Beyond visual range air combat decision-making method based on dynamic fire field and artificial potential field fusion

Abstract

The application discloses a beyond-visual-range air combat decision-making method based on dynamic fire field and artificial potential field fusion, and relates to the field of aerospace. The method comprises the following steps: establishing a beyond-visual-range air combat simulation environment, modeling enemy and friendly camps in the air combat; grouping the enemy camp in real time; constructing a time-varying killing performance model of a missile to determine a three-view dynamic fire field of an aircraft-missile-target so as to establish an artificial potential field of combat groups, and determining the potential field gradient generated by each combat group in the grouping result; improving a particle updating formula in a particle swarm optimization algorithm by using the potential field gradient generated by the combat groups; solving a target function to determine a maneuvering decision of a friendly aircraft. The improved particle swarm optimization algorithm is obtained by improving the particle updating formula in the particle swarm optimization algorithm by using the potential field gradient generated by all the combat groups in the grouping result. The application can improve the coping ability, autonomous decision-making ability and coping ability in a complex battlefield environment of the beyond-visual-range air combat.

Description

Technical Field

[0001] This application relates to the aerospace field, and in particular to a beyond-visual-range air combat decision-making method based on the fusion of dynamic fire fields and artificial potential fields. Background Technology

[0002] Currently, beyond-visual-range (BVR) air combat scenarios are becoming increasingly complex, especially in the face of multi-target coordinated engagements, where existing decision support methods and systems have certain limitations. Traditional air combat decision-making methods cannot effectively describe the threat situation in multi-target coordinated operations and struggle to consider global threats and optimize aircraft maneuver decisions in real time. Specifically, existing technologies lack an effective model that can integrate threat information from different combat units, particularly in dynamic game scenarios, making it impossible to update and optimize aircraft response strategies in real time. Summary of the Invention

[0003] The purpose of this application is to provide a beyond-visual-range air combat decision-making method based on the fusion of dynamic fire field and artificial potential field. By constructing a comprehensive situation field, it accurately describes the threat situation in multi-target cooperative operations. Combined with an improved particle swarm optimization algorithm, it efficiently fuses multi-source threat information to provide optimized maneuver decision-making schemes for fighter jets, thereby improving autonomous decision-making capabilities and the ability to cope with complex battlefield environments.

[0004] To achieve the above objectives, this application provides the following solution:

[0005] This application provides a beyond-visual-range air combat decision-making method based on the fusion of dynamic fire field and artificial potential field, including:

[0006] Establish a beyond-visual-range air combat simulation environment, model the combat units of both sides in the air combat, and obtain three-degree-of-freedom kinematic models of different combat units.

[0007] Construct an enemy combat unit coordination judgment model based on firepower coordination. Based on the overlap of enemy combat unit attack zones, the enemy camp is grouped in real time, and the grouping results are output. In the grouping results, multiple combat groups are of the type of independent combat units or coordinated combat clusters.

[0008] Construct a time-varying lethality model of the missile and determine the dynamic fire field from three perspectives: carrier aircraft, missile, and target.

[0009] Based on the dynamic fire field from the three perspectives of carrier aircraft, missile, and target, we establish the artificial potential field of our fighter jets on all combat groups in the grouping results, and determine the potential field gradient generated by each combat group in the grouping results.

[0010] Construct the objective function;

[0011] In a beyond-visual-range air combat simulation environment, based on the three-degree-of-freedom kinematic models of different combat units of both sides in air combat, an improved particle swarm optimization algorithm is used to optimize and solve the objective function to determine the maneuver decisions of our fighter jets; the improved particle swarm optimization algorithm is obtained by improving the particle update formula in the particle swarm optimization algorithm by using the potential field gradient generated by all combat groups in the grouping results.

[0012] Optionally, the beyond-visual-range air combat simulation environment is used to describe air combat at the current moment. The state space;

[0013] Air combat at the present moment The state space is:

[0014] ;

[0015] ;

[0016] in, For air combat at the present moment The state space; for Blue Team's first combat unit In this state, the blue team represents our faction; for Shi Lanfang combat units The state; for The first combat unit of the Red Army In this state, the red team represents the enemy camp; for Shi Hongfang combat units The state; The first combat unit of the Blue Team Spatial coordinates; The first combat unit of the Blue Team track deflection; The first combat unit of the Blue Team The inclination of the flight path; The first combat unit of the Blue Team The velocity vector value.

[0017] Optionally, the physical type of the combat unit is a missile or an aircraft;

[0018] The three-degree-of-freedom kinematic model is as follows:

[0019] ;

[0020] in, This refers to the position of the combat unit in the inertial coordinate system. This indicates that the entity type of the combat unit is a missile. The time indicates that the entity type of the combat unit is an aircraft; For the flight speed of combat units; For the elevation angle of the combat unit; The yaw angle of the combat unit; For the roll angle of the combat unit; For combat unit normal overload; For tangential overload of combat units; For combat unit normal overload exist Components along the axial direction; For combat unit normal overload exist Components along the axial direction;

[0021] When constructing a three-degree-of-freedom kinematic model of a missile, , To control the quantity;

[0022] in, For missile normal overload exist Components along the axial direction; For missile normal overload exist Components along the axial direction;

[0023] When constructing a three-degree-of-freedom kinematic model of an aircraft, with To control the quantity;

[0024] in, This is for tangential overload of the aircraft; This is for aircraft normal overload; The roll angle of the aircraft.

