CN111888767B - Missile ash box simulator trajectory registration method in simulation environment - Google Patents

Missile ash box simulator trajectory registration method in simulation environment Download PDF

Info

Publication number
CN111888767B
CN111888767B CN202010639747.7A CN202010639747A CN111888767B CN 111888767 B CN111888767 B CN 111888767B CN 202010639747 A CN202010639747 A CN 202010639747A CN 111888767 B CN111888767 B CN 111888767B
Authority
CN
China
Prior art keywords
missile
box
chromosome
simulator
typical design
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.)
Active
Application number
CN202010639747.7A
Other languages
Chinese (zh)
Other versions
CN111888767A (en
Inventor
赵艺阳
朴海音
詹光
杨振
周德云
孔维仁
张凯
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Northwestern Polytechnical University
Original Assignee
Northwestern Polytechnical University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Northwestern Polytechnical University filed Critical Northwestern Polytechnical University
Priority to CN202010639747.7A priority Critical patent/CN111888767B/en
Publication of CN111888767A publication Critical patent/CN111888767A/en
Application granted granted Critical
Publication of CN111888767B publication Critical patent/CN111888767B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
    • A63F13/00—Video games, i.e. games using an electronically generated display having two or more dimensions
    • A63F13/55—Controlling game characters or game objects based on the game progress
    • A63F13/57—Simulating properties, behaviour or motion of objects in the game world, e.g. computing tyre load in a car race game
    • A63F13/573—Simulating properties, behaviour or motion of objects in the game world, e.g. computing tyre load in a car race game using trajectories of game objects, e.g. of a golf ball according to the point of impact
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
    • A63F2300/00—Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game
    • A63F2300/80—Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game specially adapted for executing a specific type of game
    • A63F2300/8076—Shooting
    • 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
    • Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention provides a missile ash box simulator trajectory registration method in a simulated environment, which is used for acquiring the initial positions of a missile and an enemy plane at the launching moment to obtain N groups of typical design points; the method comprises the steps of taking a missile simulation system to be identified as a gray box simulator, taking a known missile model as a white box simulator, collecting missile path data output by the white box simulator, initializing each parameter to be identified in a random mode, evaluating the output of the white box simulator under each chromosome by using trajectory data generated by the gray box, and continuously optimizing population. The method is simple and easy to implement, not only avoids complicated sensitivity solution, but also avoids possible numerical problems, and can obviously enhance the identification accuracy, thereby effectively improving the reliability and the accuracy of parameter identification.

