CN115993835A - Target maneuver intention prediction-based short-distance air combat maneuver decision method and system - Google Patents
Target maneuver intention prediction-based short-distance air combat maneuver decision method and system Download PDFInfo
- Publication number
- CN115993835A CN115993835A CN202211689864.XA CN202211689864A CN115993835A CN 115993835 A CN115993835 A CN 115993835A CN 202211689864 A CN202211689864 A CN 202211689864A CN 115993835 A CN115993835 A CN 115993835A
- Authority
- CN
- China
- Prior art keywords
- unmanned
- enemy
- maneuver
- fighter
- unmanned fighter
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The invention discloses a near-distance air combat maneuver decision-making method and a near-distance air combat maneuver decision-making system based on target maneuver intention prediction, which are used for establishing an air combat evaluation model of an unmanned combat aircraft in the near-distance air combat countering process aiming at the air combat conditions of the unmanned combat aircraft of both parties; on the other hand, a self-adaptive self-unmanned fighter maneuver decision-making framework for the change of the target maneuver intention in the short-distance air combat process is constructed, after the intention prediction is added, compared with the same type decision-making method, the method can keep the dominant situation in most maneuver decision-making time, and the self-unmanned fighter maneuver decision-making framework can make maneuver decisions on the target at multiple moments without delaying the fighter.
Description
Technical Field
The invention relates to maneuver intention prediction and countermeasure decision-making for UCAV, in particular to a maneuver decision-making method and a maneuver decision-making system for a short-distance air combat based on target maneuver intention prediction.
Background
With the development of the age, the information industry represented by artificial intelligence is continuously going deep into various industries of human beings. The military field, in particular the capturing of the empty rights, is an important factor for measuring the trend of a battle, and the appearance of unmanned fighter aircraft (UCAV) is a key for playing the battle efficiency. At present, research on unmanned aerial vehicles is tightened in various countries, and particularly in the field of air combat decision-making, how to realize intelligent decision-making of unmanned aerial vehicles or construct a set of intelligent auxiliary decision-making system becomes a subject of research of various countries.
However, due to the drastic change of the air combat situation, the situation that the own UCAV only makes maneuvering decisions on the current moment of the target is easy to miss, and how to make judgment on the next intentional intention of the target UCAV is a problem which cannot be ignored. Meanwhile, the current situation assessment method does not have objective judgment criteria. These factors make intent detection and maneuver decision-making of UCAV challenging.
The final decision quantity of the air combat maneuver decision is often the optimal maneuver control quantity under the air combat situation in the current or future period, so the air combat situation assessment is the basis of the UCAV maneuver decision, and important information support is provided for the UCAV maneuver decision in the air combat. Currently, the existing methods have the following problems: (1) in the 1v1 short-distance air combat countermeasure process, the air combat situation is changed drastically, and the situation that the own UCAV only makes maneuvering decisions on the current moment of the target is easy to miss, so how to make judgment on the next intentional intention of the target UCAV is a problem to be considered. (2) The existing situation assessment method does not have objective judgment criteria, and how to give more detailed judgment criteria on the basis of intent assessment is also a problem to be considered.
Disclosure of Invention
The invention aims to provide a near-distance air combat maneuver decision-making method and system based on target maneuver intention prediction, so as to solve the problem that in the prior art, in the 1v1 near-distance air combat countermeasure process, the air combat situation is changed drastically, and own unmanned fighter only makes maneuver decisions on the current moment of a target so as to easily miss opportunities.
In order to achieve the above purpose, the present invention provides the following technical solutions:
a close-range air combat maneuver decision method based on target maneuver intention prediction comprises the following steps:
s1: acquiring state information of the unmanned fighter and the unmanned fighter of the enemy;
s2: according to the state information of the unmanned fighter plane and the unmanned fighter plane, three-degree-of-freedom models of the unmanned fighter plane and the unmanned fighter plane are built;
s3: constructing an air combat evaluation model of the unmanned fighter and the enemy fighter according to the three-degree-of-freedom models of the unmanned fighter and the unmanned fighter;
s4: constructing attack and escape intention functions of the enemy unmanned fighter plane according to the state information of the enemy unmanned fighter plane;
s5: predicting the intention of the unmanned fighter against the enemy according to the attack and escape intention function of the unmanned fighter against the enemy;
s6: according to the air combat evaluation models of the unmanned fighter and the unmanned fighter of the enemy and the intention of the unmanned fighter of the enemy, the comprehensive situation function value of the unmanned fighter of the enemy is obtained through fuzzy reasoning;
s7: and performing traversal calculation on the comprehensive situation function values of the enemy unmanned fighter plane to obtain maneuvering decisions of the unmanned fighter plane.