[0025] Optionally, the grouping result is:

[0026] ;

[0027] Where G represents the grouping result of the enemy faction; This is the first combat group in the enemy camp; This is the second combat group within the enemy camp; The first in the enemy camp One operational group;

[0028] When the operational group is an independent operational unit, the number of operational units in the operational group is 1;

[0029] When the combat group is a collaborative combat cluster, the number of combat units in the combat group is greater than 1;

[0030] For any combat unit in the aforementioned cooperative combat cluster There is at least one combat unit in the collaborative combat cluster. satisfy: , ;

[0031] in, For combat units The attack zone; For combat units The attack zone.

[0032] Optionally, the missile's time-varying lethality model is as follows:

[0033] ;

[0034] ;

[0035] in, For the time-varying lethality of missiles; This refers to the overall probability level of ballistic interception. These are the initial ballistic conditions, used to describe the target's initial position coordinates in the aircraft's heading system and the ballistic flight time; (x, y, z) represent the missile's initial position coordinates in the aircraft's heading system and the ballistic flight time. Spatial coordinates at the moment of launch; Weighting of the impact of ballistic interception level. ; Quantify the coverage level; This is the maximum attack range; This refers to the distance between the projectile and the target.

[0036] Optionally, the method for determining the maximum attack range is as follows:

[0037] Acquire the missile-target parameters when the target is in its initial position; the missile-target parameters include the missile flight time and the relative velocity between the missile and the target.

[0038] Determine whether the missile-target parameters meet the cumulative interception probability triggering condition to obtain the first determination result; the cumulative interception probability triggering condition is that the missile flight time is less than the missile's rated working time and the relative speed between the missile and the target is greater than the minimum limit speed.

[0039] If the first judgment result is negative, then the distance between the missile and the initial position is shortened by a preset step size, and the process returns to the step "obtain missile-target parameters when the target is in the initial position";

[0040] If the first judgment result is yes, then calculate the probability of intercepting the target at the handover moment of the mid-mode guidance;

[0041] Determine whether the cutoff condition is met to obtain a second determination result; the cutoff condition is that the interception probability is less than the preset interception probability requirement and the target distance is less than the miss distance requirement.

[0042] If the second judgment result is negative, then the distance between the missile and the initial position is shortened by a preset step size, and the process returns to the step "obtain missile-target parameters when the target is in the initial position";

[0043] If the second judgment result is yes, then the distance between our combat unit and the initial position is determined to be the maximum attack range.

[0044] Optionally, the three-view dynamic fire field of carrier aircraft-missile-target includes: the fire field from the perspective of carrier aircraft, the time-varying lethality from the perspective of missile, and the hit probability information from the perspective of target.

[0045] The fire field from the aircraft's perspective was generated before missile launch using an "all-aspect attack centered on the aircraft" fire control mode.

[0046] The method for constructing a fire field for an independent combat unit is as follows: set the initial velocity of the target and make the target's direction of motion point towards the carrier aircraft; use a missile time-varying lethality calculation model to determine the missile time-varying lethality at each discrete point in the fire field; after the target is detected, calculate the missile time-varying lethality based on the carrier aircraft and target status, and update the fire field information of the sector where the target is located.

[0047] The method for constructing a fire field for a coordinated combat cluster is as follows: acquire multiple maneuvering modes; determine the interception probability of a target in the aggregation zone when it uses different maneuvering modes; use a missile time-varying lethality calculation model to determine the missile time-varying lethality at each discrete point in the fire field; correct the equal lethality envelope corresponding to the maneuvering mode; when there is a target, acquire the target's current maneuvering mode; use the target's subsequent maneuvering modes to correct the equal lethality envelope.

[0048] The time-varying lethality from the missile's perspective is generated in the mid-course guidance phase after missile launch, using a "missile-centric all-aspect attack" fire control mode.

[0049] The hit probability information from the target's perspective is generated in the air-to-air missile at the end of the handover using the "target-centered all-aspect attack" fire control mode.

[0050] The hit probability information includes: time-varying cumulative interception probability and formation aggregation area interception probability;

[0051] The time-varying cumulative interception probability is:

[0052] ;

[0053] in, The time-varying cumulative interception probability of a single platform for a target; N is the number of aircraft in the formation; For the first in the formation The probability of an aircraft intercepting a target;

[0054] The interception probability of the formation aggregation area is:

[0055] ;

[0056] in, The probability of interception in the formation aggregation area; The probability of multi-platform interception when the target is maneuvering; Towards the target The probability of the aircraft taking flight.

[0057] Optionally, based on the dynamic fire field from the three perspectives of the carrier aircraft, missile, and target, an artificial potential field is established for our fighter jets on all combat groups in the grouping results, and the potential field gradient generated by each combat group in the grouping results is determined, including:

[0058] When our fighter jets do not fall into the artificial potential field of the enemy's combat group, we use the first potential field formula to determine the artificial potential field of our fighter jets on the enemy's combat group, and use the first potential field gradient formula to determine the potential field gradient generated by the enemy's combat group.

[0059] When our fighter jets fall into the artificial potential field of the enemy's combat group, the second potential field formula is used to determine the artificial potential field of our fighter jets on the enemy's combat group, and the second potential field gradient formula is used to determine the potential field gradient generated by the enemy's combat group.