Description

Missile ash box simulator trajectory registration method in simulation environment
Technical Field
The invention belongs to the technical field of parameter identification and computer simulation, and particularly relates to a missile trajectory registration method in a simulation environment.
Background
The trajectory registration is a process of utilizing the existing missile simulation system, starting from an initial value of a parameter to be identified according to a certain algorithm under the same input condition, calculating a trajectory curve according to a missile model, carrying out error comparison with trajectory curves output by other missile simulation systems of the same type but with unknown parameters, and gradually adjusting parameter values according to the algorithm to minimize the trajectory curve error, and belongs to the problem of minimum optimization under parameter identification. The whole missile simulation system needing to be identified can be regarded as a ' grey box ', and the system obeys basic physical laws, such as Newton's second law, rigid body dynamics principle and the like, namely the whole system identifies specific parameters of the missile under the condition that dynamics, kinematic equations and other mathematical models and laws are known. One typical application of ballistic registration is: air combat simulation games often require trajectory simulation and solution problems involving multiple missiles, which typically build trajectory models for each missile separately. On the premise of ensuring certain resolving accuracy, a certain trajectory simulation resolving model can be reused by setting different input parameters so as to support the trajectory simulation functions of various missiles and reduce the software complexity.
The method has the advantages of high convergence speed and small calculated amount. But also has the limitation that when the system has nonlinearity or time lag, the gradient value can not be obtained. Meanwhile, the method has the problem of sensitivity to the initial value of the parameter to be identified, and the method starts from a group of specific parameters in the optimization design, so that the optimization result is easy to sink in the vicinity of the initial point to form a local optimal solution, thereby causing great difficulty in the selection of the initial value of the parameter during optimization. And when the sensitivity is solved, some numerical problems are brought, so that the accuracy of the parameter identification result is greatly influenced, and even the divergence of the identification process is caused in serious cases, so that an effective result cannot be obtained.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a missile ash box simulator trajectory registration method in a simulation environment, which solves the problem of parameter identification by adopting a genetic algorithm, has simple and feasible identification thought and method, avoids not only complicated sensitivity solution, but also possible numerical problems, and can obviously enhance the identification accuracy, thereby effectively improving the reliability and the accuracy of parameter identification.
The technical scheme adopted by the invention for solving the technical problem comprises the following steps:
step 1, collecting the initial positions of a missile and an enemy plane at the launching moment as a group of typical design points, and repeating the process for N times to obtain N groups of typical design points;
step 2, taking the missile simulation system to be identified as a gray box simulator, collecting missile path data of the gray box simulator, and collecting a missile track position set output by the gray box simulator at a kth typical design point
Figure GDA0003756956890000021
In the formula:
Figure GDA0003756956890000022
-the i-th position coordinates of the missile output by the gray-box simulator at the k-th typical design point;
n is the number of typical design points;
t k -ballistic simulation time output by the gray-box simulator at the input kth typical design point;
t-simulation step length;
step 3, regarding the known missile model as a white-box simulator, transmitting the same input in the step 2 into the missile model, collecting missile path data output by the missile model, and collecting a missile trajectory position set output by the white-box simulator at the kth typical design point
Figure GDA0003756956890000023
Wherein:
Figure GDA0003756956890000024
-the i-th position coordinate of the missile output by the white-box simulator at the k-th typical design point;
Figure GDA0003756956890000025
-ballistic simulation time output by the white-box model at the input kth typical design point;
step 4, setting parameters to be identified in the ash box simulator
Figure GDA0003756956890000026
N p The number of the parameters to be identified is,
Figure GDA0003756956890000027
denotes the firsti parameters theta to be identified i Optimizing range of p is Represents the search interval of the parameter, satisfies mod (p) imax -p imin ,p is ) 0, mod (·) denotes a remainder operator; initializing each parameter to be identified in a random mode, wherein a chromosome coding mode is real number coding, and a random initialization method is adopted for population generation;
step 5, evaluating the output of the white box simulator under each chromosome by using the ballistic trajectory data generated by the gray box simulator, and respectively designing penalty terms generated by the incapability of registering the ballistic trajectory and the ballistic time, wherein at the jth typical design point of the ith chromosome, the penalty terms are respectively as follows:
Figure GDA0003756956890000028
Figure GDA0003756956890000031
in the formula:
Figure GDA0003756956890000032
penalty term due to failure of the ballistic trajectory to register;
Figure GDA0003756956890000033
penalty term due to the fact that the trajectory lengths output by the white-box and gray-box simulators cannot be registered in time series at the jth typical design point of the ith chromosome,
Figure GDA0003756956890000034
when the ballistic length output by the gray box simulator is larger than that of the white box simulator, the penalty term needs to be calculated,
Figure GDA0003756956890000035
