CN115019238A - Group target dynamic behavior identification method based on hidden Markov model - Google Patents
Group target dynamic behavior identification method based on hidden Markov model Download PDFInfo
- Publication number
- CN115019238A CN115019238A CN202210779849.8A CN202210779849A CN115019238A CN 115019238 A CN115019238 A CN 115019238A CN 202210779849 A CN202210779849 A CN 202210779849A CN 115019238 A CN115019238 A CN 115019238A
- Authority
- CN
- China
- Prior art keywords
- behavior
- group target
- group
- markov model
- hidden markov
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 27
- 230000006399 behavior Effects 0.000 claims abstract description 152
- 230000015572 biosynthetic process Effects 0.000 claims abstract description 89
- 239000011159 matrix material Substances 0.000 claims abstract description 61
- 230000007704 transition Effects 0.000 claims abstract description 35
- 230000009466 transformation Effects 0.000 claims abstract description 16
- 238000004445 quantitative analysis Methods 0.000 claims abstract description 4
- 238000013507 mapping Methods 0.000 claims description 10
- 230000007123 defense Effects 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000012935 Averaging Methods 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 abstract description 12
- 238000005755 formation reaction Methods 0.000 description 75
- 230000009471 action Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000003542 behavioural effect Effects 0.000 description 2
- 241000826860 Trapezium Species 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000007363 ring formation reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/84—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using probabilistic graphical models from image or video features, e.g. Markov models or Bayesian networks
- G06V10/85—Markov-related models; Markov random fields
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Databases & Information Systems (AREA)
- Computational Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Computing Systems (AREA)
- Multimedia (AREA)
- Mathematical Analysis (AREA)
- Medical Informatics (AREA)
- Probability & Statistics with Applications (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Algebra (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Complex Calculations (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a hidden Markov model-based group target dynamic behavior identification method, which comprises the following steps: calculating Archimedes spiral coefficients of all standard formation of the group targets according to the real-time dynamic position coordinate information data of the group targets; determining a behavior state set of the group targets; constructing a group target behavior state transition probability matrix and a group target formation transformation observation probability matrix; constructing a hidden Markov model for group target behavior analysis of a continuous time sequence; calculating the Archimedes spiral coefficient of each frame of the group target to obtain an Archimedes spiral coefficient sequence; and taking the obtained Archimedes spiral coefficient sequence as an observation sequence to be brought into a hidden Markov model, deducing the behavior of the group target, and carrying out quantitative analysis and identification on the group target behavior. The invention constructs a hidden Markov model to realize the purpose of carrying out quantitative identification and analysis on the group target behaviors, and realizes the continuous dynamic analysis and identification of the group target behaviors and the acquisition of the battlefield situation.
Description
Technical Field
The invention belongs to the field of group target identification, relates to the technologies of group target formation identification, group target behavior quantitative description and group target behavior analysis, and particularly relates to a hidden Markov model-based group target dynamic behavior identification method.
Background
When the ships finish their missions, in order to achieve certain effects, formation is often formed, such as battle formation and carrier formation, to concentrate the forces to achieve the intended purpose. The formation of ships is suitable for various battle formations, such as a longitudinal formation, a transverse formation, a herringbone formation and the like, and is the basis of various task formations. The ship formation is the concrete embodiment of centralizing the force of force, and the effect is obvious. According to the battlefield battle situation, the enemy formation is correctly identified and predicted, the enemy machine for converting the formation is seized to launch the attack, and the ships of our party are reasonably organized, properly formed, flexibly directed and skillfully maneuvered to achieve the best battle effect.
In future wars, with more and more exquisite weaponry, the formation scale of the naval vessels tends to be reduced, and when the naval vessels fight, the formation is more evacuated, and the maneuvering is more flexible. In order to improve the quick response capability of the ships, the sailing formation and the fighting formation tend to be consistent gradually. The array of sea wars can properly exert the integral advantages, analyze the enemy array, grasp the enemy array transformation machine, find attack points and realize the breakthrough of the war bureau.