[0060] The formula for the first potential field is:

[0061] ;

[0062] ;

[0063] in, This represents the state vector of the combat groups of our fighter jets and the enemy's fighter jets; The artificial potential field for our fighter jets against the enemy's combat groups; This gives us the advantage; To repel potential fields; A positive gain coefficient; For the enemy's territory; This marks the boundary of the no-fly zone. The threshold for safe distance;

[0064] The gradient formula for the first potential field is:

[0065] ;

[0066] in, The formula for the gradient of the potential field; The potential field gradient generated by the enemy's combat group; This is the direction from which our firepower momentum increases the fastest. The negative gradient of the potential field is directed away from the threat.

[0067] The formula for the second potential field is:

[0068] ;

[0069] The formula for the gradient of the second potential field is:

[0070] ;

[0071] in, This is the direction from which the enemy's firepower decreases the fastest.

[0072] Optionally, the objective function is:

[0073] ;

[0074] ;

[0075] ;

[0076] ;

[0077] Where J is the objective function; This enhances the combined firepower of our aircraft against enemy combat units; The gravitational pull of our aircraft on enemy combat units within the potential field; This is the repulsive force of our aircraft against enemy combat units; Let be the potential field function generated by the red side against the blue side's combat units; The location of the Blue Force combat unit in space; Let be the potential field function generated by the blue team against the red team's combat units; The location of the Red Force combat unit in space; This is the first derivative of the fighter jet's state vector with respect to time. The differential equations that describe the dynamics of a fighter jet, i.e., the dynamic model; This represents the state vector of the fighter jet. For the control of fighter jets; For the fighter jet's overload; This is the maximum permissible overload for the fighter jet; The angular velocity of the fighter jet; This is the maximum permissible angular velocity of the fighter jet.

[0078] Alternatively, the improved particle update formula is:

[0079] ;

[0080] ;

[0081] in, For the (t+1)th iteration, the d-dimensional space is... The velocity of each particle; Inertial weight; For the t-th iteration, the d-dimensional space is... The velocity of each particle; , , and All are acceleration constants; , , and All Uniformly distributed random numbers; For the t-th iteration, the d-dimensional space is... The optimal position of each particle; For the t-th iteration, the d-dimensional space is... The position of each particle; Let be the group's optimal position in d-dimensional space after t iterations; The potential gradient generated by the first combat group The calculated position parameters; The potential gradient generated by the m-th combat group The calculated position parameters; For the (t+1)th iteration, the d-dimensional space is... The position of each particle.

[0082] According to the specific embodiments provided in this application, the following technical effects are disclosed:

[0083] This application provides a beyond-visual-range (BVR) air combat decision-making method based on the fusion of dynamic firefields and artificial potential fields. First, a BVR air combat simulation environment is established to model the core elements of air combat—air-to-air missiles and fighter jets—providing a reliable simulation platform for subsequent decision support. Second, a collaborative judgment model for enemy combat units based on fire coordination is constructed. Based on the overlap of enemy combat units' attack zones, they are grouped in real-time into independent combat units or collaborative combat clusters, providing support for subsequent tactical analysis and decision-making. A dynamic firefield is generated by establishing a missile time-varying lethality model, and artificial potential fields are established for the fighter jet targeting independent combat units / collaborative combat clusters based on the dynamic firefields from the perspectives of the aircraft, missiles, and targets. For different combat groups, each group generates an independent potential field gradient, representing the threat level perceived by the fighter jet from each combat unit or cluster. To enable the fighter jet to simultaneously respond to the gradient influence of all combat groups, an improved particle swarm optimization algorithm is adopted, effectively combining the gradients of each group of artificial potential fields for optimization, enhancing the optimization capability in each potential field gradient direction. In dynamic game scenarios of multi-target cooperative combat in beyond-visual-range air combat, existing methods lack effective models to describe the threat situation of multi-target cooperation and struggle to consider global threats and optimize maneuver decisions in real time. This invention constructs a comprehensive situational awareness field that reflects the dynamic threats of independent units / cooperative clusters and improves the particle swarm optimization algorithm to efficiently fuse multi-source threat gradients, guiding fighter jets to make maneuver decisions, thereby enhancing the autonomous decision-making capability and the ability to cope with complex battlefield environments.

[0084] This application constructs a comprehensive situational awareness field reflecting the dynamic threats of independent combat units and cooperative combat clusters, effectively describing the threat situation in multi-target cooperative operations. Compared with existing technologies, this invention solves the problem that traditional methods cannot update and optimize fighter jet response strategies in real time. It can identify the coordination of enemy combat units in real time, group them into independent combat units or cooperative combat clusters, and generate independent potential field gradients according to different combat groups, enabling fighter jets to flexibly respond to different combat situations and enhancing tactical flexibility and adaptability. By introducing the fusion of dynamic fire fields and artificial potential fields, and combining them with an improved particle swarm optimization algorithm, it can efficiently fuse multi-source threat gradients, guiding fighter jets to make real-time optimized maneuver decisions and improving the autonomous decision-making capabilities of fighter jets. In complex beyond-visual-range air combat environments, it can achieve more accurate and faster tactical responses. Attached Figure Description

[0085] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0086] Figure 1 This is a flowchart of a beyond-visual-range air combat decision-making method based on the fusion of dynamic fire field and artificial potential field in one embodiment of this application;

[0087] Figure 2 This is a flowchart illustrating the calculation of the maximum attackable range of a missile before / after firing in one embodiment of this application;

[0088] Figure 3 This is a flowchart illustrating the calculation of the lethality envelope in one embodiment of this application;

[0089] Figure 4 This is a composite diagram of the power of an artificial potential field in one embodiment of this application. Detailed Implementation

[0090] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0091] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0092] In one exemplary embodiment, such as Figure 1 As shown, a beyond-visual-range air combat decision-making method based on the fusion of dynamic fire field and artificial potential field is provided, including:

[0093] Step 101: Establish a beyond-visual-range air combat simulation environment, model the combat units of both sides in the air combat, and obtain the three-degree-of-freedom kinematic models of different combat units.