when the ballistic length output by the white-box simulator is greater than the grayWhen the box simulator is used, penalty items need to be calculated;
T s -a ballistic data sampling interval;
Figure GDA0003756956890000036
two groups of trajectories passing through the ballistic data samples T in time series s The number of traces of the rear overlap;
Figure GDA0003756956890000037
two groups of trajectories passing through the ballistic data samples T in time series s The latter non-overlapping portion;
Figure GDA0003756956890000038
in the formula:
t j -ballistic simulation time of the gray-box simulator at the j-th typical design point of the input;
Figure GDA0003756956890000039
-ballistic simulation time of the white-box simulator at the jth typical design point of the ith chromosome;
Figure GDA00037569568900000310
the number of overlapped tracks of the output tracks of the white box model and the output tracks of the gray box model after sampling by a simulation step length T on a time sequence is determined at a jth typical design point of the ith chromosome;
Figure GDA00037569568900000311
at the jth typical design point of the ith chromosome, taking the maximum track number of the two tracks after the output tracks of the white box model and the gray box model are sampled by a simulation step length T on a time sequence;
fix () -round-down operation;
the fitness function of the ith chromosome at the jth typical design point is:
Figure GDA0003756956890000041
in the formula:
α 1 —
Figure GDA0003756956890000042
a penalty term coefficient;
α 2 —
Figure GDA0003756956890000043
a penalty term coefficient;
the final fitness function of the ith chromosome is
Figure GDA0003756956890000044
Step 6, judging whether the current iteration number step is equal to the set maximum iteration number maxstep, if step is equal to maxstep, stopping the algorithm, and outputting the current dominant chromosome; if step ≠ maxstep, continuing to execute the following steps;
step 7, for the g generation population theta (g) The evolution is carried out, and the specific evolution method is as follows:
is provided with
Figure GDA0003756956890000048
The solution of the ith chromosome of the g generation population is corresponding to the fitness function of f i Calculating the probability
Figure GDA0003756956890000046
For selecting the chromosome of the i-th chromosome,
Figure GDA0003756956890000047
is the maximum fitness value among all chromosomes of the population of the g-th generation;
let the cross probability P ic According to the g generation calculated in the selection strategy described aboveProbability P of individual chromosome group selection i (g) Selection of chromosome C using roulette method 1 And C 2 Corresponding fitness value of f c1 And f c2 For the ith parameter to be identified according to the cross probability P ic Operated in a crossover operation to yield two new individuals C' 1 And C' 2 Fitness value of New Individual f' c1 And f' c2 Calculated by step 5, if f' c1 ＞f c1 Then receive C' 1 (ii) a If 'f' c2 ＞f c2 Then receive C' 2 ；
Let the probability of variation P im Selecting probability P of each chromosome of the population of the g generation according to the probability P calculated in the selection strategy i (g) Selection of chromosome M using the roulette method 1 With a fitness value of f m1 For the ith parameter to be identified, according to the variation probability P im Performing mutation operation to obtain new individual M' 1 Fitness value of New Individual f' m1 Calculated by step 5, if f' m1 ＞f m1 And then accept M' 1 Otherwise, accept M 1 ；
And 8, calculating the optimal solution of the new population after the current evolution, adding 1 to the iteration number step, and returning to the step 6.
In the step 1, N is an integer in the range of [5,10 ].
In the step 2, the value range of the simulation step length T is [0.02,0.5 ].
In the step 5, the sampling interval T of the ballistic data s Take [1, 5]]An integer in between;
Figure GDA0003756956890000051
penalty term coefficient alpha 1 Taking out 2;
Figure GDA0003756956890000052
penalty term coefficient alpha 2 And taking 2.
In step 6, maxstep takes 50.
In the step 7, the cross probability P ic 0.66, mutation probability P im ＝0.34。
The invention has the beneficial effects that: the method adopts the genetic algorithm to solve the problem of parameter identification, and the identification thought and method are simple and easy to implement, thereby not only avoiding fussy sensitivity solution, but also avoiding the possible numerical problems, and also obviously enhancing the identification accuracy, thereby effectively improving the reliability and the accuracy of parameter identification. The parameter identification based on the genetic algorithm is insensitive to the initial value problem, and simultaneously does not require continuity and conductibility of the target function, and does not have the rigorous requirement of the traditional optimization method on the search space, so the method has extremely strong robustness. In addition, as a random optimization method, the genetic algorithm can simultaneously search different regions of a parameter space to be identified in each generation, and points the search direction to a region with higher probability for finding a better solution, so that the genetic algorithm has good global optimization characteristics, can simultaneously process a plurality of points in the search space, increases the possibility of converging to the global optimal solution, and is particularly suitable for processing the parameter identification problem.
Drawings
FIG. 1 is a block diagram of the system architecture of the present invention;
fig. 2 is a flow chart of the method of the present invention.
Detailed Description
The invention provides a trajectory registration method based on an improved genetic algorithm, which is used for identifying specific parameters of a missile, takes a missile model needing to be multiplexed in an air combat simulation game as a gray box simulator, takes a general missile model in the game as a white box simulator, and realizes trajectory registration by determining parameters to be identified in the white box simulator to fit a trajectory generated by the gray box simulator. The structural block diagram of the recognition system is shown in fig. 1.
The method comprises the following steps:
step 1: in the air combat simulation game, the initial positions of the missiles and the enemy planes of the missiles at the launching time are collected as a group of typical design points, and the step is repeated for N times; each set of typical design points reflects the friend or foe situation in the current environment. The value of N influences the running time of the algorithm and the track fitting precision, and N can be an integer between 5 and 10 to ensure the performance of the algorithm; taking the N groups of typical design points as input of a gray/white box simulator;
step 2: taking the missile simulation system to be identified as a gray box simulator, collecting the missile path data of the gray box simulator, and recording the data as:
Figure GDA0003756956890000061
in the formula:
B k -a set of missile trajectory positions output by the gray box simulator at the kth typical design point;
Figure GDA0003756956890000062
-the i-th position coordinates of the missile output by the gray-box simulator at the k-th typical design point;
n is the number of typical design points, and the value size is specified in the step 1;
t k -ballistic simulation time output by the gray-box model at the input kth typical design point;
t-simulation step length; the value size of the T-shaped variable can influence the running time of the algorithm and the track fitting precision, and the T-shaped variable can be in a range of [0.02,0.5] to ensure the performance of the algorithm;
and step 3: regarding the universal missile model in the game as a white-box simulator, transmitting the same input in the step 2 into the simulation model, collecting the missile path data output by the simulation model, and recording the data as the missile path data
Figure GDA0003756956890000063
Wherein:
w k -a set of missile trajectory positions output by the white-box simulator at the kth typical design point;
Figure GDA0003756956890000064
-the i-th position coordinate of the missile output by the white-box simulator at the k-th typical design point;
n is the number of typical design points, and the value size is specified in the step 1;
Figure GDA0003756956890000065
-ballistic simulation time output by the white-box model at the input kth typical design point;
t-simulation step length; the value size is consistent with that in the step 2;
and 4, step 4: initializing population, setting the parameters to be identified in the gray box model
Figure GDA0003756956890000066
N p The number of the parameters to be identified is,
Figure GDA0003756956890000067
represents the ith parameter theta to be identified i Optimizing range of p is Represents the search interval of the parameter, satisfies mod (p) imax -p imin ,p is ) 0, mod (·) denotes a remainder operator. And initializing each parameter to be identified in a random mode. The chromosome coding mode is real number coding, and the population generation adopts a random initialization method.
And 5: calculating a fitness function, evaluating the output of a white box simulator under each chromosome by using trajectory track data generated by a gray box, and respectively designing penalty items generated by incapability of registering trajectory tracks and trajectory time for improving parameter identification precision, wherein the penalty items are respectively calculated at a jth typical design point of an ith chromosome:
Figure GDA0003756956890000071
Figure GDA0003756956890000072
in the formula:
Figure GDA0003756956890000073
penalty term due to failure of the ballistic trajectory to register; the calculation method is that the white box and the gray box simulator are added up by the position errors generated by the superposition part of trajectory tracks output by the white box and the gray box on the time sequence at the jth typical design point of the ith chromosome;
Figure GDA0003756956890000074
penalty term due to the fact that the trajectory lengths output by the white-box and gray-box simulators cannot be registered in time series at the jth typical design point of the ith chromosome,
Figure GDA0003756956890000075
when the ballistic length output by the gray box simulator is larger than that of the white box simulator, the penalty term needs to be calculated,
Figure GDA0003756956890000076
when the trajectory length output by the white box simulator is greater than that of the gray box simulator, a penalty item needs to be calculated;
T s ballistic data sampling interval, typically taken as [1, 5]]An integer in between;
Figure GDA0003756956890000077
two groups of trajectories passing through the ballistic data samples T in time series s Number of traces of the rear overlap portion.
Figure GDA0003756956890000078
Two groups of trajectories passing through the ballistic data samples T in time series s The latter non-overlapping portion;
Figure GDA0003756956890000079
the expression is as follows:
Figure GDA00037569568900000710
in the formula:
t j -ballistic simulation time of the gray-box model at the input jth typical design point;
Figure GDA0003756956890000081
-ballistic simulation time of the white-box model at the jth typical design point of the ith chromosome;
t-simulation step length; the value size is consistent with that in the step 2;
Figure GDA0003756956890000082
the number of overlapped tracks of the output tracks of the white box model and the output tracks of the gray box model after sampling by a simulation step length T on a time sequence is determined at a jth typical design point of the ith chromosome;
Figure GDA0003756956890000083
at the jth typical design point of the ith chromosome, taking the maximum track number of the two tracks after the output tracks of the white box model and the gray box model are sampled by a simulation step length T on a time sequence;
T s ballistic data sampling interval, typically taken as [1, 5]]An integer in between;
fix () -round-down operation;
the fitness function of the ith chromosome at the jth typical design point is:
Figure GDA0003756956890000084
in the formula:
α 1 —
Figure GDA0003756956890000085
penalty coefficient, generally 2;
α 2 —
Figure GDA0003756956890000086
a penalty term coefficient, which is generally 2;
the final fitness function for the ith chromosome is:
Figure GDA0003756956890000087