The formation is used for concentrating the force, and the formation is a necessary form for exerting firepower. The group target behaviors have certain dependency on the formation of the group targets, the standard formation of the group targets is quantitatively described and analyzed, and the behaviors of the group targets are analyzed and recognized on the basis of the quantitative description and the analysis.
Disclosure of Invention
The purpose of the invention is as follows: in order to solve the problems that template matching, Bayesian networks, behavior recognition based on expert knowledge and the like are mostly adopted in the existing group target behavior recognition, the group target behavior recognition depends too much on the expert knowledge and does not have the real-time performance of dynamic analysis, a group target dynamic behavior recognition method based on a hidden Markov model is provided, and continuous dynamic analysis recognition of the group target behavior is realized to obtain the battlefield situation; based on Archimedes spiral coefficients, a relation probability matrix between a standard group target formation and a group target behavior and a state transition matrix of a group target behavior state set are established, and a hidden Markov model is established to realize the purpose of carrying out quantitative recognition analysis on the group target behavior.
The technical scheme is as follows: in order to achieve the above object, the present invention provides a hidden markov model-based group target dynamic behavior identification method, which comprises the following steps:
s1: calculating Archimedes spiral coefficients of all standard formation of the group targets according to the real-time dynamic position coordinate information data of the group targets;
s2: determining a behavior state set of the group target;
s3: according to the behavior state set of the group target, a group target behavior state transition probability matrix and a group target formation transformation observation probability matrix are constructed;
s4: constructing a hidden Markov model for analyzing the group target behaviors of the continuous time sequence according to the group target behavior state transition probability matrix and the group target formation transformation observation probability matrix;
s5: acquiring the position coordinates of continuous frames of a group target, and calculating the Archimedes spiral coefficient of each frame of the group target to obtain an Archimedes spiral coefficient sequence;
s6: and taking the obtained Archimedes spiral coefficient sequence as an observation sequence to be brought into a hidden Markov model, deducing the group target behavior, and carrying out quantitative analysis and identification on the group target behavior.
Further, the step S1 specifically includes the following steps:
a1: determining a classic standard formation of group targets, e.g., a platoon, a column, a wedge formation, a trapeze formation, etc.;
a2: acquiring horizontal and vertical coordinate information of each formation;
a3: and calculating the Archimedes spiral coefficient of each classical standard group target formation.
Further, the method for calculating the archimedes' spiral coefficient in step a3 includes:
b1: converting the (x, y) coordinates to polar coordinates (ρ, θ);
b2: calculating the Archimedes spiral coefficient of each object in the group target, wherein the calculation formula is as follows:
a＝(ρ 1 -ρ 2 )/(θ 1 -θ 2 +2*pi)
wherein pi has a value of pi, wherein (p) 1 ,θ 1 ) Is the polar coordinate of an object within a cluster object, (p) 2 ,θ 2 ) Polar coordinates of another object within a cluster target;
b3: and averaging the Archimedes spiral coefficients between every two objects in the group target to serve as the final Archimedes spiral coefficient of the group target at the moment.
Further, the behavior state set of the group target in the step S2 includes attack behavior, convoy behavior, defense behavior, shield behavior and retreat behavior.
Further, the method for constructing the group target behavior state transition probability matrix in step S3 includes:
c1: constructing a state transition matrix A with the size of m × m for the group targets, wherein m is the number of behaviors of the group targets to be researched;
c2: determining probability values of the group targets from the behaviors e to f by using historical empirical data, determining probability values of the group targets from the behaviors e to f, and storing the probability values into positions corresponding to the state transition matrix, for example, P 12 The probability value is the state transition from behavior one to behavior two.
Further, the method for constructing the observation probability matrix of the group target formation transformation in the step S3 includes:
d1: constructing an observation probability matrix B with the size of m x n for the group targets, wherein m is the number of behaviors of the group targets to be researched, and n is the number of classical standard formation of the group targets to be researched;
d2: by usingHistorical empirical data determines the probability value of each classical standard formation that a group target is likely to achieve for action e, and stores the probability value in a position corresponding to an observed probability matrix, for example, P 12 And the probability value of the formation where the group target finishes the action one time is the formation two is represented.
Further, the hidden markov model constructed in the step S4 is composed of a hidden state chain, an observation chain, a state transition matrix a, an observation probability matrix B, and an initial state chain;
hidden Markov model hidden state chain is a behavior state set of the group target; and the observation chain of the hidden Markov model is a group target formation Archimedes spiral coefficient sequence.
Further, the step S6 is specifically:
e1: the group target behavior state transition matrix A and the group target formation transformation observation probability matrix B form a mapping probability relation from a standard formation Archimedes spiral coefficient to a group target behavior, and the specific matrix formula form is as follows:
the mapping relation is as follows: a and B;
e2: and deducing the group target behavior from the Archimedes spiral coefficient information in the observation chain based on the mapping relation, and finishing the behavior recognition result of the group target.
Has the advantages that: compared with the prior art, the method realizes the continuous dynamic analysis and identification of the group target behaviors and the acquisition of the battlefield situation, and solves the problems that the existing group target behavior identification is too dependent on expert knowledge and does not have the real-time performance of dynamic analysis; based on Archimedes spiral coefficients, a relation probability matrix between a standard group target formation and a group target behavior and a state transition matrix of a group target behavior state set are established, and a hidden Markov model is established to realize the purpose of carrying out quantitative recognition analysis on the group target behavior.
Drawings
FIG. 1 is an overall flow diagram of the present invention;
FIG. 2 is a dynamic change diagram of an Archimedes spiral of 55 continuous frames of a ship group target;
fig. 3 is a diagram showing the recognition result of the behavior of a certain ship group at certain times within a 55-frame time period of the certain ship group.
Detailed Description
The present invention is further illustrated by the following figures and specific examples, which are to be understood as illustrative only and not as limiting the scope of the invention, which is to be given the full breadth of the appended claims and any and all equivalent modifications thereof which may occur to those skilled in the art upon reading the present specification.
The invention provides a hidden Markov model-based group target dynamic behavior identification method, which comprises the following steps as shown in figure 1:
s1: calculating Archimedes spiral coefficients of all standard formation of the group targets according to the real-time dynamic position coordinate information data of the group targets, and specifically comprising the following steps:
a1: determining a classical standard formation of group targets, e.g., a platoon, a column, a wedge, a trapezium, etc.;
a2: acquiring horizontal and vertical coordinate information of each formation;
a3: calculating the Archimedes spiral coefficient of each classical standard group target formation, wherein the calculation method comprises the following steps:
b1: converting the (x, y) coordinates to polar coordinates (ρ, θ);
b2: calculating the Archimedes spiral coefficient of each object in the group target, wherein the calculation formula is as follows:
a＝(ρ 1 -ρ 2 )/(θ 1 -θ 2 +2*pi)
wherein pi has a value of pi, wherein (p) 1 ,θ 1 ) Is the polar coordinate of an object within a cluster object, (p) 2 ,θ 2 ) Is the polar coordinate of some other object within a cluster target;
b3: and averaging the Archimedes spiral coefficients between every two objects in the group target to serve as the final Archimedes spiral coefficient of the group target at the moment.
S2: determining a behavioral state set of the group targets:
the set of behavioral states of the group target includes an attack behavior, a convoy behavior, a defense behavior, a shield behavior, and a retreat behavior.
S3: according to the behavior state set of the group target, constructing a group target behavior state transition probability matrix and a group target formation transformation observation probability matrix:
the method for constructing the group target behavior state transition probability matrix comprises the following steps:
c1: constructing a state transition matrix A with the size of m × m for the group targets, wherein m is the number of behaviors of the group targets to be researched;
c2: determining probability values of the group targets from the behaviors e to f by using historical empirical data, determining probability values of the group targets from the behaviors e to f, and storing the probability values into positions corresponding to the state transition matrix, for example, P 12 The probability value is the state transition from behavior one to behavior two.