[0094] The game is a two-team competition, with red and blue teams representing the opposing sides: blue (our team) and red (the enemy team). The blue team's combat units are designated as follows: The Red Force's combat units are respectively denoted as .by For example, in time, It has a state in a spatial coordinate system:

[0095] .

[0096] in: for Spatial coordinates; The track deflection angle, i.e. The direction of the velocity vector is in Projection on a plane and The angle between the axes is positive in the clockwise direction. The inclination angle of the flight path, i.e. The direction of the velocity vector is The angle between two planes is positive when it faces upwards; for The magnitude of the velocity vector.

[0097] in, For the current moment of the entire air battle State space:

[0098] .

[0099] in, for Shi Lanfang The status of each combat unit for Shi Hongfang The status of each combat unit.

[0100] To establish a beyond-visual-range (BVR) air combat simulation environment, it is necessary to model the two main elements involved in the air combat process: air-to-air missiles and fighter jets. To focus the research on air combat decision-making and simplify the problem by ignoring the effects of angle of attack and sideslip angle, a three-degree-of-freedom kinematic model of the missile and aircraft is established, as follows:

[0101] .

[0102] In the formula: They represent missiles and aircraft, respectively. This indicates the position of the aircraft in the inertial coordinate system; Indicates flight speed; These represent pitch angle, yaw angle, and roll angle, respectively. Indicates tangential overload; Normal overload exist The components in two directions along the axis. When building a missile model, using... To control the quantity. When building the aircraft model, with... To control the quantity.

[0103] Step 102: Construct a collaborative judgment model for enemy combat units based on firepower coordination. Based on the overlap of enemy combat unit attack zones, group the enemy forces in real time and output the grouping results. Multiple combat groups in the grouping results can be categorized as independent combat units or collaborative combat clusters.

[0104] Firstly, in dynamic beyond-visual-range air combat scenarios, our airborne sensor system can accurately identify both friendly and enemy factions, as well as the type of enemy combat unit (aircraft / missile). Based on this, it can perform real-time coordination assessments of the detected enemy combat units, determining whether the enemy is an independent combat unit or a coordinated combat cluster. Assuming a set of combat units... For combat units Its attack zone can be represented as a spatial region. .

[0105] Taking a dual-aircraft unit as an example, if the attack zones of the two combat units overlap... If two machines have coordinated attack capabilities, they are considered a coordinated combat cluster. Based on the above, the following grouping results can be output. Each group It is either an independent combat unit or a coordinated combat cluster.

[0106] Step 103: Construct a time-varying lethality model of the missile and determine the dynamic fire field from three perspectives: carrier aircraft, missile, and target.

[0107] Step 104: Based on the dynamic fire field from the three perspectives of carrier aircraft, missile, and target, establish the artificial potential field of our fighter jets on all combat groups in the grouping results, and determine the potential field gradient generated by each combat group in the grouping results.

[0108] A dynamic fire field is generated based on a time-varying lethality model of a missile. By constructing a time-varying lethality calculation model for air-to-air missiles, the direction and magnitude of the fastest change in missile lethality at a certain location can be given. This can provide decision-making assistance information for pilots to position themselves for attack in the next moment, and its gradient can dynamically characterize the non-uniformity of the spatial distribution of lethality.

[0109] The time-varying lethality of air-to-air missiles is defined on two levels: First, before missile launch, the maximum attack range of the missile under a given probability of interception is calculated, and the coverage of the missile-target real-time distance is used to characterize the missile's time-varying lethality. Second, after missile launch and before the end of mid-course guidance, the cumulative probability of interception by trajectory extrapolation is calculated, and the lethality is characterized based on the coverage of the maximum attack range after launch and the real-time distance between the missile and the target at the extrapolation time.

[0110] This application provides a rapid calculation of the maximum attackable area before and after missile launch based on real-time intercept probability calculation. The flowchart for the rapid calculation of the maximum attackable area before and after launch is shown below. Figure 2 As shown. Based on the missile's maximum operating time. Minimum approach velocity of the target Maximum available overload of the missile and maximum line-of-sight pitch angle and azimuth Using constraints such as [variables], the calculation of the cumulative interception probability determines whether the handover time of mid-course guidance meets the preset interception probability requirements. If not, the calculation ends, and the distance step size is reduced by a certain amount based on the current initial missile-target distance. Conversely, the range is expanded; after the preset interception conditions are met, it is determined whether the missile's miss distance meets the requirements when the missile and target meet. By repeating this calculation, the maximum attack range before / after missile launch can be obtained. .