step 6: judging whether the current iteration number step is equal to the maximum iteration number maxstep, if step is equal to maxstep, stopping the algorithm, and outputting the current dominant chromosome; if step ≠ maxstep, the following steps are continued. To take the fitting accuracy and the algorithm performance into account, maxstep can be 50.
And 7: for the g generation population theta (g) Evolution was carried out. The evolution operators are divided into two types, the first type is a selection operator, the second type is a cross operator, and the specific evolution method comprises the following steps:
1, selecting a strategy: is provided with
Figure GDA0003756956890000088
The solution of the ith chromosome of the population of the g generation corresponds to a fitness function of f i According to the probability P i (g) The calculation method for selecting the ith chromosome is as follows:
Figure GDA0003756956890000091
in the formula:
Figure GDA0003756956890000092
-maximum fitness value in all chromosomes of the population of the g-th generation;
2, a crossover operator: let the cross probability P ic Generally, take P ic 0.66, the probability P of each chromosome being selected for the population of the g-th generation calculated in the selection strategy as described above i (g) Selection of chromosome C using the roulette method 1 And C 2 Their fitness value is f c1 And f c2 For the ith parameter to be identified according to the cross probability P ic Operated in a crossover operation to yield two new individuals C' 1 And C' 2 Fitness value of New Individual f' c1 And f' c2 Calculated by step 5, if f' c1 ＞f c1 Then receive C' 1 (ii) a If' c2 ＞f c2 Then receive C' 2 。
3 mutation operator: let the probability of variation P im Taking P generally im 0.34, the probability P of each chromosome being selected for the population of the g-th generation calculated in the selection strategy as described above i (g) Selection of chromosome M using the roulette method 1 With a fitness value of f m1 For the ith parameter to be identified, according to the variation probability P im Performing mutation operation to generate new individual M' 1 Fitness value of New Individual f' m1 Calculated by step 5, if f' m1 ＞f m1 Then receive M' 1 Otherwise, accept M 1 。
And 8: and (5) calculating the optimal solution of the new population after the current evolution, adding 1 to the iteration number step, and returning to the step 6.
The flow illustrates a method of registering a trajectory. Collecting the ballistic data by taking a plurality of groups of typical design points under different conditions as the input of the ash box model; and then, transmitting the input into another missile whitebox simulator to be tuned, evaluating an output result of the missile whitebox simulator by using the acquired missile path data, and performing optimization search on key parameters in a whitebox model through an improved genetic algorithm to perform trajectory registration. The invention can obtain a more satisfactory registration result.
The flow chart of the method of the invention integrating the above steps is shown in fig. 2.
The present invention will be further described with reference to the following drawings and examples, which include, but are not limited to, the following examples.
The invention is implemented according to the flow chart shown in fig. 2 as follows:
step 1: in the air combat simulation game, the initial positions of the missiles and the enemy planes of the missiles at the launching time are collected as a group of typical design points, and the step is repeated for N times; each set of typical design points reflects the friend or foe situation in the current environment. The value of N influences the running time of the algorithm and the track fitting precision, and N can be an integer between 5 and 10 to ensure the performance of the algorithm; taking the N groups of typical design points as input of a gray/white box simulator;
and 2, step: taking a missile simulation system to be identified as a gray box simulator, and collecting missile path data by taking N groups of typical design points as the input of the simulator, wherein the data are recorded as:
Figure GDA0003756956890000101
in the formula:
B k -a set of missile trajectory positions output by the gray box simulator at the kth typical design point;
Figure GDA0003756956890000102
-the i-th position coordinates of the missile output by the gray-box simulator at the k-th typical design point;
n is the number of typical design points, and the value size is specified in the step 1;
t k -ballistic simulation time output by the gray-box model at the input kth typical design point;
t-simulation step length; the value size of the T-shaped variable can influence the running time of the algorithm and the track fitting precision, and the T-shaped variable can be in a range of [0.02,0.5] to ensure the performance of the algorithm;
and step 3: and (3) regarding a general missile model in the game as a white box simulator, transmitting the same input in the step (2) into the simulation model, collecting the missile data output by the simulation model, and recording the data as:
Figure GDA0003756956890000103
wherein:
w k -a set of missile trajectory positions output by the white-box simulator at the kth typical design point;
Figure GDA0003756956890000104
-the i-th position coordinate of the missile output by the white-box simulator at the k-th typical design point;
n is the number of typical design points, and the value size is specified in the step 1;
Figure GDA0003756956890000105
-ballistic simulation time output by the white-box model at the input kth typical design point;
t-simulation step length; the value size is consistent with that in the step 2;
and 4, step 4: initializing population, and setting parameters to be identified in the gray box model
Figure GDA0003756956890000106
N p The number of the parameters to be identified is,
Figure GDA0003756956890000107
represents the ith parameter theta to be identified i Optimizing range of (1), p is Represents the search interval of the parameter, satisfies mod (p) imax -p imin ,p is ) 0, mod (·) denotes a remainder operator. And initializing each parameter to be identified in a random mode. The chromosome coding mode is real number coding, and the population generation adopts a random initialization method.
And 5: calculating a fitness function, namely evaluating the trajectory output by the white box simulator under each chromosome by using trajectory data output by the gray box, respectively designing penalty terms generated by the fact that trajectory trajectories and trajectory time cannot be registered in order to improve parameter identification accuracy, wherein the penalty terms are respectively calculated at the jth typical design point of the ith chromosome:
Figure GDA0003756956890000111
Figure GDA0003756956890000112
in the formula:
Figure GDA0003756956890000113
penalty term due to failure of the ballistic trajectory to register; the calculation method is that the white box and the gray box simulator are added up by the position errors generated by the superposition part of trajectory tracks output by the white box and the gray box on the time sequence at the jth typical design point of the ith chromosome;
Figure GDA0003756956890000114
penalty term due to non-registration of the ballistic trajectory lengths output by the white-box and gray-box simulators in time series at the jth typical design point of the ith chromosome,
Figure GDA0003756956890000115
when the ballistic length output by the gray box simulator is larger than that of the white box simulator, the penalty term needs to be calculated,
Figure GDA0003756956890000116
the penalty item needs to be calculated when the ballistic length output by the white box simulator is greater than that of the gray box simulator;
T s sampling intervals for track data, typically taken [1, 5]]An integer in between;
Figure GDA0003756956890000117
two groups of trajectories in timeChannel data sampling T on the inter sequence s Number of traces of the rear overlap portion.
Figure GDA0003756956890000118
Two groups of trajectories pass through the ballistic data sample T in time series s The latter non-overlapping portion;
Figure GDA0003756956890000119
the expression is as follows:
Figure GDA00037569568900001110
in the formula:
t j -ballistic simulation time of the gray-box model at the input jth typical design point;
Figure GDA0003756956890000121
-ballistic simulation time of the white-box model at the jth typical design point of the ith chromosome;
t-simulation step length; the value size is consistent with that in the step 2;
Figure GDA0003756956890000122
the number of overlapped tracks of the output tracks of the white box model and the output tracks of the gray box model after sampling by a simulation step length T on a time sequence is determined at a jth typical design point of the ith chromosome;
Figure GDA0003756956890000123
at the jth typical design point of the ith chromosome, taking the maximum track number of the two tracks after the output tracks of the white box model and the gray box model are sampled by a simulation step length T on a time sequence;
T s -ballistic data sampling intervalGenerally take [1, 5]]An integer in between;
fix () -round-down operation;
the fitness function of the ith chromosome at the jth typical design point is:
Figure GDA0003756956890000124
in the formula:
α 1 —
Figure GDA0003756956890000125
penalty coefficient, generally 2;
α 2 —
Figure GDA0003756956890000126
penalty coefficient, generally 2;
the final fitness function for the ith chromosome is:
Figure GDA0003756956890000127
step 6: judging whether the current iteration number step is equal to the maximum iteration number maxstep, if step is equal to maxstep, stopping the algorithm, and outputting the current dominant chromosome; if step ≠ maxstep, the following steps continue. To take both the fitting accuracy and the algorithm performance into account, maxstep may take 50.
And 7: for the g generation population theta (g) Evolution was carried out. The evolution operators are divided into two types, the first type is a selection operator, the second type is a cross operator, and the specific evolution method comprises the following steps:
4, selecting a strategy: is provided with
Figure GDA0003756956890000128
The solution of the ith chromosome of the population of the g generation corresponds to a fitness function of f i According to the probability P i (g) The calculation method for selecting the ith chromosome is as follows:
Figure GDA0003756956890000129
in the formula:
Figure GDA0003756956890000131
-maximum fitness value among all chromosomes of the population of the g-th generation;
5, a crossover operator: let the cross probability be P ic Generally, take P ic 0.66, the probability P of each chromosome being selected for the population of the g-th generation calculated in the selection strategy as described above i (g) Selection of chromosome C using roulette method 1 And C 2 Their fitness value is f c1 And f c2 For the ith parameter to be identified according to the cross probability P ic Operated in a crossover operation to yield two new individuals C' 1 And C' 2 Fitness value of New Individual f' c1 And f' c2 Calculated by step 5, if f' c1 ＞f c1 Then receive C' 1 (ii) a If' c2 ＞f c2 Then receive C' 2 。
6 mutation operator: let the mutation probability P im Taking P generally im 0.34, the probability P of each chromosome being selected for the population of the g-th generation calculated in the selection strategy as described above i (g) Selection of chromosome M using the roulette method 1 With a fitness value of f m1 For the ith parameter to be identified, according to the variation probability P im Performing mutation operation to obtain new individual M' 1 Fitness value of New Individual f' m1 Calculated by step 5, if f' m1 ＞f m1 Then receive M' 1 Otherwise, accept M 1 。
And 8: and (5) calculating the optimal solution of the new population after the current evolution, adding 1 to the iteration number step, and returning to the step 6.
The flow illustrates a method of registering a trajectory. Collecting the ballistic data by taking a plurality of groups of typical design points under different conditions as the input of the ash box model; and then, the input is transmitted into another guided missile white box simulator to be tuned, the output result of the guided missile white box simulator is evaluated by utilizing the acquired missile path data, and the optimization search is carried out on key parameters in a white box model through an improved genetic algorithm to carry out trajectory registration. The invention can obtain a satisfactory registration result.
The flow chart of the method of the invention, which integrates the above steps, is shown in fig. 2.