The method for constructing the observation probability matrix of the group target formation transformation comprises the following steps:
d1: constructing an observation probability matrix B with the size of m x n for the group targets, wherein m is the number of behaviors of the group targets to be researched, and n is the number of classical standard formation of the group targets to be researched;
d2: determining the probability value of each classical standard formation that the group target may be in to achieve the purpose of behavior e by using historical empirical data, and storing the probability value in a position corresponding to an observation probability matrix, for example, P 12 And the probability value of the formation where the group target finishes the action one time is the formation two.
S4: and constructing a hidden Markov model for analyzing the group target behaviors of the continuous time sequence according to the group target behavior state transition probability matrix and the group target formation transformation observation probability matrix:
the constructed hidden Markov model is composed of a hidden state chain, an observation chain, a state transition matrix A, an observation probability matrix B and an initial state chain;
hidden Markov model hidden state chain is a behavior state set of the group target; and the observation chain of the hidden Markov model is a group target formation Archimedes spiral coefficient sequence.
S5: acquiring the position coordinates of continuous frames of a group target, and calculating the Archimedes spiral coefficient of each frame of the group target to obtain an Archimedes spiral coefficient sequence;
s6: taking the obtained Archimedes spiral coefficient sequence as an observation sequence to be brought into a hidden Markov model, deducing the behavior of the group target, and carrying out quantitative analysis and identification on the group target behavior, wherein the method specifically comprises the following steps:
e1: the group target behavior state transition matrix A and the group target formation transformation observation probability matrix B form a mapping probability relation from a standard formation Archimedes spiral coefficient to a group target behavior, and the specific matrix formula form is as follows:
the mapping relation is as follows: a and B;
e2: and deducing the group target behavior from the Archimedes spiral coefficient information in the observation chain based on the mapping relation, and finishing the behavior recognition result of the group target.
In this embodiment, the method is applied to dynamic behavior recognition of a ship group, and the specific process is as follows:
the method comprises the following steps: calculating the Archimedes spiral coefficient of the standard formation of the ship group to be researched, which is as follows:
(1a) determining the classic standard formation commonly used by a ship group in the sea warfare, namely a horizontal formation, a longitudinal formation, a wedge formation, a trapezoid formation and a ring formation;
(1b) acquiring coordinate position information (x, y) of each standard formation;
(1c) calculating the Archimedes spiral coefficient A of the standard formation of each ship group 1 ,A 2 ,A 3 ,A 4 ,A 5 Wherein A is 1 ,A 2 ,A 3 ,A 4 ,A 5 The Archimedes spiral coefficient values of the horizontal team, the vertical team, the wedge team, the trapezoid team and the annular team respectively; wherein, the step of calculating the Archimedes spiral coefficient comprises the following steps:
(1c1) first converting the (x, y) coordinates to polar coordinates (ρ, θ);
(1c2) the formula for calculating the coefficient of the Archimedes spiral is as follows:
a＝(ρ 1 -ρ 2 )/(θ 1 -θ 2 +2*pi)
wherein pi has a value of pi, wherein (p) 1 ,θ 1 ) Is the polar coordinate of a ship in the ship group (rho) 2 ,θ 2 ) Is the polar coordinate of another ship in the ship group.
(1c3) And averaging the Archimedes spiral coefficients calculated between every two ships in the ship group to serve as the final Archimedes spiral coefficient of the group target at the moment.
Step two: determining a state set in a hidden Markov model aiming at a ship group, wherein the state set comprises an attack behavior, a convoy behavior, a defense behavior, a shield behavior and a retreat behavior. Abstraction is a notation to represent a state set as S ═ S 1 ,S 2 ,S 3 ,S 4 ,S 5 In which S is 1 ,S 2 ,S 3 ,S 4 ,S 5 Respectively corresponding to the convoy behavior, the attack behavior, the defense behavior, the shield behavior and the retreat behavior of the ship group;
step three: constructing a ship group behavior state transition probability matrix, namely a state transition probability from an attack behavior to a defense behavior, a state transition probability from a shield behavior to a retreat behavior, and a state transition probability from the attack behavior to the retreat behavior, wherein the specific conditions are as follows:
(3a) constructing a state transition matrix A with the size of mxm for the group target, wherein m is the number of the behaviors of the ship group to be researched;
(3b) determining probability values of the group targets from behavior one to behavior two by using historical empirical data, determining probability values of the group targets from behavior one to behavior two, and storing the probability values in positions corresponding to the state transition matrix, for example, P 12 Storing the position of a first row and a second column of a state transition matrix for the state transition probability value from the convoy behavior to the attack behavior;
step four: constructing a ship formation transformation observation probability matrix, wherein the probability of a longitudinal formation is the probability of a wedge formation under attack; under the defense behavior, the probability of being a horizontal team is the probability of being a longitudinal team; the method comprises the following specific steps:
(4a) constructing an observation probability matrix B with the size of mxn for a group target, wherein m is the number of behaviors of the ship group to be researched, and n is the number of standard formation of the ship group to be researched;
(4b) determining probability values of various standard formation forms possibly located by the ship group and storing the probability values into a position corresponding to an observation probability matrix, for example, P 12 Representing the probability value of a longitudinal formation when the ship group finishes the convoy behavior, and storing the probability value in the positions of a first row and a second row of an observation probability matrix;
step five: and constructing a hidden Markov model for ship group behavior analysis based on the second step to the fourth step. The hidden Markov hidden chain is the behavior of each ship group, including the behavior of escort, attack, defense, shield and retreat; the observation chain is the Archimedes spiral coefficient of each standard formation, namely a horizontal formation, a vertical formation, a wedge formation, a trapezoid formation and an annular formation; the method comprises the following specific steps:
(5a) establishing a mapping probability relation between a ship group standard formation Archimedes spiral coefficient and ship group behaviors by using the behavior state transition matrix A and the formation transformation observation probability matrix B of the group targets established in the third step and the fourth step, and establishing a hidden Markov model for identifying and analyzing the ship group behaviors;
(5b) the observation chain of the hidden Markov model is the Archimedes spiral coefficient of the ship group at each continuous moment, and the hidden state chain is the behavior of the ship group at each continuous moment.
Step six: obtaining the position coordinates of continuous frames of a ship group, and calculating the Archimedes spiral coefficient A of continuous 55 frames of the ship group 1 ,…A 55 The calculation method is as in steps (1c1) - (1c3), and dynamic changes of the archimedean spiral of 55 continuous frames of the ship group target shown in fig. 2 are obtained;
step seven: the acquired archimedes spiral coefficient sequence is taken as an observation sequence to be brought into a constructed hidden markov model for the ship group, the ship group behavior is quantitatively analyzed, the behavior of the ship group is deduced, ship group behavior sequence sets { S1, S2, S3, S2, S5 and S1} at certain moments identified in a 55-frame time period are output, and the ship group behavior identification result acquired in the embodiment is specifically shown in fig. 3.
Claims (8)
1. A hidden Markov model-based group target dynamic behavior identification method is characterized by comprising the following steps:
s1: calculating Archimedes spiral coefficients of all standard formation of the group targets according to the real-time dynamic position coordinate information data of the group targets;
s2: determining a behavior state set of the group target;
s3: according to the behavior state set of the group target, a group target behavior state transition probability matrix and a group target formation transformation observation probability matrix are constructed;
s4: constructing a hidden Markov model for analyzing the group target behaviors of the continuous time sequence according to the group target behavior state transition probability matrix and the group target formation transformation observation probability matrix;
s5: acquiring the position coordinates of continuous frames of a group target, and calculating the Archimedes spiral coefficient of each frame of the group target to obtain an Archimedes spiral coefficient sequence;
s6: and taking the obtained Archimedes spiral coefficient sequence as an observation sequence to be brought into a hidden Markov model, deducing the behavior of the group target, and carrying out quantitative analysis and identification on the group target behavior.