[0111] The maximum attack range is determined as follows: The missile-target parameters are obtained when the target is in its initial position. These parameters include the missile's flight time and the relative velocity between the missile and the target. The first judgment result is obtained by determining whether the missile-target parameters meet the cumulative interception probability trigger condition. The cumulative interception probability trigger condition is that the missile's flight time is less than its rated operating time, and the relative velocity between the missile and the target is greater than the minimum speed limit. If the first judgment result is negative, the distance between the missile and its initial position is shortened by a preset step size, and the process returns to the step "Obtain the missile-target parameters when the target is in its initial position". If the first judgment result is positive, the interception probability at the mid-mode guidance handover moment is calculated. The second judgment result is obtained by determining whether the cutoff condition is met. The cutoff condition is that the interception probability is less than the preset interception probability requirement, and the missile-target distance is less than the miss distance requirement. If the second judgment result is negative, the distance between the missile and its initial position is shortened by a preset step size, and the process returns to the step "Obtain the missile-target parameters when the target is in its initial position".

[0112] If the second judgment result is yes, then the distance between our combat unit and the initial position is determined to be the maximum attack range.

[0113] Utilizing the distance between the target and the target and maximum attack range Establish quantitative indicators for coverage:

[0114] .

[0115] So, in relation to the overall probability of interception of the ballistic trajectory Together, these constitute the two basic elements of designing a time-varying lethality performance model. Therefore, the calculation model for the missile's time-varying lethality performance is as follows:

[0116] .

[0117] In the formula: These are the initial ballistic conditions, used to describe the target's initial position coordinates in the aircraft's heading system and the ballistic flight time; (x, y, z) represent the missile's initial position coordinates in the aircraft's heading system and the ballistic flight time. Spatial coordinates at the moment of launch; Weighting of the impact of ballistic interception level. .

[0118] Based on different accuracy requirements, the time-varying lethality of the missile against a target at a certain location in space is sampled and calculated. According to the requirement of displaying equal lethality envelopes, target points with the same lethality are connected to form equal lethality envelopes. The calculation process is as follows: Figure 3 As shown.

[0119] The dynamic fire field based on the three perspectives of the carrier aircraft, missile, and target are as follows: The first refers to the fire field generated before missile launch using the "carrier aircraft-centered omnidirectional attack" fire control mode (carrier aircraft perspective), which provides information on the overall firepower capability under the dynamic game situation of enemy and friendly formations, enhances the pilot's fire control situation awareness, and improves the ability to detect and launch the enemy first; The second refers to the time-varying lethality generated during the mid-course guidance phase after missile launch using the "missile-centered omnidirectional attack" fire control mode (missile perspective), which provides lethality information under the maneuvering confrontation situation between the missile and the target, enhances the pilot's fire control decision-making awareness, and improves the ability to destroy the enemy first; The third refers to the hit probability information generated during the mid-course and terminal guidance phase of air-to-air missiles using the "target-centered omnidirectional attack" fire control mode (target perspective), which improves the ability to escape the enemy first.

[0120] If the fighter jet has no target, it will divide the entire area into multiple sectors centered on the aircraft to form a fire field, and set the initial position of the target at multiple distance intervals.

[0121] For the fire field of an independent combat unit, assuming the target's initial velocity magnitude and direction are pointing towards the carrier aircraft, the discrete points in the fire field are obtained according to the missile's time-varying lethality calculation model. Upon target detection, the missile's time-varying kill performance is calculated based on the aircraft and target status. Simultaneously, the fire field information for the target's azimuth sector is updated, primarily reflected in changes to the kill performance envelope. When multiple targets appear in the same or multiple azimuth sectors, the kill performance envelopes for different targets in different sectors are updated.

[0122] For the fire field of a coordinated combat cluster, the multi-aircraft fire field aggregation model first assumes different maneuvering patterns of the target, calculates the target's interception probability in the aggregation zone using the full probability formula, then calculates the kill performance of multiple aircraft against the target, and corrects the corresponding equal kill performance envelope. When a target is present, its current maneuvering pattern is already known. At this point, the formation fire field aggregation model mainly considers the target's possible subsequent maneuvers and their impact on the fire fields of each aircraft.

[0123] The time-varying cumulative interception probability of a single platform for a target is calculated based on the target maneuvering assumption. This allows us to obtain the interception probability of the target across multiple platforms:

[0124] .

[0125] In the formula: The number of aircraft in the formation; It is the first in the formation The probability of an aircraft intercepting a target.

[0126] The probability of interception in the formation aggregation zone can be obtained using the law of total probability:

[0127] .

[0128] In the formula: The probability of multi-platform interception when the target is maneuvering. ; Towards the target The probability of the aircraft taking flight. The size is set according to the missile's current motion state.

[0129] Therefore, the main function of the multi-aircraft fire field aggregation model is to address the problem that the different attack ranges of the same target calculated from different aircraft attack perspectives cannot be effectively combined in the target-centered omnidirectional attack mode. Based on the maneuver assumption of the target, it calculates the cumulative interception probability of the target in the aggregation zone, and then obtains the equal kill performance envelope of the two aircraft against the target and the aggregated fire field, so as to quantify the target kill capability aggregation effect of multiple fire nodes.

[0130] For missiles that have already been launched, it is necessary to calculate their time-varying kill performance against single or multiple targets based on their real-time position, and display their magnitude and direction in the form of single or multiple kill vectors.

[0131] When establishing an artificial potential field for a missile against a fighter jet, only the repulsive potential field is considered, and the gravitational potential field is not considered.

[0132] Based on the missile's time-varying lethality calculation model, the discrete points in the fire field were obtained. This allows for the establishment of an artificial potential field.