Preferably, the my unmanned fighter and enemy unmanned fighter status information in S1 includes: three-dimensional coordinates of level and altitude, flight rate, pitch angle, yaw angle, tangential overload, normal overload, and roll angle of my unmanned fighter and of enemy unmanned fighter.
Preferably, the sideslip angle during flight of the unmanned fighter of my and the unmanned fighter of enemy is equal to 0 °.
Preferably, the construction of the S3 hollow war evaluation model comprises the construction of four situation functions of angle, distance, speed and height and a comprehensive situation function.
Preferably, bayesian theory is used in S5 to predict enemy unmanned fighter intent.
Preferably, four types of results of predicting the intent of the unmanned fighter plane of the enemy are near attack, far attack, near escape and far escape, respectively.
Preferably, the specific method for obtaining the comprehensive situation function value of the unmanned fighter plane of the enemy through the fuzzy reasoning method in the S6 is that firstly, a fuzzy decision membership function is constructed to convert the angle, the speed, the distance and the height of the unmanned fighter plane of the enemy into a fuzzy language, and then a fuzzy rule decision tree is constructed to calculate the decision weight of the comprehensive situation function.
Preferably, the fuzzy decision membership function employs a rectangular membership function or a single point membership function.
Preferably, in S7, a rolling time domain control method is adopted to perform traversal and calculation on the comprehensive situation function value of the unmanned fighter plane of the enemy.
A near air combat maneuver decision system based on target maneuver intention prediction comprises the following steps:
an information acquisition module: the method is used for acquiring state information of the unmanned fighter of the my and the unmanned fighter of the enemy;
and a model building module: the three-degree-of-freedom model is used for building three-degree-of-freedom models of the unmanned fighter plane and the unmanned fighter plane according to the state information of the unmanned fighter plane and the unmanned fighter plane;
the air combat evaluation model building module: the method is used for constructing an air combat evaluation model of the unmanned fighter and the enemy fighter according to the three-degree-of-freedom model of the unmanned fighter and the enemy fighter;
the intention function construction module: the method comprises the steps of constructing an attack and escape intention function of an enemy unmanned fighter plane according to state information of the enemy unmanned fighter plane;
an intention prediction module: the method is used for predicting the intention of the unmanned fighter plane of the enemy according to the attack and escape intention function of the unmanned fighter plane of the enemy;
and a fuzzy reasoning module: the comprehensive situation function value of the enemy unmanned fighter is obtained through fuzzy reasoning according to the air combat evaluation model of the enemy unmanned fighter and the intention of the target unmanned fighter;
decision module: the method is used for performing traversal calculation on the comprehensive situation function values of the enemy unmanned fighter plane to obtain maneuvering decisions of the unmanned fighter plane. Compared with the prior art, the invention has the following beneficial effects: the invention provides a near-distance air combat maneuver decision-making method based on target maneuver intention prediction, which aims at the air combat situations of unmanned fighters of two parties, on one hand, establishes an air combat evaluation model of the unmanned fighters of the my in the process of the near-distance air combat countering; on the other hand, a self-adaptive self-unmanned fighter maneuver decision-making framework for the change of the target maneuver intention in the short-distance air combat process is constructed, after the intention prediction is added, compared with the same type decision-making method, the method can keep the dominant situation in most maneuver decision-making time, and the self-unmanned fighter maneuver decision-making framework can make maneuver decisions on the target at multiple moments without delaying the fighter.
Further, a close-range air combat maneuver decision system based on target maneuver intention prediction is provided, so that maneuver decisions of the unmanned fighter plane on the my side can be made more effectively.
Drawings
FIG. 1 is a flow chart of a close-range air combat maneuver decision method air combat maneuver decision based on target maneuver intention prediction;
FIG. 2 is a UCAV maneuver decision process based on rolling time domain of the present invention;
FIG. 3 is relative situation information of the unmanned aerial vehicle according to the present invention;
fig. 4 is an angle situation of the inventive friend or foe unmanned aerial vehicle;
FIG. 5 is a reachable set calculation flow according to the invention;
FIG. 6 is a process of enemy UCAV intent prediction in accordance with the present invention;
FIG. 7 is a diagram of a short-range fuzzy inference rule tree in accordance with the present invention;
fig. 8 is a remote fuzzy inference rule tree of the present invention.