2. The hidden markov model-based group target dynamic behavior recognition method according to claim 1, wherein the step S1 specifically comprises the steps of:
a1: determining a classic standard formation of the group target;
a2: acquiring horizontal and vertical coordinate information of each formation;
a3: and calculating the Archimedes spiral coefficient of each classical standard group target formation.
3. The hidden markov model-based group target dynamic behavior identification method according to claim 2, wherein the archimedes spiral coefficient in step a3 is calculated by:
b1: converting the (x, y) coordinates to polar coordinates (ρ, θ);
b2: calculating the Archimedes spiral coefficient of each object in the group target, wherein the calculation formula is as follows:
a＝(ρ 1 -ρ 2 )/(θ 1 -θ 2 +2*pi)
wherein pi has a value of pi, wherein (p) 1 ,θ 1 ) Is the polar coordinate of an object within a cluster object, (p) 2 ,θ 2 ) Is the polar coordinate of some other object within a cluster target;
b3: and averaging the Archimedes spiral coefficients between every two objects in the group target to serve as the final Archimedes spiral coefficient of the group target at the moment.
4. The hidden markov model-based group target dynamic behavior identification method of claim 1, wherein the behavior state set of the group target in the step S2 comprises attack behavior, convoy behavior, defense behavior, shield behavior and retreat behavior.
5. The hidden markov model-based group target dynamic behavior identification method according to claim 1, wherein the group target behavior state transition probability matrix in the step S3 is constructed by:
c1: constructing a state transition matrix A with the size of m × m for the group targets, wherein m is the number of behaviors of the group targets to be researched;
c2: and determining the probability value of the group target from the behavior e to the behavior f by using historical empirical data, determining the probability value of the group target from the behavior e to the behavior f, and storing the probability value in a position corresponding to the state transition matrix.
6. The hidden markov model-based group target dynamic behavior recognition method of claim 1, wherein the group target formation transformation observation probability matrix in step S3 is constructed by the following steps:
d1: constructing an observation probability matrix B with the size of m x n for the group targets, wherein m is the number of behaviors of the group targets to be researched, and n is the number of classical standard formation of the group targets to be researched;
d2: and determining the probability values of each classical standard formation possibly positioned by the group target according to the historical experience data to realize the purpose of the behavior e, and storing the probability values into the corresponding positions of the observation probability matrix.
7. The hidden markov model-based group target dynamic behavior recognition method of claim 1, wherein the hidden markov model constructed in the step S4 is composed of a hidden state chain, an observation chain, a state transition matrix a, an observation probability matrix B, and an initial state chain;
hidden Markov model hidden state chain is a behavior state set of the group target; and the observation chain of the hidden Markov model is a group target formation Archimedes spiral coefficient sequence.
8. The hidden markov model-based group target dynamic behavior identification method according to claim 7, wherein the step S6 specifically comprises:
e1: the group target behavior state transition matrix A and the group target formation transformation observation probability matrix B form a mapping probability relation from a standard formation Archimedes spiral coefficient to a group target behavior, and the specific matrix formula form is as follows:
the mapping relation is as follows: a and B;
e2: and deducing the group target behavior from the Archimedes spiral coefficient information in the observation chain based on the mapping relation, and finishing the behavior recognition result of the group target.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210779849.8A CN115019238B (en) | 2022-07-04 | Group target dynamic behavior identification method based on hidden Markov model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210779849.8A CN115019238B (en) | 2022-07-04 | Group target dynamic behavior identification method based on hidden Markov model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115019238A true CN115019238A (en) | 2022-09-06 |
| CN115019238B CN115019238B (en) | 2024-09-24 |
Family
ID=