[0133] Fighter jets, along with their onboard equipment and weapons, exert varying degrees of influence on the surrounding airspace. This influence is quantitatively described using an artificial potential field, forming a virtual field that can be categorized into gravitational potential and repulsive potential. The gravitational potential represents the effect of the fighter jet and its onboard weapons on the airspace; the repulsive potential represents the threat exerted by the fighter jet on other combat units within the combat airspace. Therefore, the degree of threat posed by a fighter jet to a point in the air battlefield, i.e., its power, can be described by the gradient of the artificial potential field function. Power is divided into gravitational and repulsive forces; gravity represents the magnitude of the threat posed by the fighter jet to a point, while repulsion represents the magnitude of the threat received by the fighter jet at a point.

[0134] The gravitational pull of my aircraft on enemy combat units within the potential field is:

[0135] .

[0136] The repulsive force of our aircraft against enemy combat units is:

[0137] .

[0138] The combined firepower of our aircraft against enemy combat units is:

[0139] .

[0140] The power synthesis of artificial potential fields, such as Figure 4 As shown.

[0141] Establishing artificial potential fields for fighter jets targeting independent combat units / cooperative combat clusters mainly includes two scenarios: ① Our forces have not entered the enemy's artificial potential field. In this case, we move along the direction in which our own firepower potential increases the most, and treat the enemy's artificial potential field as a no-fly zone; ② Our forces have entered the enemy's artificial potential field. We move as far as possible along the direction in which the enemy's firepower potential decreases the most and the direction in which our own firepower potential increases the most. Assuming our potential field is... The enemy's position is The direction in which our firepower increases the fastest is... The direction in which the enemy's firepower decreases the fastest is ,in, This is the state vector of our fighter jets and the enemy's combat group.

[0142] Regarding the first scenario: Besides considering the direction from which our firepower potential increases the fastest... In addition, define no-fly zones To avoid entering the no-fly zone, we can introduce a repulsive potential field (which generates a repulsive force when approaching the boundary of the no-fly zone):

[0143] .

[0144] in, It is a threshold for a safe distance. It is a positive gain coefficient.

[0145] The potential field is then:

[0146] .

[0147] The potential gradient is:

[0148] .

[0149] Regarding the second scenario: considering the direction in which our firepower potential increases the fastest, The direction in which the enemy's firepower decreases the fastest is .

[0150] The potential field is then:

[0151] .

[0152] The potential gradient is:

[0153] .

[0154] For different combat groups Each group will generate an independent potential gradient. The fighter jet needs to respond to the gradient effects of all groups simultaneously in order to guide the fighter jet's maneuver decisions.

[0155] Step 105: Construct the objective function.

[0156] The decision-making process in S3 can be viewed as a multi-objective optimization problem, with our fighter jet in the following state: Our side targets the enemy's first combat unit or coordinated cluster The generated artificial field function is The total artificial potential field function is defined as:

[0157] .

[0158] The maneuver decision problem is transformed into the following constrained optimization problem:

[0159] .

[0160] in, For dynamic constraints, For overload constraints, For angular velocity constraints.

[0161] Step 106: In a beyond-visual-range air combat simulation environment, based on the three-degree-of-freedom kinematic models of different combat units of both sides in the air combat, an improved particle swarm optimization algorithm is used to optimize and solve the objective function to determine the maneuver decisions of our fighter jets. The improved particle swarm optimization algorithm is obtained by modifying the particle update formula in the particle swarm optimization algorithm using the potential field gradients generated by all combat groups in the grouping results.

[0162] An improved particle swarm optimization (PSO) algorithm is employed, effectively combining the artificial potential field gradients of each combat group to optimize and solve maneuver control variables, guiding the aircraft to make maneuver decisions. The basic idea of ​​PSO is: in In 3D space, the first In the nth iteration, the 1st The position of each particle is The speed is ,particle The best position in the past was The optimal position for the group is In the next iteration, the particle adjusts its velocity and position, and after N iterations, it searches for the optimal solution.

[0163] .

[0164] .

[0165] In the formula, Inertial weight; , Here is the acceleration constant; , for Uniformly distributed random numbers.

[0166] Due to the nonlinearity and piecewise nature of the superimposed artificial potential field function, the artificial potential field method may get stuck in local optima. In this case, a particle swarm optimization (PSO) algorithm is used in conjunction with the artificial potential field to perform optimization within the maneuverable range of the current flight direction. The fitness function is the sum of the superimposed artificial potential fields within the search region. An improved PSO algorithm is used to enhance optimization along the gradient directions of each potential field. The improved PSO algorithm is as follows:

[0167] .

[0168] .

[0169] In the formula, Here is the acceleration constant; for Uniformly distributed random numbers; Indicates by The calculated position parameters are then optimized using the algorithm described above to obtain the aircraft's control variables. After obtaining these control variables, the UAV's state for the next moment is updated by solving the UAV's kinematic and dynamic equations, thus enabling dynamic maneuvers. The updated state information is then used as input for the next moment's state information, and this process is repeated to complete the UAV's maneuver decision-making process.

[0170] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0171] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0172] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchain. The processors involved in the embodiments provided in this application may be, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc.