Detailed Description
The invention will now be described in further detail with reference to specific examples, which are intended to illustrate, but not to limit, the invention.
The invention provides a near-distance air combat maneuver decision method based on target maneuver intention prediction, which is shown in fig. 1-8, and comprises the steps of firstly establishing a three-dimensional space kinematics and dynamics model of two UCAV, establishing a UCAV air combat situation assessment model, acquiring the next moment state quantity from a target UCAV reachable set by adopting escape and attack membership functions, calculating the next moment state quantity acquired by the target UCAV based on a Bayesian theory, acquiring the pre-taking intention of the target UCAV at the next moment, inputting a corresponding fuzzy inference decision model, acquiring the optimal situation weight of the UCAV of a host under the condition of predicting the target maneuver, traversing the acquired situation function by adopting a UCAV maneuver decision method based on a rolling time domain, acquiring the host optimal maneuver, and completing one-time decision. The method specifically comprises the following steps:
step 1: and acquiring UCAV state information of both sides of the friend and foe. The state of my UCAV is known, and the state of the enemy UCAV can be acquired by the airborne radar of the my in real time, and mainly comprises the information of the motion state. Aiming at the parameter measurement of the target UCAV, and the like, a great deal of researches exist, the invention provides a short-distance air combat maneuver decision method based on the target maneuver intention prediction, so the parameter measurement of the target UCAV does not belong to the research content of the invention, the maneuver decision and the intention analysis can be carried out only by utilizing the obtained parameter information of the UCAV of the two parties, and the specific details of the measurement are not explained. The state vector of UCAV can be expressed as x= [ x, y, z, ψ, θ] T Wherein x, y and z represent the level and height coordinates of UCAV; v denotes the UCAV flight rate; psi is pitch angle; θ is the yaw angle; the control vector is
Wherein n is x And n z Tangential overload and normal overload of UCAV, respectively, < ->
Is the roll angle.
Step 2: and constructing a UCAV three-degree-of-freedom model. Based on a ground coordinate system, a dynamic model and a dynamic model of the unmanned aerial vehicle of both the enemy and the me are established through reasonable assumption. The UCAV has three degrees of freedom of motion in the flight process, can be simplified into one particle, and has physical descriptive quantities such as flight speed, three-dimensional coordinates, attitude Euler angles and the like in a three-dimensional space. In order to focus the research on the combat decision method, the method makes the following four assumptions on the UCAV flight model design:
1) The UCAV does not take into account air resistance nor air flow rate during flight, i.e. the speed of the unmanned aerial vehicle is determined solely by its own maneuver.
2) UCAV has no sideslip in the flight process, namely the sideslip angle is equal to 0 degrees.
3) The UCAV has constant quality, and the gravity acceleration, the air density and other factors do not change along with the change of the flying environment.
4) The ground reference system is always in a static state, and the influence of the rotation of the earth on modeling is ignored.
Based on the setting, defining a three-dimensional space dynamics and a kinematic equation of UCAV under a ground inertial coordinate system as shown in a formula (1):
step 3: and (5) constructing an air combat evaluation model. And constructing four attitude functions of angle, distance, speed and height and a comprehensive attitude function according to UCAV attitude relations of the two parties of the enemy. The air combat situation information comprises s= [ a, R, V, H ], wherein a comprises an azimuth angle and an entry angle of my UCAV, R represents a relative distance vector of the my UCAV and the enemy UCAV, V represents a speed vector of the UCAV, H represents a height of the UCAV, and fig. 3 is a schematic diagram of the air combat UCAV relative situation.
(1) Angular situation dominance function
The advantage of my attack is greater when the my UCAV azimuth is within the radar search angle range or within the missile attack range or the enemy UCAV entry angle is less than the my UCAV escape angle. According to the relation, the azimuth angle and enemy entrance angle dominance functions of the unmanned aerial vehicle can be respectively constructed. The azimuth angle dominance function established by the method is as follows:
the entry angle dominance function constructed by the method is as follows:
in air combat, to achieve the final attack condition, only one of the angle parameters cannot be considered independently, and the UCAV azimuth angle and the target entry angle of the user need to be considered simultaneously, so that the angle situation dominance function is constructed as follows:
Φ A ＝Φ p Φ q (4)
in phi, phi A ∈[0,1]As an angular situation dominance function, the bigger the situation value, the bigger the dominance.