Claims (6)

1. A missile ash box simulator trajectory registration method in a simulation environment is characterized by comprising the following steps:
step 1, collecting the initial positions of a missile and an enemy plane at the launching moment as a group of typical design points, and repeating the process for N times to obtain N groups of typical design points;
step 2, taking the missile simulation system to be identified as a gray box simulator, collecting missile path data of the gray box simulator, and collecting missile track position sets output by the gray box simulator at the kth typical design point
Figure FDA0003756956880000011
In the formula:
Figure FDA0003756956880000012
-the j position coordinate of the missile output by the gray box simulator at the k typical design point;
n is the number of typical design points;
t k -ballistic simulation time output by the gray-box simulator at the input kth typical design point;
t-simulation step length;
step 3, regarding the known missile model as a white-box simulator, transmitting the same input in the step 2 into the missile model, collecting missile path data output by the missile model, and collecting a missile trajectory position set output by the white-box simulator at the kth typical design point
Figure FDA0003756956880000013
Wherein:
Figure FDA0003756956880000014
the j position coordinate of the missile output by the white box simulator at the k typical design point;
Figure FDA0003756956880000015
-ballistic simulation time output by the white-box model at the input kth typical design point;
step 4, setting parameters to be identified in the ash box simulator
Figure FDA0003756956880000016
N p The number of the parameters to be identified is,
Figure FDA0003756956880000017
represents the ith parameter theta to be identified i Optimizing range of p is Represents the search interval of the parameter, satisfies mod (p) imax -p imin ,p is ) 0, mod (·) denotes a remainder operator; initializing each parameter to be identified in a random mode, wherein a chromosome coding mode is real number coding, and a random initialization method is adopted for population generation;
step 5, evaluating the output of the white box simulator under each chromosome by using the ballistic trajectory data generated by the gray box simulator, and respectively designing penalty terms generated by the incapability of registering the ballistic trajectory and the ballistic time, wherein at the jth typical design point of the ith chromosome, the penalty terms are respectively as follows:
Figure FDA0003756956880000021
Figure FDA0003756956880000022
in the formula:
Figure FDA0003756956880000023
penalty term due to failure of the ballistic trajectory to register;
Figure FDA0003756956880000024
penalty term due to non-registration of the ballistic trajectory lengths output by the white-box and gray-box simulators in time series at the jth typical design point of the ith chromosome,
Figure FDA0003756956880000025
when the ballistic length output by the gray box simulator is larger than that of the white box simulator, the penalty term needs to be calculated,
Figure FDA0003756956880000026
when the trajectory length output by the white box simulator is greater than that of the gray box simulator, a penalty item needs to be calculated;
T s -a ballistic data sampling interval;
Figure FDA0003756956880000027
two groups of trajectories passing through the ballistic data samples T in time series s The number of traces of the rear overlap;
Figure FDA0003756956880000028
two groups of trajectories pass through the ballistic data sample T in time series s The latter non-overlapping portion;
Figure FDA0003756956880000029
in the formula:
t j -ballistic simulation time of the gray-box simulator at the j-th typical design point of the input;
Figure FDA00037569568800000210
-ballistic simulation time of the white-box simulator at the jth typical design point of the ith chromosome;
Figure FDA00037569568800000211
the number of overlapped tracks of the output tracks of the white box model and the gray box model after sampling by a simulation step length T on a time sequence is determined at a jth typical design point of the ith chromosome;
Figure FDA00037569568800000212
at the jth typical design point of the ith chromosome, taking the maximum track number of the two tracks after the output tracks of the white box model and the gray box model are sampled by a simulation step length T on a time sequence;
fix () -round-down operation;
the fitness function of the ith chromosome at the jth typical design point is:
Figure FDA0003756956880000031
in the formula:
α 1 —
Figure FDA0003756956880000032
a penalty term coefficient;
α 2 —
Figure FDA0003756956880000033
a penalty term coefficient;
the final fitness function of the ith chromosome is
Figure FDA0003756956880000034
Step 6, judging whether the current iteration number step is equal to the set maximum iteration number maxstep or not, if step is maxstep, stopping the algorithm, and outputting the current dominant chromosome; if step ≠ maxstep, continuing to execute the following steps;
step 7, for the g generation population theta (g) The evolution is carried out, and the specific evolution method is as follows:
is provided with
Figure FDA0003756956880000035
The solution of the ith chromosome of the population of the g generation corresponds to a fitness function of f i Calculating the probability
Figure FDA0003756956880000036
For selecting the chromosome of the i-th chromosome,
Figure FDA0003756956880000037
is the maximum fitness value among all chromosomes of the population of the g-th generation;
let the cross probability P ic Selecting probability P of each chromosome of the population of the g generation according to the probability P calculated in the selection strategy i (g) Selection of chromosome C using the roulette method 1 And C 2 Corresponding fitness value of f c1 And f c2 For the ith parameter to be identified according to the cross probability P ic Operated in a crossover operation to yield two new individuals C' 1 And C' 2 Fitness value of New Individual f' c1 And f' c2 Calculated by step 5, if' c1 ＞f c1 Then receive C' 1 (ii) a If' c2 ＞f c2 Then receive C' 2 ；
Let the mutation probability P im The g-th generation population is stained according to the above-mentioned respective stains calculated in the selection strategyProbability P of being selected i (g) Selection of chromosome M using the roulette method 1 With a fitness value of f m1 For the ith parameter to be identified, according to the variation probability P im Performing mutation operation to obtain new individual M' 1 Fitness value of New Individual f' m1 Calculated by step 5, if f' m1 ＞f m1 Then receive M' 1 Otherwise, accept M 1 ；
And 8, calculating the optimal solution of the new population after the current evolution, adding 1 to the iteration step, and returning to the step 6.
2. The missile ash box simulator trajectory registration method in a simulated environment of claim 1, wherein: in the step 1, N is an integer in the range of [5,10 ].
3. The missile ash box simulator trajectory registration method in a simulated environment of claim 1, wherein: in the step 2, the value range of the simulation step length T is [0.02,0.5 ].
4. The missile ash box simulator trajectory registration method in a simulated environment of claim 1, wherein: in the step 5, the sampling interval T of the ballistic data s Take [1, 5]]An integer in between;
Figure FDA0003756956880000041
penalty term coefficient alpha 1 Taking out 2;
Figure FDA0003756956880000042
penalty term coefficient alpha 2 And taking 2.
5. The missile ash box simulator trajectory registration method in a simulated environment of claim 1, wherein: in step 6, maxstep takes 50.
6. The method of claim 1 in a simulated environmentThe trajectory registration method of the ash-ejecting box simulator is characterized by comprising the following steps: in the step 7, the cross probability P ic 0.66, mutation probability P im ＝0.34。
CN202010639747.7A 2020-07-06 2020-07-06 Missile ash box simulator trajectory registration method in simulation environment Active CN111888767B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010639747.7A CN111888767B (en) 2020-07-06 2020-07-06 Missile ash box simulator trajectory registration method in simulation environment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010639747.7A CN111888767B (en) 2020-07-06 2020-07-06 Missile ash box simulator trajectory registration method in simulation environment

Publications (2)

Publication Number Publication Date
CN111888767A CN111888767A (en) 2020-11-06
CN111888767B true CN111888767B (en) 2022-08-30

Family

ID=73191594

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010639747.7A Active CN111888767B (en) 2020-07-06 2020-07-06 Missile ash box simulator trajectory registration method in simulation environment

Country Status (1)

Country Link
CN (1) CN111888767B (en)

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107589665A (en) * 2017-08-25 2018-01-16 中国人民解放军火箭军装备研究院 A kind of trajectory fast Optimization and device based on gradient particle group's algorithm

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106599537B (en) * 2016-11-17 2019-03-29 西北工业大学 A kind of extensive weapon-target assignment method based on multiple target Clonal Evolutionary Algorithms
CN111240353B (en) * 2020-01-07 2021-06-15 南京航空航天大学 Unmanned aerial vehicle collaborative air combat decision method based on genetic fuzzy tree

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107589665A (en) * 2017-08-25 2018-01-16 中国人民解放军火箭军装备研究院 A kind of trajectory fast Optimization and device based on gradient particle group's algorithm

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于遗传算法的弹道优化算法;张乐蓉等;《四川兵工学报》;20130525(第05期);全文 *

Also Published As

Publication number Publication date
CN111888767A (en) 2020-11-06

Similar Documents

Publication Publication Date Title
WO2021134871A1 (en) Forensics method for synthesized face image based on local binary pattern and deep learning
CN108564129B (en) Trajectory data classification method based on generation countermeasure network
CN111126134B (en) Radar radiation source deep learning identification method based on non-fingerprint signal eliminator
CN108596327B (en) Seismic velocity spectrum artificial intelligence picking method based on deep learning
CN109085469A (en) A kind of method and system of the signal type of the signal of cable local discharge for identification
CN114170461B (en) Noise-containing label image classification method based on feature space reorganization for teacher and student architecture
CN112884059B (en) Small sample radar working mode classification method fusing priori knowledge
CN112861257B (en) Aircraft fire control system precision sensitivity analysis method based on neural network
CN111123232B (en) Radar individual identification system with task adaptability
CN107832789B (en) Feature weighting K nearest neighbor fault diagnosis method based on average influence value data transformation
CN110082738B (en) Radar target identification method based on Gaussian mixture and tensor recurrent neural network
CN110807291B (en) On-site situation future guiding technology based on mimicry countermeasure learning mechanism
CN109940614B (en) Mechanical arm multi-scene rapid motion planning method integrating memory mechanism
CN112001433A (en) Flight path association method, system, equipment and readable storage medium
CN115329594B (en) Large-scale missile cluster attack and defense confrontation simulation acceleration method and system
CN103885867B (en) Online evaluation method of performance of analog circuit
CN111144462B (en) Unknown individual identification method and device for radar signals
CN110889207B (en) Deep learning-based intelligent assessment method for credibility of system combination model
CN110223342B (en) Space target size estimation method based on deep neural network
CN114627047A (en) Improved SORT algorithm-based fish counting statistical method
CN111888767B (en) Missile ash box simulator trajectory registration method in simulation environment
CN105183624B (en) A kind of emulation back method based on Data Matching
CN112560881A (en) Object identification method and device and data processing method
CN116051960A (en) Unmanned aerial vehicle group flight behavior recognition method based on deep learning
CN114067155B (en) Image classification method, device, product and storage medium based on meta learning

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
GR01 Patent grant
GR01 Patent grant