used to design the corresponding fuzzy language subsets and membership functions of each fuzzy variable; 32) Design a fuzzy algorithm with four maneuver decision factors as input; The output is defuzzified to obtain variable weight ω A ,ω R ,ω V ,ω H ; Step 4: Setting variable weight maneuver decision function; using four maneuver decision factors K A ,K R ,K V ,K H and variable weight ω A ,ω R ,ω V ,ω H Establish variable weight maneuver decision function; Step 5: Optimize the basic maneuver action library of the fighter jet, traverse the action library, select the maneuver action that maximizes the value of the variable-weight maneuver decision function and execute the corresponding instructions to realize the autonomous maneuver decision of the fighter jet in the close-range air combat scenario; The step 31 is specifically as follows: Accurate amount K A The fuzzy language subset is: Blur Amount The membership function of is: Accurate amount K R The fuzzy language subset is: Blur Amount The membership function of is: Accurate amount K V The fuzzy language subset is: Blur Amount The membership function of is: Accurate amount K H The fuzzy language subset is: Blur Amount The membership function of is: The fuzzy algorithm described in step 32 is specifically as follows: according to four input variables and four output variables, the fuzzy rules include 32 types of air combat situation classifications, which can reflect the air combat situation in more detail and transform the abstract air combat situation into a specific weight factor in the advantage function; When our side is in an advantageous situation, we should shorten the distance between the two sides, put the enemy in an inescapable zone, maintain the angle advantage, and expand the height advantage and speed advantage. Therefore, in an advantageous situation, the distance factor has the largest weight, followed by the angle factor, and finally the height factor and speed factor. When we are at a disadvantage, we should increase the distance between us, increase the lead angle of the enemy aircraft, and escape from the attack area of ​​the enemy aircraft; therefore, the distance factor and angle factor under disadvantage are the main factors; When the two sides are in a balance of power, our aircraft should increase the angle to occupy the advantageous position as soon as possible, and then increase the height factor and speed to obtain a greater energy advantage; When both sides are at a disadvantage, our aircraft should increase the angle factor, altitude factor and speed factor to gain greater energy advantage and battlefield initiative.

2. According to the air combat maneuver decision design method based on fuzzy reasoning as claimed in claim 1, It is characterized in that In step 1, in the inertial coordinate system, the three-degree-of-freedom kinematic equation of each aircraft is: In the formula, [x, y, z, V, θ, ψ] is the state vector of the model, where x, y, z represent the position of the center of mass of the aircraft in the inertial system, V represents the total speed of the aircraft along the body axis, θ represents the pitch angle of the aircraft in the inertial system, and ψ represents the yaw angle of the aircraft in the inertial system; The kinetic model is specifically: In the formula, [n x ,n z ,γ] is the control vector of the aircraft; g represents the acceleration of gravity; tangential overload n x Along the direction of flight speed, forward is positive; normal overload n z Perpendicular to the flight velocity vector, vertically upward with the direction of the aircraft's head, it is positive; γ is the track roll angle, and it is positive when the aircraft tilts to the right.

3. According to the air combat maneuver decision design method based on fuzzy reasoning as claimed in claim 1, It is characterized in that The calculation method of each maneuver decision factor in step 2 is: 21) Angle advantage factor K A ∈[0,1]; K A The calculation method is: in, Indicates the azimuth of the enemy aircraft; Indicates the azimuth of our fighter plane; 22) Distance advantage factor K R ∈[0,1]; K R The calculation method is Among them, R is the distance between the enemy fighter and our fighter; Rm r The maximum range of missiles carried by our fighter planes; Rm b The maximum range of missiles carried by enemy aircraft; Rr r Rr is the maximum detection distance of our fighter radar; b is the maximum detection distance of the enemy fighter radar; 23) Speed ​​advantage factor K V ∈[0,1]; Among them, V r is the airspeed of our fighter, V b is the airspeed of the enemy aircraft; 24) Height advantage factor K H ∈[0,1]: K H The calculation method is: Where: ΔZ=Z r -Z b The height of our aircraft minus the height of the enemy aircraft.

4. According to the air combat maneuver decision design method based on fuzzy reasoning as claimed in claim 3, It is characterized in that In step 21: when the azimuth of our fighter plane is Azimuth of enemy aircraft When K A =0, at this time the enemy fighter is right behind our fighter, our fighter is being followed by the enemy fighter, and the angle advantage value takes the minimum value of 0; The azimuth of our fighter plane Azimuth of enemy aircraft When K A =0.5, at this time our fighter and the enemy fighter are flying parallel to each other, and the angle advantage factors of both sides are the same, which is 0.5; The azimuth of our fighter plane Azimuth of enemy aircraft When K A =1, at this time our fighter is right behind the enemy fighter, our attack is very advantageous, and the angle advantage factor is the maximum value 1.

5. According to the air combat maneuver decision design method based on fuzzy reasoning as claimed in claim 3, It is characterized in that In step 22, the distance advantage factor is related to the missile range and radar detection range of the enemy and our fighters, and needs to be classified and discussed: There are three equal situations in formula (4): Both the enemy and our fighters are within the missile attack range of the other side, both are beyond the radar detection range of the other side, and both are beyond the missile attack range of both sides, but within the radar detection range of the other side, that is, they can only sense their presence but cannot attack. All three situations are K. R =0.5; If Party A's fighter is within the maximum range of Party B's airborne missiles, but Party B's fighter is beyond the maximum range of Party A's airborne missiles, Party A's radar can only detect Party B but cannot attack it. Party B is now in a great advantage in air combat, and the distance advantage factor is 0.5 < K R <0.7, Party A is at a disadvantage in air combat, and the distance advantage factor is 0.3 <K R <0.

5.

6. According to the air combat maneuver decision design method based on fuzzy reasoning as claimed in claim 1, It is characterized in that In step 32, the centroid method is used to solve the fuzziness: Where μ(z) is the fuzzy output variable The corresponding output memberships, a and b are the four maneuver decision factors K A ,K R ,K V ,K H The corresponding lower and upper limits of the range respectively.

7. According to the air combat maneuver decision design method based on fuzzy reasoning as claimed in claim 1, It is characterized in that The variable weight maneuver decision function established by using four maneuver decision factors and variable weights in step 4 is P=ω A ·K A +ω R ·K R +ω V ·K V +ω H ·K H ; This function can adaptively adjust the weights of maneuver decision factors according to the results of situation assessment to ensure the fighter's dominant position.

Images