(2) Distance situation dominance function
During the combat process, my UCAV finds and captures enemies mainly through radar search, wherein effective killing is caused to the enemies mainly through airborne weapons, but as the distance between two parties increases, the capturing capacity of the radar and the killing power of the weapons gradually decrease, and the threat to the enemies decreases. Therefore, the distance situation dominance function is constructed by analyzing the attack distance, the non-escapable range and the search range of the radar of the weapon as in formula (5):
wherein D is Rmax Searching a range for the maximum of the radar; d (D) Mmax Is the maximum attack distance of the air-to-air missile; (D) MKmin ,D MKmax ) Is a target UCAV non-escapable range.
(3) Velocity situation dominance function
According to different speed requirements at different distances, the method introduces time-varying optimal air combat speed V best 。V best The optimal speed is higher when the distance is longer, so that the speed is increased to quickly reach the preset position; when the target is in the attack area, the target is maneuvered at a similar speed under the condition of keeping a certain speed advantage on the target, so as to achieve the attack condition. Based on the above, the construction speed situation is excellentPotential function:
in the optimum air combat speed V best As formula (7):
wherein V is r ，V b The respective UCAV speed, V, is My UCAV and target UCAV speed best For the current moment, the optimal speed of the UCAV is My, V max Maximum speed achievable for the enemy's UCAV.
(4) Height situation dominance function
The higher the aircraft flight altitude, the better it will be with respect to the target UCAV. The higher the my UCAV height relative to the target UCAV, the greater the energy maneuver advantage to provide for weapon firing, the greater the air combat advantage. However, the relative height of both enemy and UCAV must be maintained within a relative area until the target UCAV is locked within the escape-free range of the UCAV of the my and the winning of the battle is finally obtained. Thus, the situation dominance function of the build distance is as in equation (8):
the comprehensive air combat situation dominance function is shown as (11):
wherein omega is i (i=1, 2,3, 4) is the weight occupied by each situation dominance function.
Step 4: the attack and escape intent functions are constructed. And calculating a next time reachable set of the target UCAV, and selecting the maneuver which is most fit with the attack and escape intents from the reachable set. First at the systemThe state reachable set which can be reached under the specific constraint is constructed. At t 0 At the moment, the system starts from the initial state, takes all control quantities in the control constraint set, and at t f A set of all states reached. The ordinary differential equation of the state of the construction target is shown in the formula (10):
wherein X represents a state quantity of the blue fighter plane (including information such as a position, a speed, and an azimuth of the target); u represents a control amount of the target; t is time. The process of computing the reachable set is shown in fig. 5.
UCAV needs to be at t k Time to target UCAV next time (t k+1 Moment) of the UCAV is predicted, and a target UCAV at t is obtained according to a target control variable set k+1 The reachable set of time is
Prediction of target UCAV intent, i.e. from +.>
The most similar state element to the current state is selected.
If the target UCAV is the attack, the situation of the most adverse to the UCAV is maneuvered towards the my, so that the state elements in the reachable set can be evaluated by using the entry angle dominance function (3) as the target UCAV attack membership function to obtain a dominance membership evaluation value set of the states in the reachable set
The lowest membership representation is the most threatening to UCAV. Therefore, the state of the offensive intention with the corresponding element in the reachable set as the target is described as +.>
Similarly, if the target UCAV feels non-hostileAfter UCAV, the target is intended to escape, quickly leaves the battlefield and is away from the UCAV, the distance is increased in the escaping process, and the target entering angle is smaller. Thus, for escape intent, an escape membership function can be used to evaluate:
using the pair of escape membership functions of (11)
All elements in the list are evaluated to obtain a membership evaluation value set of each element>
Taking the element in the reachable set corresponding to the minimum membership value as escape intention state, and recording as
Step 5: and predicting the target intention based on Bayesian theory. Constructing a Bayesian inference model of four intentions, namely a near attack, a far attack, a near escape and a far escape, and bringing the attack maneuver and the escape maneuver selected in the step 4 into the four Bayesian inference models, wherein the final intention of the target UCAV at the next moment is the highest probability;
the behavior of the extended target UCAV herein is intended to be 4: first, a short-range attack; secondly, remote attack; thirdly, short-distance escape; fourth, long-distance escape. And for judging four intentions, evaluating the four action intentions of the target UCAV reachable set by adopting a Bayesian theory mechanism, and taking the calculated maximum intention probability as a target UCAV intention prediction result. The target UCAV intent prediction process based on the bayesian theory is shown in fig. 6.