[0173] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0174] This application uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. In summary, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for beyond-visual-range air combat decision-making based on the fusion of dynamic fire field and artificial potential field, characterized in that, The application relates to a method for determining a fighter plane maneuvering decision in an over-the-horizon air combat simulation environment. The method comprises the following steps: An over-the-horizon air combat simulation environment is established, and combat units of enemy and friendly camps in air combat are modeled to obtain a three-degree-of-freedom kinematics model of different combat units; A cooperative judgment model of enemy combat units based on fire cooperation is constructed, the enemy camp is grouped in real time according to the overlapping condition of attack zones of the enemy combat units, and a grouping result is output; the types of multiple combat groups in the grouping result are independent combat units or cooperative combat clusters; A missile time-varying killing performance model is constructed, and a three-view dynamic fire field of an aircraft-missile-target is determined; Based on the three-view dynamic fire field of the aircraft-missile-target, an artificial potential field of the fighter plane to all combat groups in the grouping result is established, and the potential field gradient generated by each combat group in the grouping result is determined; A target function is constructed; In the over-the-horizon air combat simulation environment, the three-degree-of-freedom kinematics model of different combat units of enemy and friendly camps in air combat is used, and an improved particle swarm optimization algorithm is used to optimize and solve the target function to determine the fighter plane maneuvering decision; the improved particle swarm optimization algorithm is obtained by improving a particle updating formula in the particle swarm optimization algorithm by using the potential field gradient generated by all combat groups in the grouping result; Based on the three-view dynamic fire field of the aircraft-missile-target, an artificial potential field of the fighter plane to all combat groups in the grouping result is established, and the potential field gradient generated by each combat group in the grouping result is determined, including: When the fighter plane does not fall into the artificial potential field of the opponent combat group, a first potential field formula is used to determine the artificial potential field of the fighter plane to the opponent combat group, and a first potential field gradient formula is used to determine the potential field gradient generated by the opponent combat group; When the fighter plane falls into the artificial potential field of the opponent combat group, a second potential field formula is used to determine the artificial potential field of the fighter plane to the opponent combat group, and a second potential field gradient formula is used to determine the potential field gradient generated by the opponent combat group; ； ； wherein, is a state vector of the warplane of the side and the combat group of the opposite side; is an artificial potential field of the warplane of the side to the combat group of the opposite side; is a potential field of the side; is a repulsive potential field; is a positive gain coefficient; is an enemy potential field; is a no-fly zone boundary; is a threshold value of a safety distance; The first potential field formula is: ； wherein, is a potential field gradient formula; is a potential field gradient generated by the opposing combat group; is the direction in which the potential field of our firepower increases the fastest, is the negative gradient of the repulsive potential field, pointing away from the threat; The first potential field gradient formula is: ； The second potential field formula is: ； wherein is the direction in which the enemy fire potential decreases the fastest; The second potential field gradient formula is: ； ； in, For the (t+1)th iteration, the d-dimensional space is... The velocity of each particle; Inertial weights; For the t-th iteration, the d-dimensional space is... The velocity of each particle; , , and All are acceleration constants; , , and All Uniformly distributed random numbers; For the t-th iteration, the d-dimensional space is... The optimal position of each particle; For the t-th iteration, the d-dimensional space is... The position of each particle; Let be the group's optimal position in d-dimensional space after t iterations; The potential gradient generated by the first combat group The calculated position parameters; The potential gradient generated by the m-th combat group The calculated position parameters; For the (t+1)th iteration, the d-dimensional space is... The position of each particle.

2. The method of claim 1, wherein the method is based on a dynamic fire field and an artificial potential field fusion-based beyond-visual-range air combat decision method. The over-the-horizon air combat simulation environment is used to describe the state space of air combat at the current time ; The state space of the air combat at the current time instant is: { x, y, vx, vy, z, vz, a, b, ； ； in, For air combat at the present moment The state space; for Blue Team's first combat unit In this state, the blue team represents our faction; for Shi Lanfang combat units The state; for The first combat unit of the Red Army In this state, the red team represents the enemy camp; for Shi Hongfang combat units The state; The first combat unit of the Blue Team Spatial coordinates; The first combat unit of the Blue Team track deflection; The first combat unit of the Blue Team The inclination of the flight path; The first combat unit of the Blue Team The velocity vector value.

3. The method of claim 1, wherein the method is based on a dynamic fire field and an artificial potential field fusion-based beyond-visual-range air combat decision method. The improved particle updating formula is: The entity type of the combat unit is a missile or an aircraft; ； wherein, is the position of the combat unit in the inertial coordinate system, denotes that the entity type of the combat unit is a missile, denotes that the entity type of the combat unit is an aircraft; is the flight speed of the combat unit; is the pitch angle of the combat unit; is the yaw angle of the combat unit; is the roll angle of the combat unit; is the normal acceleration of the combat unit; is the tangential acceleration of the combat unit; is the normal acceleration of the combat unit is the component of the normal acceleration of the combat unit in the direction of the axis; is the normal acceleration of the combat unit is the component of the normal acceleration of the combat unit in the direction of the axis; When constructing the three-degree-of-freedom kinematic model of a missile, the control variable is taken as , . wherein, is the missile normal acceleration In the axial direction; is the missile normal acceleration In the axial direction; When constructing the three-degree-of-freedom kinematic model of an airplane, the control quantity is taken as ​ wherein, is the tangential acceleration of the aircraft; is the normal acceleration of the aircraft; is the roll angle of the aircraft.

4. The method of claim 1, wherein the method is based on a dynamic fire field and an artificial potential field fusion-based beyond-visual-range air combat decision method. The three-degree-of-freedom kinematics model is: ； G is the result of the enemy camp grouping; is the first combat group in the enemy camp; is the second combat group in the enemy camp; is the first combat group in the enemy camp; combat group in the enemy camp; The grouping result is: When the combat group is an independent combat unit, the number of combat units in the combat group is 1; for any combat unit in the cooperative combat cluster there is at least one combat unit in the cooperative combat cluster satisfies: , ; wherein is the attack zone of the combat unit . is the attack zone of the combat unit .

5. The method of claim 1, wherein the method is based on a dynamic fire field and an artificial potential field fusion-based beyond-visual-range air combat decision method. When the combat group is a cooperative combat cluster, the number of combat units in the combat group is greater than 1; ； ； wherein, is the time-varying kill performance of the missile; is the overall interception probability level of the trajectory; is the initial condition of the trajectory, which is used to describe the initial position coordinates of the target in the heading system of the carrier and the flight time of the trajectory; (x, y, z) is the space coordinates of the missile at the time of launching; is the time of launching; is the influence weight of the trajectory interception level, ; is the coverage degree quantitative index; is the maximum attackable distance; is the missile-target distance.

6. The method of claim 5, wherein the method is based on a dynamic fire field and an artificial potential field fusion-based beyond-visual-range air combat decision method. The missile time-varying killing performance model is: The determination method of the maximum attackable distance is: The missile-target parameters when the target is at an initial position are acquired; the missile-target parameters include a missile flight time and a missile-target relative speed; It is judged whether the missile-target parameters satisfy a cumulative interception probability trigger condition, and a first judgment result is obtained; the cumulative interception probability trigger condition is that the missile flight time is less than a missile rated working time, and the missile-target relative speed is greater than a minimum limited speed; If the first judgment result is no, the distance between the missile and the initial position is shortened by a preset step, and the step of "acquiring the missile-target parameters when the target is at the initial position" is returned; If the first judgment result is yes, the interception probability of the target at a midcourse guidance handover moment is calculated. judging whether a cut-off condition is met to obtain a second judging result; the cut-off condition is that a probability of interception is less than a preset probability of interception requirement and a missile-target distance is less than a miss distance requirement; if the second judging result is no, shortening a distance between the missile and the initial position by a preset step length, and returning to the step of obtaining the missile-target parameters when the target is at the initial position; if the second judging result is yes, determining that a distance between the combat unit and the initial position is a maximum attackable distance.

7. The method of claim 1, wherein the method is based on a dynamic fire field and an artificial potential field fusion-based beyond-visual-range air combat decision method. the three-view dynamic fire field of the aircraft-missile-target includes a fire field in a view of the aircraft, time-varying killing performance in a view of the missile, and hit probability information in a view of the target; the fire field in the view of the aircraft is generated by using an "aircraft-centered omnidirectional attack" fire control mode before the missile is launched; the fire field construction method for the independent combat unit is that an initial target speed is set to be zero, and a target motion direction is set to point to the aircraft; a time-varying killing performance of the missile at each discrete point in the fire field is determined by using a missile time-varying killing performance calculation model; after the target is found, the time-varying killing performance of the missile is calculated based on the states of the aircraft and the target, and the fire field information of the target in a bearing sector is updated; the fire field construction method for the cooperative combat cluster is that multiple maneuvering modes are obtained; a probability of interception of the target in an aggregation area in different maneuvering modes is determined, the time-varying killing performance of the missile at each discrete point in the fire field is determined by using the missile time-varying killing performance calculation model, and an equal killing performance contour corresponding to the maneuvering mode is corrected; a current maneuvering mode of the target is obtained when the target exists; the equal killing performance contour is corrected by using a subsequent maneuvering mode of the target; the time-varying killing performance in the view of the missile is generated by using an "missile-centered omnidirectional attack" fire control mode during a mid-guidance stage after the missile is launched; the hit probability information in the view of the target is generated by using a "target-centered omnidirectional attack" fire control mode when an air-to-air missile is handed over in the terminal stage; the hit probability information includes time-varying cumulative interception probability and formation aggregation area interception probability; the time-varying cumulative interception probability is: ； wherein, is the time-varying cumulative intercept probability of a single platform against a target; N is the number of formation aircraft; is the intercept probability of the i-th aircraft in the formation against a target; is the intercept probability of the i-th aircraft in the formation against a target; the formation aggregation area interception probability is: ； where, is the probability of interception by the formation aggregation zone; is the probability of multi-platform interception when the target maneuvers; is the probability of the target heading toward the is the probability of the launch aircraft flight.

8. The method of claim 1, wherein the method is based on a dynamic fire field and an artificial potential field fusion-based beyond-visual-range air combat decision method. the target function is: ； ； ； ； where J is the objective function; is the combined force of the red force combat units on the blue force combat units; is the gravitational force of the blue force combat units on the red force combat units in the potential field; is the repulsive force of the red force combat units on the blue force combat units; is the potential field function generated by the red force combat units on the blue force combat units; is the position of the blue force combat units in space; is the potential field function generated by the blue force combat units on the red force combat units; is the position of the red force combat units in space; is the first derivative of the state vector of the combat aircraft with respect to time; is the differential equation describing the dynamics of the combat aircraft, i.e., the dynamics model; is the state vector of the combat aircraft; is the control variable of the combat aircraft; is the overload of the combat aircraft; is the maximum allowable overload of the combat aircraft; is the angular velocity of the combat aircraft; is the maximum allowable angular velocity of the combat aircraft.

Images