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116502055A (en) * | 2023-01-10 | 2023-07-28 | 昆明理工大学 | Multi-dimensional characteristic dynamic abnormal integral model based on quasi-Markov model |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103150611A (en) * | 2013-03-08 | 2013-06-12 | 北京理工大学 | Hierarchical prediction method of II type diabetes mellitus incidence probability |
| CN108170680A (en) * | 2017-12-29 | 2018-06-15 | 厦门市美亚柏科信息股份有限公司 | Keyword recognition method, terminal device and storage medium based on Hidden Markov Model |
| US20190025851A1 (en) * | 2017-07-21 | 2019-01-24 | AI Incorporated | Polymorphic path planning for robotic devices |
| CN111914910A (en) * | 2020-07-15 | 2020-11-10 | 南京邮电大学 | Airplane formation identification method based on multi-observation-point shape context |
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103150611A (en) * | 2013-03-08 | 2013-06-12 | 北京理工大学 | Hierarchical prediction method of II type diabetes mellitus incidence probability |
| US20190025851A1 (en) * | 2017-07-21 | 2019-01-24 | AI Incorporated | Polymorphic path planning for robotic devices |
| CN108170680A (en) * | 2017-12-29 | 2018-06-15 | 厦门市美亚柏科信息股份有限公司 | Keyword recognition method, terminal device and storage medium based on Hidden Markov Model |
| CN111914910A (en) * | 2020-07-15 | 2020-11-10 | 南京邮电大学 | Airplane formation identification method based on multi-observation-point shape context |
Non-Patent Citations (1)
| Title |
|---|
| 李海敏;: "反潜机采用阿基米德螺旋线搜潜问题研究", 指挥控制与仿真, no. 06, 15 December 2014 (2014-12-15), pages 47 - 51 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116502055A (en) * | 2023-01-10 | 2023-07-28 | 昆明理工大学 | Multi-dimensional characteristic dynamic abnormal integral model based on quasi-Markov model |
| CN116502055B (en) * | 2023-01-10 | 2024-05-03 | 昆明理工大学 | Multi-dimensional characteristic dynamic abnormal integral model based on quasi-Markov model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110929394B (en) | Combined combat system modeling method based on super network theory and storage medium | |
| CN110348708B (en) | Ground target dynamic threat assessment method based on extreme learning machine | |
| CN113435644B (en) | Emergency prediction method based on deep bidirectional long-short term memory neural network | |
| CN107330560A (en) | A kind of multitask coordinated distribution method of isomery aircraft for considering temporal constraint | |
| CN112418171B (en) | Zebra fish spatial attitude and heart position estimation method based on deep learning | |
| CN102222240B (en) | DSmT (Dezert-Smarandache Theory)-based image target multi-characteristic fusion recognition method | |
| CN108415045B (en) | Method and device for searching marine moving target by remote sensing satellite | |
| CN116050795A (en) | Unmanned ship cluster task scheduling and collaborative countermeasure method based on MADDPG | |
| CN115525058A (en) | Unmanned underwater vehicle cluster cooperative countermeasure method based on deep reinforcement learning | |
| CN113324545A (en) | Multi-unmanned aerial vehicle collaborative task planning method based on hybrid enhanced intelligence | |
| CN115019238A (en) | Group target dynamic behavior identification method based on hidden Markov model | |
| CN109299491B (en) | Meta-model modeling method based on dynamic influence graph strategy and using method | |
| CN115019238B (en) | Group target dynamic behavior identification method based on hidden Markov model | |
| CN115755971A (en) | Cooperative confrontation task allocation method for sea-air integrated unmanned intelligent equipment | |
| CN110390337B (en) | Ship individual identification method | |
| CN113255234B (en) | Method for carrying out online target distribution on missile groups | |
| Lu et al. | Control of divergence in an extended invertible logistic map | |
| CN113128698B (en) | Reinforced learning method for multi-unmanned aerial vehicle cooperative confrontation decision | |
| Liu et al. | Effective 3D boundary learning via a nonlocal deformable network | |
| Ni et al. | 3D real-time path planning for AUV based on improved bio-inspired neural network | |
| Wang et al. | Coactive design of human-machine collaborative damage assessment using UAV images and decision trees | |
| Tian et al. | Application of Full Connection Network in Submarine Formation Recognition | |
| Chen et al. | Ship recognition based on improved forwards-backwards algorithm | |
| Zhi et al. | SAR maritime object recognition based on convolutional neural network | |
| Li et al. | Double Deep Q-learning for Anti-saturation Attack Problem of Warship Group |
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 |