The target UCAV intent prediction type Γ is defined as a near-range tap 1, a far-range tap 2, a near-range escape tap 3, and a far-range escape tap 4. In the target UCAV intention prediction process, only the intention of the target UCAV at the next moment is predicted, and the UCAV of the two parties is predictedThe height is limited by overload in a short time, and the variation is small. In order to simplify the model and improve the calculation efficiency, three factors of the azimuth angle of the UCAV on the my side, the relative height and the relative speed of the UCAV on the two sides of the friend and foe are ignored in the target intention prediction process, so that the target intention prediction is determined by the state S= { q, D } of the target UCAV. The target UCAV at t can be calculated by the following formula k+1 Probability of intent prediction result at moment:
in the method, in the process of the invention,
at t for target UCAV k+1 Time status (I)>
Is the target UCAV intent type. In the air combat process, the probability of occurrence of the target UCAV state:
combining formula (12) and formula (13), the target UCAV is at t k+1 Probability of intent prediction result at moment:
since the air combat is a dynamically changing process, the process of target UCAV intent prediction may be approximated as a markov decision process. Therefore, the result of target UCAV intention prediction is only related to the predicted state at the next moment, while the prior probability
Are independent of each other and satisfy->
Thus, there is:
since the elements q, R in the state variable S are independent of the state of the situation classification, a joint conditional probability density function can be obtained:
in the method, in the process of the invention,
is a conditional probability function.
In order to make the conditional probability function more fit to the actual requirement of the air combat, the threshold coefficient epsilon=0.8 is considered, so that the aim of avoiding the influence of excessive 0 on the prediction result caused by the joint conditional probability density function is achieved, and the definition of the conditional probability function is shown in the table 1.
Table 1 intent prediction state condition probability table
Step 6: and solving the state potential weight factors by a fuzzy reasoning method. And (3) constructing four intention fuzzy inference models of near attack, far attack, near escape and far escape, and carrying the target UCAV intention obtained in the step (5) into the corresponding fuzzy inference model to calculate to obtain the weight value and the comprehensive situation function of each situation function in the UCAV 9.
Firstly, a fuzzy decision membership function is constructed, and UCAV air combat factors are converted into fuzzy languages. The situational dominance factors are respectively as follows:
(1) Angle blur situation dominance factor:
(2) Distance blur situation dominance factor:
(3) Speed blur situation dominance factor:
(4) High blur situation dominance factor:
(5) Angle, speed, distance, and height blur situation weighting factors:
wherein: omega i Representing the precise output quantity after deblurring;
represents fuzzy output, E A 、E D 、E V 、E H Representing the blurred input amounts, respectively. In order to reduce the complexity of fuzzy reasoning calculation, the mapping from input to output is conveniently and rapidly carried out according to a fuzzy rule, and a rectangular membership function and a single-point membership function are adopted in the text.
According to expert experience, two kinds of fuzzy inference rule trees, namely a short-distance one and a long-distance one, are designed as shown in figures 7 and 8, each fuzzy rule tree comprises 16 inference rules, and the serial numbers of the fuzzy rule trees are Y in sequence i (i=1,..16), subscript i denotes the ith rule,the fuzzy rule tree weight range settings are shown in table 2.
Based on expert knowledge, 16 fuzzy rules of 4 different intentions are respectively interpreted: (1) when the target UCAV is about to take a short-range attack, the my UCAV firstly judges the merits of the angle of the friend and foe, and if the my has the merits of the angle, the other short-board situations are preferentially considered to be promoted; if there is no angular advantage on my side, consider to prefer to raise the angular situation while lowering the distance situation. (2) When a target UCAV is about to take a long-distance attack, the my UCAV firstly judges the merits of the angle of the friend and foe, and if the my has the merits of the angle, the lifting distance and the height situation are considered; if there is no angular advantage on my side, the angular situation is considered to be elevated preferentially, while the distance and altitude situation are elevated. (3) When the target UCAV is about to take close escape, the my UCAV firstly judges the merits of the angle of the friend and foe, and if the my has the merits of the angle, the lifting height and the speed situation are considered; if there is no angular advantage on my side, the angular situation is considered to be elevated preferentially while the altitude situation is elevated. (4) When the target UCAV is about to take long-distance escape, the my UCAV firstly judges the merits of the angle of the friend and foe, and if the my has the merits of the angle, the situation of the lifting distance is considered; if there is no angular advantage on my side, the angular situation is considered to be elevated preferentially, while the distance and altitude situation are elevated.
TABLE 2 fuzzy rule tree weight range setting
Defuzzification using barycentric method to obtain precise quantities:
wherein: μ (z) is a membership function of the fuzzy set where the output z is located; z 0 The exact quantity obtained for fuzzy reasoning. z 0 May not be consistent with the range of values actually output, and requires domain transformation:
wherein:
is a proportional molecule [ u ] max ,u min ]Is the variation range of the actual control quantity; [ z ] max ,z min ]Is z 0 Is described.
Step 7: maneuvering decision based on a rolling time domain control method. And (3) performing traversal solution on the comprehensive situation function value determined in the step (6) through a rolling time domain method, wherein the obtained optimal solution is the optimal maneuver to be selected at the next moment of the UCAV. Based on the strong timeliness of the air combat, the whole maneuver reception process is subjected to time and space discretization, and the optimal maneuver strategy is obtained in a segmented mode, so that the whole maneuver reception process is optimal. The scrolling time domain procedure is shown in fig. 2. Wherein T is i (i∈[1,n]) Is the ith process of the n discrete processes. The variable range of the control amount is divided into m control sequences based on the continuity of the two state changes and considering the mobility of the UCAV. The optimal sequence must be selected from the control sequences to make the maneuver decision when making the maneuver decision. u (u) k (k∈[1,n]) The intelligent control variable optimization method is a control variable, and the optimization of the control variable by using an intelligent algorithm is excessively long in time consumption and does not meet the real-time requirement of air combat. Therefore, attempts are made herein to divide control amounts
When each control variable is divided into 10 grades, the current state measured value is taken as an initial condition, and is regarded as a track optimization problem to be solved, and the optimal control solution u is calculated online * Executing control u in time domain T * Until the system obtains a new status measurement and takes it as a new initial condition. Optimal control solution u * And the control variable gradient is set as shown in formulas (24) (25):
a near air combat maneuver decision system based on target maneuver attempt prediction, comprising:
an information acquisition module: the method is used for acquiring state information of the unmanned fighter of the my and the unmanned fighter of the enemy;
and a model building module: the three-degree-of-freedom model is used for building three-degree-of-freedom models of the unmanned fighter plane and the unmanned fighter plane according to the state information of the unmanned fighter plane and the unmanned fighter plane;
the air combat evaluation model building module: the method is used for constructing an air combat evaluation model of the unmanned fighter and the enemy fighter according to the three-degree-of-freedom model of the unmanned fighter and the enemy fighter;
the intention function construction module: the method comprises the steps of constructing an attack and escape intention function of an enemy unmanned fighter plane according to state information of the enemy unmanned fighter plane;
an intention prediction module: the method is used for predicting the intention of the unmanned fighter plane of the enemy according to the attack and escape intention function of the unmanned fighter plane of the enemy;
and a fuzzy reasoning module: the comprehensive situation function value of the enemy unmanned fighter is obtained through fuzzy reasoning according to the air combat evaluation model of the enemy unmanned fighter and the intention of the target unmanned fighter;
decision module: the method is used for performing traversal calculation on the comprehensive situation function values of the enemy unmanned fighter plane to obtain maneuvering decisions of the unmanned fighter plane.
Although the embodiments of the present invention have been described above with reference to the accompanying drawings, the present invention is not limited to the above-described specific embodiments and application fields, which are merely illustrative, instructive, and not restrictive. Those skilled in the art, having the benefit of this disclosure, may make many forms without departing from the scope of the invention as claimed.
Claims (10)
1. The short-distance air combat maneuver decision-making method based on target maneuver intention prediction is characterized by comprising the following steps of:
s1: acquiring state information of the unmanned fighter and the unmanned fighter of the enemy;
s2: according to the state information of the unmanned fighter plane and the unmanned fighter plane, three-degree-of-freedom models of the unmanned fighter plane and the unmanned fighter plane are built;
s3: constructing an air combat evaluation model of the unmanned fighter and the enemy fighter according to the three-degree-of-freedom models of the unmanned fighter and the unmanned fighter;
s4: constructing attack and escape intention functions of the enemy unmanned fighter plane according to the state information of the enemy unmanned fighter plane;
s5: predicting the intention of the unmanned fighter against the enemy according to the attack and escape intention function of the unmanned fighter against the enemy;
s6: according to the air combat evaluation models of the unmanned fighter and the unmanned fighter of the enemy and the intention of the unmanned fighter of the enemy, the comprehensive situation function value of the unmanned fighter of the enemy is obtained through fuzzy reasoning;
s7: and performing traversal calculation on the comprehensive situation function values of the enemy unmanned fighter plane to obtain maneuvering decisions of the unmanned fighter plane.
2. The near air combat maneuver decision method based on target maneuver attempt prediction as claimed in claim 1 wherein the my unmanned fighter and enemy unmanned fighter status information in S1 includes: three-dimensional coordinates of level and altitude, flight rate, pitch angle, yaw angle, tangential overload, normal overload, and roll angle of my unmanned fighter and of enemy unmanned fighter.
3. The near air combat maneuver decision method based on target maneuver attempt prediction as claimed in claim 1 wherein the sideslip angle during flight of my unmanned fighter and enemy unmanned fighter is equal to 0 °.
4. The near air combat maneuver decision method based on target maneuver attempt prediction as recited in claim 1, wherein the construction of the S3 air combat assessment model includes the construction of four attitude functions of angle, distance, speed and altitude and a comprehensive attitude function.
5. The near air combat maneuver decision method based on target maneuver attempt prediction as defined in claim 1, wherein the enemy unmanned combat aircraft intent is predicted in S5 using bayesian theory.
6. The method for predicting maneuver attempt at close range air combat maneuver decision as defined in claim 5 wherein the four outcomes of the enemy unmanned combat maneuver attempt are predicted as close range attack, far range attack, close range escape and far range escape respectively.
7. The near air combat maneuver decision method based on the target maneuver intention prediction according to claim 1, wherein the specific method for obtaining the comprehensive situation function value of the enemy unmanned fighter plane in S6 through the fuzzy reasoning method is that firstly, a fuzzy decision membership function is constructed to convert the angle, speed, distance and height of the enemy unmanned fighter plane into a fuzzy language, and then a fuzzy rule decision tree is constructed to calculate the decision weight of the comprehensive situation function.
8. The target maneuver attempt prediction based short distance air combat maneuver decision method as recited in claim 7, wherein the fuzzy decision membership function is a rectangular membership function or a single point membership function.
9. The near air combat maneuver decision method based on the target maneuver attempt prediction according to claim 1, wherein in S7, the rolling time domain control method is used to traverse the integrated situation function values of the enemy unmanned fighter plane.
10. A target maneuver intention prediction-based short-distance air combat maneuver decision system, a target maneuver intention prediction-based short-distance air combat maneuver decision method as claimed in any one of claims 1-9, comprising:
an information acquisition module: the method is used for acquiring state information of the unmanned fighter of the my and the unmanned fighter of the enemy;
and a model building module: the three-degree-of-freedom model is used for building three-degree-of-freedom models of the unmanned fighter plane and the unmanned fighter plane according to the state information of the unmanned fighter plane and the unmanned fighter plane;
the air combat evaluation model building module: the method is used for constructing an air combat evaluation model of the unmanned fighter and the enemy fighter according to the three-degree-of-freedom model of the unmanned fighter and the enemy fighter;
the intention function construction module: the method comprises the steps of constructing an attack and escape intention function of an enemy unmanned fighter plane according to state information of the enemy unmanned fighter plane;
an intention prediction module: the method is used for predicting the intention of the unmanned fighter plane of the enemy according to the attack and escape intention function of the unmanned fighter plane of the enemy;
and a fuzzy reasoning module: the comprehensive situation function value of the enemy unmanned fighter is obtained through fuzzy reasoning according to the air combat evaluation model of the enemy unmanned fighter and the intention of the target unmanned fighter;
decision module: the method is used for performing traversal calculation on the comprehensive situation function values of the enemy unmanned fighter plane to obtain maneuvering decisions of the unmanned fighter plane.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211689864.XA CN115993835A (en) | 2022-12-27 | 2022-12-27 | Target maneuver intention prediction-based short-distance air combat maneuver decision method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211689864.XA CN115993835A (en) | 2022-12-27 | 2022-12-27 | Target maneuver intention prediction-based short-distance air combat maneuver decision method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115993835A true CN115993835A (en) | 2023-04-21 |
Family
ID=85989901
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211689864.XA Pending CN115993835A (en) | 2022-12-27 | 2022-12-27 | Target maneuver intention prediction-based short-distance air combat maneuver decision method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115993835A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116993010A (en) * | 2023-07-28 | 2023-11-03 | 南通大学 | Fixed wing unmanned aerial vehicle situation prediction method based on Bayesian neural network |
| CN117540626A (en) * | 2023-10-30 | 2024-02-09 | 南通大学 | Fixed wing unmanned aerial vehicle situation prediction method based on Bayesian neural network |
| CN117687322A (en) * | 2024-02-04 | 2024-03-12 | 青岛哈尔滨工程大学创新发展中心 | AUV cluster countermeasure simulation system and method considering individual faults |
| CN117540626B (en) * | 2023-10-30 | 2024-05-14 | 南通大学 | Fixed wing unmanned aerial vehicle situation prediction method based on Bayesian neural network |
-
2022
- 2022-12-27 CN CN202211689864.XA patent/CN115993835A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116993010A (en) * | 2023-07-28 | 2023-11-03 | 南通大学 | Fixed wing unmanned aerial vehicle situation prediction method based on Bayesian neural network |
| CN116993010B (en) * | 2023-07-28 | 2024-02-06 | 南通大学 | Fixed wing unmanned aerial vehicle situation prediction method based on Bayesian neural network |
| CN117540626A (en) * | 2023-10-30 | 2024-02-09 | 南通大学 | Fixed wing unmanned aerial vehicle situation prediction method based on Bayesian neural network |
| CN117540626B (en) * | 2023-10-30 | 2024-05-14 | 南通大学 | Fixed wing unmanned aerial vehicle situation prediction method based on Bayesian neural network |
| CN117687322A (en) * | 2024-02-04 | 2024-03-12 | 青岛哈尔滨工程大学创新发展中心 | AUV cluster countermeasure simulation system and method considering individual faults |
| CN117687322B (en) * | 2024-02-04 | 2024-05-03 | 青岛哈尔滨工程大学创新发展中心 | AUV cluster countermeasure simulation system and method considering individual faults |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115993835A (en) | Target maneuver intention prediction-based short-distance air combat maneuver decision method and system | |
| Zhang et al. | An improved constrained differential evolution algorithm for unmanned aerial vehicle global route planning | |
| CN114413906B (en) | Three-dimensional trajectory planning method based on improved particle swarm optimization algorithm | |
| CN107063255B (en) | Three-dimensional route planning method based on improved drosophila optimization algorithm | |
| CN112215283A (en) | Close-range air combat intelligent decision method based on manned/unmanned aerial vehicle system | |
| CN114330115B (en) | Neural network air combat maneuver decision-making method based on particle swarm search | |
| CN114492805A (en) | Air combat maneuver decision design method based on fuzzy reasoning | |
| Fu et al. | The overview for UAV air-combat decision method | |
| CN111443728A (en) | Chaos wolf optimization-based unmanned aerial vehicle formation control method | |
| CN115951709A (en) | Multi-unmanned aerial vehicle air combat strategy generation method based on TD3 | |
| Wu et al. | Visual range maneuver decision of unmanned combat aerial vehicle based on fuzzy reasoning | |
| CN113625569A (en) | Small unmanned aerial vehicle prevention and control hybrid decision method and system based on deep reinforcement learning and rule driving | |
| CN113139331A (en) | Air-to-air missile situation perception and decision method based on Bayesian network | |
| Yuan et al. | Research on UCAV maneuvering decision method based on heuristic reinforcement learning | |
| CN114638339A (en) | Intelligent agent task allocation method based on deep reinforcement learning | |
| CN112305913A (en) | Multi-UUV collaborative dynamic maneuver decision method based on intuitive fuzzy game | |
| CN115688268A (en) | Aircraft near-distance air combat situation assessment adaptive weight design method | |
| CN113741186B (en) | Double-aircraft air combat decision-making method based on near-end strategy optimization | |
| Xianyong et al. | Research on maneuvering decision algorithm based on improved deep deterministic policy gradient | |
| CN115268496B (en) | Unmanned aerial vehicle air countermeasure maneuvering controller and design method thereof | |
| Meng et al. | One-to-one close air combat maneuver decision method based on target maneuver intention prediction | |
| Karli et al. | Air combat learning from F-16 flight information | |
| Scukins et al. | Enhancing situation awareness in beyond visual range air combat with reinforcement learning-based decision support | |
| CN115145295A (en) | Online autonomous flight path optimization control method for unmanned aerial vehicle in dynamic environment | |
| CN113625739A (en) | Expert system optimization method based on heuristic maneuver selection algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |