WO2021174765A1 - 基于多无人机协同博弈对抗的控制系统 - Google Patents

基于多无人机协同博弈对抗的控制系统 Download PDF

Info

Publication number
WO2021174765A1
WO2021174765A1 PCT/CN2020/108774 CN2020108774W WO2021174765A1 WO 2021174765 A1 WO2021174765 A1 WO 2021174765A1 CN 2020108774 W CN2020108774 W CN 2020108774W WO 2021174765 A1 WO2021174765 A1 WO 2021174765A1
Authority
WO
WIPO (PCT)
Prior art keywords
uav
confrontation
module
current
game
Prior art date
Application number
PCT/CN2020/108774
Other languages
English (en)
French (fr)
Inventor
刘振
蒲志强
丘腾海
易建强
Original Assignee
中国科学院自动化研究所
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 中国科学院自动化研究所 filed Critical 中国科学院自动化研究所
Priority to US17/261,036 priority Critical patent/US11669110B2/en
Publication of WO2021174765A1 publication Critical patent/WO2021174765A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/10Simultaneous control of position or course in three dimensions
    • G05D1/101Simultaneous control of position or course in three dimensions specially adapted for aircraft
    • G05D1/104Simultaneous control of position or course in three dimensions specially adapted for aircraft involving a plurality of aircrafts, e.g. formation flying
    • FMECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F41WEAPONS
    • F41HARMOUR; ARMOURED TURRETS; ARMOURED OR ARMED VEHICLES; MEANS OF ATTACK OR DEFENCE, e.g. CAMOUFLAGE, IN GENERAL
    • F41H13/00Means of attack or defence not otherwise provided for
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/0088Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/0094Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots involving pointing a payload, e.g. camera, weapon, sensor, towards a fixed or moving target

Definitions

  • the invention belongs to the field of multi-UAV game confrontation, and specifically relates to a control system based on multi-UAV cooperative game confrontation.
  • Multi-UAV cluster technology has become a key technology in the field of UAV applications.
  • multi-UAV coordinated game confrontation can make full use of the reconnaissance, strike and evaluation capabilities of each drone, and improve the overall combat effectiveness and mission execution efficiency of the UAV system, thus becoming the trend of future air combat confrontation.
  • Multi-UAV cooperative game confrontation means that two or more drones cooperate with each other to complete the designated confrontation task.
  • the research of UAV game confrontation was mainly 1V1.
  • the multi-aircraft cooperative game confrontation there are such things as rapid time-varying environmental information, unstable communication transmission, multi-UAV anti-collision, multi-UAV information fusion and task allocation, and UAV group coordinated action decision-making, etc.
  • the conventional single-machine game confrontation system is difficult to directly apply to the research of multi-machine game confrontation.
  • the first aspect of the present invention proposes a multi-unmanned A control system for aircraft cooperative game confrontation.
  • the system includes a management module and a UAV formation module for both sides of the game, a situation assessment module, a decision-making module, and a cooperative task allocation module:
  • the management module is configured to store the first information sent by the drone formation module and send it to the situation assessment module;
  • the UAV formation module is configured to obtain the status information of each UAV in the current party and the UAV formation of the opposing party when the UAV of the opposing party is found during the cruise, as the first information, and send it to
  • the management module is also configured to control the drones of the current party to execute the control instructions sent by the coordinated task allocation module;
  • the situation assessment module is configured to obtain the situation assessment information of the current drones based on the first information and by pre-setting multiple preset types of assessment methods;
  • the decision-making module is configured to count all the maneuvering strategies of the drones of each party based on the acquired number of drones of the two parties in the game, and combine the situation assessment information to obtain the relative differences between the drones of the current party and the opponent.
  • the optimal situational advantage value of the UAV is to construct a situation matrix of the current UAV formation relative to the opponent UAV formation; based on the situation matrix, obtain the current confrontation strategy through a preset strategy selection method;
  • the cooperative task allocation module is configured to obtain the confrontation target that the current drone responds to according to the situation matrix, and combine the confrontation strategy and the optimal situation evaluation value to generate the current drone's
  • the control instruction is sent to the UAV formation module.
  • system further includes a visual display module
  • the visual display module is configured to obtain the first state information and the game confrontation image for output display.
  • the management module is further configured to perform system settings and game confrontation performance analysis
  • the system settings include simulation scheduling management, multi-UAV system initialization of both game parties, and discrete event trigger management; wherein, the simulation scheduling management includes setting management of system simulation duration, step size, simulation start time, and simulation end time;
  • the initialization of the multi-UAV system of the two parties in the game is configured to initially set the flight status of the multi-UAVs of the two parties in the game;
  • the game confrontation performance analysis includes winning rate result analysis, process trend analysis, and confrontation time analysis; wherein, the winning rate result analysis is configured to display the winning rates of both parties in the game in the form of a graph; the process trend analysis is configured to display the game in the form of a curve The trend of the situation of the two parties; the confrontation time analysis is configured to display the game time of the two parties in the form of a graph.
  • the status information includes the position, speed, attitude angle, and control input information of the drone; the control input information includes the longitudinal overload, normal overload, and roll angle of the drone.
  • the situation assessment information includes maneuverability assessment information and visual field capability assessment information;
  • the visual field capability assessment information includes a distance assessment value, an azimuth angle assessment value, and an entry angle assessment value;
  • the maneuverability assessment The information includes energy evaluation value and air combat performance evaluation value.
  • the calculation method of the distance evaluation value is:
  • T i D is the distance evaluation value
  • D i , D Rmax , D Mmax , D Mmin , D MKmax , and D MKmin represent the relative distance and fire distance between the ith drone of the current party and any drone of the opponent.
  • the calculation method of the azimuth angle evaluation value is:
  • ⁇ i , ⁇ Rmax , ⁇ Mmax , ⁇ MKmax respectively represent the azimuth angle of the current i-th UAV and any UAV of the opposing party, the maximum search azimuth of the fire control radar, and the maximum search azimuth of the air-to-air missile Angle, the maximum non-escape angle of the air-to-air missile, T i ⁇ is the azimuth angle evaluation value.
  • the calculation method of the entry angle evaluation value is:
  • T i p is the entry angle evaluation value
  • the calculation method of the energy evaluation value is:
  • I represents the energy value of UAVs
  • H i is the i-drone flying height
  • V i is the i UAV flight speed
  • g is the gravitational acceleration coefficient
  • E T is any one of the enemy's
  • T i E is the energy evaluation value of the i-th UAV.
  • the calculation method of the air combat performance evaluation value is:
  • B i, A i, D i represent the i-th parameter mobility UAVs, fire detection, and the ability to measure the parameters measured parameter, with Respectively represent the control efficiency coefficient, survivability coefficient, and range coefficient of the i-th UAV, and T i c is the air combat performance evaluation value of the i-th UAV.
  • the decision-making module “get the optimal situational advantage values of the current drones relative to the opponent drones, and construct the current drone formation relative to the opponent drones. Formation situation matrix”, its method is:
  • the maneuvering strategy includes drones Horizontal maneuvering strategy and longitudinal maneuvering strategy, in which the horizontal maneuvering strategy determines the change value of the UAV's track deflection angle, and the longitudinal maneuvering strategy determines the UAV height change value;
  • the optimal situation advantage value of the current drones relative to the opponent drones is obtained through the min-max principle, and the current drone formation relative to the The situation matrix of the opposing drone formation.
  • the value is compared with the set minimum and maximum situational advantage threshold to obtain the confrontation strategy of the current UAV formation.
  • the method is: corresponding to the maximum value sequentially selected according to the situation matrix To obtain the opponent's drone that the current drone responds to.
  • the invention provides a simple and fast simulation environment for the design, verification and evaluation of multi-UAV cooperative game confrontation.
  • the present invention evaluates the situation information of the current drones relative to the opponents' drones by analyzing the status information of the drones of the enemy and the enemy, and combines the possibilities of the drones of the enemy and the enemy.
  • the emergence of the maneuver strategy combination obtains the optimal situational advantage value and obtains the confrontation strategy of the current UAV formation.
  • the task instructions of each drone are assigned, and the tactical execution process is visualized through the display module, which provides a simple and fast simulation environment for the design, verification and evaluation of multi-UAV cooperative game confrontation.
  • the game confrontation control system of the present invention adopts a modular design idea, and the model, perception, decision-making and distribution modules are independently designed, and the scalability is strong.
  • the model, perception, decision-making and distribution modules are independently designed, and the scalability is strong.
  • perception algorithms, decision-making algorithms, and allocation algorithms there is no need to make major changes to the system framework. You only need to replace the corresponding modules under the condition of ensuring the same input and output data format, which improves the coordination of multiple UAVs. Simulation efficiency of air combat tactics.
  • FIG. 1 is a schematic diagram of the framework of a control system based on multi-UAV cooperative game confrontation according to an embodiment of the present invention
  • FIG. 2 is a schematic diagram of the software interface of the visual display module of the control system based on multi-UAV cooperative game confrontation according to an embodiment of the present invention
  • Fig. 3 is a schematic diagram of a control flow of a control system based on a multi-UAV cooperative game confrontation according to an embodiment of the present invention.
  • a control system based on multi-UAV cooperative game confrontation of the present invention includes the following modules: a management module and a UAV formation module of both parties in the game, a situation assessment module, a decision-making module, and a cooperative task allocation module:
  • the management module is configured to store the first information sent by the drone formation module and send it to the situation assessment module;
  • the UAV formation module is configured to obtain the status information of each UAV in the current party and the UAV formation of the opposing party when the UAV of the opposing party is found during the cruise, as the first information, and send it to
  • the management module is also configured to control the drones of the current party to execute the control instructions sent by the coordinated task allocation module;
  • the situation assessment module is configured to obtain the situation assessment information of the current drones based on the first information and by pre-setting multiple preset types of assessment methods;
  • the decision-making module is configured to count all the maneuvering strategies of the drones of each party based on the acquired number of drones of the two parties in the game, and combine the situation assessment information to obtain the relative differences between the drones of the current party and the opponent.
  • the optimal situational advantage value of the UAV is to construct a situation matrix of the current UAV formation relative to the opponent UAV formation; based on the situation matrix, obtain the current confrontation strategy through a preset strategy selection method;
  • the cooperative task allocation module is configured to obtain the confrontation target that the current drone responds to according to the situation matrix, and combine the confrontation strategy and the optimal situation evaluation value to generate the current drone's
  • the control instruction is sent to the UAV formation module.
  • the system of the present invention includes a visual display module, a management module, and a UAV formation module of both parties in the game, a situation assessment module, a decision-making module, and a collaborative task allocation module. Since the functions of the UAV formation module, situation assessment module, decision module and collaborative task allocation module of both parties in the game are the same, in this embodiment, either one of them is described.
  • the two sides of the game are defined as red and blue, and the number of drones is N.
  • the management module is configured to store the first information sent by the UAV formation module and send it to the situation assessment module.
  • the management module is configured to obtain the status information of the drones of both parties in the game, store it, and send it to the situation assessment module; it is also configured to perform system settings and game confrontation performance analysis;
  • the system settings include simulation scheduling management, multi-UAV system initialization and discrete event trigger management for both sides of the game; simulation scheduling management, including the setting management of system simulation duration, step size, simulation start time, and simulation end time; there are no two sides of the game.
  • the man-machine system is initialized and configured to initially set the flight status of the multiple drones of both parties in the game; the discrete event trigger management is configured to perform manual discrete event settings to stop the game process between the multiple drones
  • Game confrontation performance analysis includes win rate result analysis, process trend analysis and confrontation time analysis.
  • Win rate analysis configured to display the winning rate of both parties in the game in the form of a chart
  • process trend analysis configured to display the trend of the situation of both parties in the game in the form of a curve
  • confrontation time analysis configured to display the time of the game between the two parties in the form of a chart.
  • the UAV formation module is configured to obtain the status information of each UAV in the current party and the UAV formation of the opposing party when the UAV of the opposing party is found during the cruise, as the first information, and send it to
  • the management module is also configured to control the current drones to execute the control instructions sent by the coordinated task distribution module.
  • the UAV formation module further includes the dynamics and kinematics model of each UAV, flight control system, detection system and missile model.
  • the UAV dynamics and kinematics model is shown in formula (1):
  • X i, Y i, Z i is the i UAVs position information
  • V i is the flight speed
  • ⁇ i, ⁇ i are track path angle and declination
  • n ix, n iz, ⁇ i Respectively represent longitudinal overload, normal overload and roll angle.
  • the flight control system includes a speed channel controller, an altitude channel controller, and a track declination controller. The details are as follows:
  • u i1 is the speed channel control quantity
  • u i2 is the altitude channel control quantity
  • u i3 is the track deflection angle control quantity.
  • k iV > 0 is the speed controller parameter
  • e iV V i -V ic
  • Vic is the flight speed command of the i-th UAV
  • g is the gravitational acceleration coefficient
  • k iz , k i ⁇ > 0 are the altitude controller parameters
  • ⁇ ic represents the flight track inclination command of the i-th UAV
  • e iz Z i -Z ic
  • e i ⁇ ⁇ i - ⁇ ic
  • Z ic is the flight altitude command of the i-th UAV.
  • k i ⁇ > 0 is the track deflection angle controller parameter
  • e i ⁇ ⁇ i - ⁇ ic
  • ⁇ ic is the flight track deflection angle command of the i-th UAV.
  • the detection system model consists of the maximum search distance of the fire control radar, the maximum search azimuth of the fire control radar, the maximum attack distance of the air-to-air missile, the minimum attack distance of the air-to-air missile, the maximum search azimuth of the air-to-air missile, the maximum non-escape distance, and the air-to-air missile.
  • the missile model includes the missile dynamics and kinematics model and the guidance model. Among them, the missile dynamics and kinematics model is shown in formula (8):
  • Is the position information of the jth missile Is the flight speed, ballistic inclination and ballistic deflection of the jth missile, with Respectively represent the overload of the three axes of the missile.
  • the status information of the drone includes: the position, speed, attitude angle, and control input information of the drone; the control input information includes the longitudinal overload, normal overload, and roll angle of the drone.
  • the situation assessment module is configured to obtain the situation assessment information of the current drones based on the first information and by pre-setting multiple preset types of assessment methods.
  • situation assessment information includes maneuverability assessment information and visual field ability assessment information
  • visual field ability assessment information includes distance assessment value, azimuth angle assessment value and entry angle assessment value.
  • the calculation method of the distance evaluation value is shown in formula (10):
  • T i D is the distance evaluation value
  • D i , D Rmax , D Mmax , D Mmin , D MKmax , and D MKmin represent the relative distance and fire distance between the ith drone of the current party and any drone of the opponent.
  • ⁇ i , ⁇ Rmax , ⁇ Mmax , ⁇ MKmax respectively represent the azimuth angle of the current i-th UAV and any UAV of the opposing party, the maximum search azimuth of the fire control radar, and the maximum search azimuth of the air-to-air missile Angle, the maximum non-escape angle of the air-to-air missile, T i ⁇ is the azimuth angle evaluation value.
  • T i p is the entry angle evaluation value
  • the maneuverability evaluation information includes an energy evaluation value and an air combat performance evaluation value.
  • I represents the energy value of UAVs
  • H i is the i-drone flying height
  • V i is the i UAV flight speed
  • g is the gravitational acceleration coefficient
  • E T is any one of the enemy's
  • T i E is the energy evaluation value of the i-th UAV.
  • B i, A i, D i represent the i-th parameter mobility UAVs, fire detection, and the ability to measure the parameters measured parameter, with Respectively represent the control efficiency coefficient, survivability coefficient, and range coefficient of the i-th UAV, and T i c is the air combat performance evaluation value of the i-th UAV.
  • the decision-making module is configured to count all the maneuvering strategies of the drones of each party based on the acquired number of drones of the two parties in the game, and combine the situation assessment information to obtain the relative differences between the drones of the current party and the opponent.
  • the optimal situation advantage value of the UAV constructs a situation matrix of the current UAV formation relative to the opponent UAV formation; based on the situation matrix, the current confrontation strategy is obtained through a preset strategy selection method.
  • the confrontation strategy is decided based on the situation assessment information, and the overall task instruction of the multi-UAV system of the party is generated.
  • the specific treatment is as follows:
  • the number of UAVs participating in the game confrontation is n and m respectively.
  • Each UAV in the current side has N A maneuvering strategies (the maneuvering strategy is divided into UAV horizontal maneuvering strategy and vertical maneuvering strategy. Among them, the horizontal maneuver strategy determines the UAV's track deflection angle change value, and the longitudinal maneuver strategy determines the UAV height change value).
  • Each UAV of the adversary has M A types of maneuver strategies to construct the current ith unmanned aircraft.
  • the situation matrix S ij of the aircraft relative to the j-th UAV of the opposing party is shown in equation (15):
  • the confrontation strategy of the current UAV formation is decided, that is, the overall mission command of the multi UAV system of the current party.
  • the maximum value of the situation matrix S (that is, the maximum element value) is selected as S T1 , then the elements in the row and column where S T1 is removed, and the remaining maximum element value of the situation matrix S is taken as S T2 , according to
  • S T3 ..., S Tn , where n is a natural number, representing the number of maximum element values, construct a set of situational advantage values, and sum them to obtain the current UAV formation relative to the opponent UAV formation
  • the overall situational advantage value S T is shown in formula (17):
  • the cooperative task allocation module is configured to obtain the confrontation target that the current drone responds to according to the situation matrix, and combine the confrontation strategy and the optimal situation evaluation value to generate the current drone's
  • the control instruction is sent to the UAV formation module.
  • the current The i-th UAV of the party is assigned to respond to the j-th UAV of the opposing party, and then the i-th UAV of the current party is obtained according to the situation matrix S ij of the current i-th UAV relative to the j-th UAV of the opposing party
  • a collection of maneuver strategies for drones Combining with the confrontation strategy, the optimal maneuvering strategy is selected, and the task instruction of the drone is generated, and sent to the drone formation module to perform the corresponding task.
  • the visual display module is configured to obtain the first state information and the game confrontation image for output display.
  • multi-UAV cooperative game confrontation images and status information of each UAV are output in real time.
  • ID is the number of the drone
  • Longitude and Latitude are latitude and longitude
  • Course angle is the heading angle
  • Height is the flight height
  • Velocity is the flight speed
  • ACC is the flight acceleration
  • the button can control whether the drone flight trajectory is displayed or not
  • the "Save Trajectory” and “Import Trajectory” buttons can save the drone flight trajectory to a specified directory or import the drone flight trajectory file from the specified directory
  • read mission and
  • the “write task” button is used to read and write the mission instructions of the drone formation
  • the “two-dimensional simulation animation” button is used to dynamically display the game confrontation process of the drone formation of both parties
  • the “add waypoint” and “delete waypoint” are used to manually set the position information of the drone
  • the "Start”, “Pause”, “0.5x speed”, “2x speed” and “4x speed” buttons
  • the execution flow of the control system based on the multi-UAV cooperative game confrontation proposed by the present invention is shown in Figure 3:
  • the local drone formation performs cruise missions in a fixed formation.
  • the target is the target, it starts to perform a cooperative game confrontation task, and then conducts a situation assessment based on the status information of the drone formations of both parties, and decides the confrontation strategy of the own formation based on the situation assessment information, and then generates a single drone mission instruction in the formation.
  • the opponent's UAV is in the non-escape zone of the current missile, the current missile system is activated, and the missile is launched to strike the enemy according to the missile model.
  • control system based on the multi-UAV cooperative game confrontation only uses the division of the above-mentioned functional modules as an example.
  • the above-mentioned functions can be assigned to different functions as required.
  • Functional modules are implemented, that is, the modules or steps in the embodiments of the present invention are further decomposed or combined.
  • the modules of the above-mentioned embodiments can be combined into one module, or further divided into multiple sub-modules to complete all or the steps described above. Part of the function.
  • the names of the modules and steps involved in the embodiments of the present invention are only for distinguishing each module or step, and are not regarded as improper limitations on the present invention.

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Business, Economics & Management (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Game Theory and Decision Science (AREA)
  • Medical Informatics (AREA)
  • General Engineering & Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)

Abstract

一种基于多无人机协同博弈对抗的控制系统,旨在解决1V1博弈对抗系统难以满足多无人机协同博弈对抗研究需求的问题,为多无人机协同博弈对抗提供了仿真环境。控制系统包括管理模块以及博弈双方的无人机编队模块、态势评估模块、决策模块、协同任务分配模块;管理模块,配置为存储无人机编队获取的状态信息;无人机编队模块,配置为获取无人机的状态信息以及执行控制指令;态势评估模块,配置为根据状态信息获取态势评估信息;决策模块,配置为基于态势评估信息,获取对抗策略;协同任务分配模块,配置为基于对抗策略,结合对抗目标以及最优态势评估值,生成各无人机的控制指令。

Description

基于多无人机协同博弈对抗的控制系统 技术领域
本发明属于多无人机博弈对抗领域,具体涉及一种基于多无人机协同博弈对抗的控制系统。
背景技术
随着无人机面临的任务更加复杂化和多样化,单架无人机越来越难以满足需求,多无人机集群技术成为无人机应用领域的关键技术。尤其在空战博弈对抗领域,多无人机协同博弈对抗能充分利用各无人机的侦察、打击和评估能力,提高无人机系统整体作战效能和任务执行效率,因而成为未来空战对抗的趋势。为了模拟多无人机协同博弈对抗过程,实现对战场环境和武器装备的交互式操作,从而验证多机空战战术对抗的效果,需要构建基于多无人机协同博弈对抗的控制系统。
多无人机协同博弈对抗是指两架或两架以上的无人机相互配合完成指定对抗任务。在无人机博弈对抗研究初期,由于受近距空战影响,无人机博弈对抗研究以1V1为主。但是,在多机协同博弈对抗中,存在着诸如环境信息快时变、通信传输不稳定、多架无人机防碰撞、多架无人机信息融合与任务分配、无人机群协同动作决策等多个问题,因此,常规的单机博弈对抗系统难以直接应用于多机博弈对抗研究。
发明内容
为了解决现有技术中的上述问题,即为了解决现有1V1无人机博弈对抗系统难以满足多无人机协同博弈对抗研究需求的问题,本发明第一方面,提出了一种基于多无人机协同博弈对抗的控制系统,该 系统包括管理模块以及博弈双方的无人机编队模块、态势评估模块、决策模块、协同任务分配模块:
所述管理模块,配置为存储所述无人机编队模块发送的第一信息并发送至所述态势评估模块;
所述无人机编队模块,配置为当巡航过程中发现对抗方的无人机时,获取当前方、对抗方无人机编队中的各无人机的状态信息,作为第一信息并发送至所述管理模块;还配置为控制当前方各无人机执行所述协同任务分配模块发送的控制指令;
所述态势评估模块,配置为基于所述第一信息,通过预设多种预设类别的评估方法获取当前方各无人机的态势评估信息;
所述决策模块,配置为基于获取的博弈双方各无人机的数量,统计各方无人机所有的机动策略,并结合所述态势评估信息,获取当前方各无人机相对于对抗方各无人机的最优态势优势值,构建当前方无人机编队相对于对抗方无人机编队的态势矩阵;基于所述态势矩阵,通过预设的策略选取方法获取当前方的对抗策略;
所述协同任务分配模块,配置为根据所述态势矩阵,获取当前方无人机所应对的对抗目标,并结合所述对抗策略、所述最优态势评估值,生成当前方各无人机的控制指令,发送所述无人机编队模块。
在一些优选的实施方式中,该系统还包括视景显示模块;
所述视景显示模块,配置为获取所述第一状态信息及博弈对抗图像进行输出显示。
在一些优选的实施方式中,所述管理模块,还配置为进行系统设置和博弈对抗性能分析;
所述系统设置包括仿真调度管理、博弈双方多无人机系统初始化和离散事件触发管理;其中,所述仿真调度管理,包括系统仿真时长、步长、仿真开始时间、仿真结束时间的设置管理;所述博弈双方多 无人机系统初始化,配置为对博弈双方多无人机的飞行状态进行初始设置;所述离散事件触发管理,配置为进行人工离散事件设置,以中止多无人机双方的博弈进程;
所述博弈对抗性能分析包括胜率结果分析、过程趋势分析和对抗时间分析;其中,所述胜率结果分析,配置为以图表形式显示博弈双方胜率;所述过程趋势分析,配置为以曲线形式显示博弈双方态势变化趋势;所述对抗时间分析,配置为以图表形式显示双方博弈时间。
在一些优选的实施方式中,所述状态信息包括无人机的位置、速度、姿态角和控制输入信息;所述控制输入信息包括无人机的纵向过载、法向过载和滚转角。
在一些优选的实施方式中,所述态势评估信息包括机动能力评估信息和视野能力评估信息;所述视野能力评估信息包括距离评估值、方位角评估值和进入角评估值;所述机动能力评估信息包括能量评估值和空战性能评估值。
在一些优选的实施方式中,所述距离评估值其计算方法为:
Figure PCTCN2020108774-appb-000001
其中,T i D为距离评估值,D i、D Rmax、D Mmax、D Mmin、D MKmax、D MKmin分别表示当前方第i架无人机与对抗方任一无人机的相对距离、火控雷达的最大搜索距离、空空导弹的最大攻击距离、空空导弹的最小攻击距离、空空导弹的最大不可逃逸距离、空空导弹的最小不可逃逸距离。
在一些优选的实施方式中,所述方位角评估值其计算方法为:
Figure PCTCN2020108774-appb-000002
其中,Φ i、Φ Rmax、Φ Mmax、Φ MKmax分别表示当前方第i架无人机与对抗方任一无人机的方位角、火控雷达的最大搜索方位角、空空导弹的最大搜索方位角、空空导弹的最大不可逃逸角,T i Φ为方位角评估值。
在一些优选的实施方式中,所述进入角评估值其计算方法为:
Figure PCTCN2020108774-appb-000003
其中,p i
Figure PCTCN2020108774-appb-000004
分别表示当前方第i架无人机与对抗方任一无人机的进入角和进入角阈值,T i p为进入角评估值。
在一些优选的实施方式中,所述能量评估值其计算方法为:
Figure PCTCN2020108774-appb-000005
其中,
Figure PCTCN2020108774-appb-000006
表示第i架无人机的能量值,H i为第i架无人机的飞行高度,V i为第i架无人机的飞行速度,g为重力加速度系数,E T为对抗方任一无人机的能量值,T i E为第i架无人机的能量评估值。
在一些优选的实施方式中,所述空战性能评估值其计算方法为:
Figure PCTCN2020108774-appb-000007
其中,B i、A i、D i分别表示第i架无人机的机动性参数、火力衡量参数和探测能力衡量参数,
Figure PCTCN2020108774-appb-000008
Figure PCTCN2020108774-appb-000009
分别表示第i架无人机的操纵效能系数、生存力系数、航程系数,T i c为第i架无人机的空战性能评估值。
在一些优选的实施方式中,所述决策模块中“获取当前方各无人机相对于对抗方各无人机的最优态势优势值,构建当前方无人机编队相对于对抗方无人机编队的态势矩阵”,其方法为:
基于当前方、对抗方各无人机的机动策略,结合所述态势评估信息,构建当前方各无人机相对于对抗方各无人机的第一态势矩阵;所述机动策略包括无人机横向机动策略和纵向机动策略,其中,横向机动策略决定无人机航迹偏角变化值,纵向机动策略决定无人机高度变化值;
基于所述第一态势矩阵中的态势优势值,通过min-max原则得到当前方各无人机相对于对抗方各无人机的最优态势优势值,并构建当前方无人机编队相对于对抗方无人机编队的态势矩阵。
在一些优选的实施方式中,所述决策模块中“基于所述态势矩阵,通过预设的策略选取方法获取当前方的对抗策略”,其方法为:
依次选取所述态势矩阵中最大值,并删除所述态势矩阵中最大值对应行和列的所有元素;
对选取的最大值进行累加,得到当前方无人机编队相对于对抗方无人机编队的整体态势优势值;
基于所述整体态势优势值,将该值与设定的最小最大态势优势阈值进行比对,获取当前方无人机编队的对抗策略。
在一些优选的实施方式中,所述协同任务分配模块中“根据所述态势矩阵,获取当前方无人机所应对的对抗目标”,其方法为:根据所述态势矩阵依次选取的最大值对应的行列坐标,获取当前方无人机所应对的对抗方的无人机。
本发明的有益效果:
本发明为多无人机协同博弈对抗的设计、验证与评估提供了简单、快速的仿真环境。本发明针对不同的空战场景与任务,通过分析敌我双方各无人机的状态信息,评估当前方各无人机相对于对抗方各无人机的态势信息,并结合敌我双方各无人机可能出现的机动策略组合,获取最优的态势优势值,得到当前方无人机编队的对抗策略。基于对抗策略,分配各无人机的任务指令,并通过显示模块实现战术执行过程的可视化,为多无人机协同博弈对抗的设计、验证与评估提供了简单、快速的仿真环境。
同时,本发明的博弈对抗控制系统采用模块化设计思路,模型、感知、决策与分配模块独立设计,可扩展性强。针对不同的无人机模型、感知算法、决策算法以及分配算法,无需对系统框架进行大幅改动,只需在保证输入输出数据格式一致的条件下替换相应模块即可,提高了多无人机协同空战战术的仿真效率。
附图说明
通过阅读参照以下附图所做的对非限制性实施例所做的详细描述,本申请的其他特征、目的和优点将会变得更明显。
图1是本发明一种实施例的基于多无人机协同博弈对抗的控制系统的框架示意图;
图2是本发明一种实施例的基于多无人机协同博弈对抗的控制系统的视景显示模块软件界面的示意图;
图3是本发明一种实施例的基于多无人机协同博弈对抗的控制系统的控制流程示意图。
具体实施方式
为使本发明的目的、技术方案和优点更加清楚,下面将结合附图对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例是本发明一部分实施例,而不是全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都属于本发明保护的范围。
下面结合附图和实施例对本申请作进一步的详细说明。可以理解的是,此处所描述的具体实施例仅用于解释相关发明,而非对该发明的限定。另外还需要说明的是,为了便于描述,附图中仅示出了与有关发明相关的部分。
需要说明的是,在不冲突的情况下,本申请中的实施例及实施例中的特征可以相互组合。
本发明的一种基于多无人机协同博弈对抗的控制系统,包括以下模块:管理模块以及博弈双方的无人机编队模块、态势评估模块、决策模块、协同任务分配模块:
所述管理模块,配置为存储所述无人机编队模块发送的第一信息并发送至所述态势评估模块;
所述无人机编队模块,配置为当巡航过程中发现对抗方的无人机时,获取当前方、对抗方无人机编队中的各无人机的状态信息,作为第一信息并发送至所述管理模块;还配置为控制当前方各无人机执行所述协同任务分配模块发送的控制指令;
所述态势评估模块,配置为基于所述第一信息,通过预设多种预设类别的评估方法获取当前方各无人机的态势评估信息;
所述决策模块,配置为基于获取的博弈双方各无人机的数量,统计各方无人机所有的机动策略,并结合所述态势评估信息,获取当前方各无人机相对于对抗方各无人机的最优态势优势值,构建当前方无人机编队相对于对抗方无人机编队的态势矩阵;基于所述态势矩阵,通过预设的策略选取方法获取当前方的对抗策略;
所述协同任务分配模块,配置为根据所述态势矩阵,获取当前方无人机所应对的对抗目标,并结合所述对抗策略、所述最优态势评估值,生成当前方各无人机的控制指令,发送所述无人机编队模块。
为了更清晰地对本发明基于多无人机协同博弈对抗的控制系统进行说明,下面结合附图对本发明系统一种实施例中各功能模块进行展开详述。
如图1所示,本发明系统包括视景显示模块、管理模块以及博弈双方的无人机编队模块、态势评估模块、决策模块、协同任务分配模块。由于博弈双方的无人机编队模块、态势评估模块、决策模块及协同任务分配模块功能一致,本实施例中,任选一方进行描述。图1中将博弈双方定义为红方、蓝方,无人机数量为N。
所述管理模块,配置为存储所述无人机编队模块发送的第一信息并发送至所述态势评估模块。
在本实施例中,管理模块,配置为获取博弈双方各无人机的状态信息进行存储并发送至所述态势评估模块;还配置为进行系统设置和博弈对抗性能分析;
其中,系统设置包括仿真调度管理、博弈双方多无人机系统初始化和离散事件触发管理;仿真调度管理,包括系统仿真时长、步长、仿真开始时间、仿真结束时间的设置管理;博弈双方多无人机系统初始 化,配置为对博弈双方多无人机的飞行状态进行初始设置;所述离散事件触发管理,配置为进行人工离散事件设置,以中止多无人机双方的博弈进程
博弈对抗性能分析包括胜率结果分析、过程趋势分析和对抗时间分析。胜率结果分析,配置为以图表形式显示博弈双方胜率;过程趋势分析,配置为以曲线形式显示博弈双方态势变化趋势;对抗时间分析,配置为以图表形式显示双方博弈时间。
所述无人机编队模块,配置为当巡航过程中发现对抗方的无人机时,获取当前方、对抗方无人机编队中的各无人机的状态信息,作为第一信息并发送至所述管理模块;还配置为控制当前方各无人机执行所述协同任务分配模块发送的控制指令。
在本实施例中,无人机编队模块进一步包括各无人机的动力学与运动学模型、飞行控制系统、探测系统和导弹模型。其中,无人机动力学与运动学模型如式(1)所示:
Figure PCTCN2020108774-appb-000010
其中,X i、Y i、Z i为第i架无人机位置信息,V i为飞行速度,γ i、ψ i分别为航迹倾角和航迹偏角,n ix、n iz、φ i分别代表纵向过载、法向过载和滚转角。
飞行控制系统包括速度通道控制器、高度通道控制器、航迹偏角控制器。具体如下所示:
首先,选择虚拟控制量,如式(2)所示:
Figure PCTCN2020108774-appb-000011
其中,u i1为速度通道控制量,u i2为高度通道控制量,u i3为航迹偏角控制量。
其中,u i1的计算过程如公式(3)所示:
Figure PCTCN2020108774-appb-000012
其中,k iV>0为速度控制器参数,e iV=V i-V ic,V ic为第i架无人机的飞行速度指令,g为重力加速度系数。
u i2的计算过程如公式(4)(5)所示:
Figure PCTCN2020108774-appb-000013
Figure PCTCN2020108774-appb-000014
其中,k iz,k >0为高度控制器参数,γ ic表示第i架无人机的飞行航迹倾角指令,e iz=Z i-Z ic,e =γ iic,Z ic为第i架无人机的飞行高度指令。
u i3的计算过程如公式(6)所示:
Figure PCTCN2020108774-appb-000015
其中,k >0为航迹偏角控制器参数,e =ψ iic,ψ ic为第i架无人机的飞行航迹偏角指令。
基于u i1、u i2、u i3对虚拟控制量进行解耦,其解耦得到的结如式(7)所示:
Figure PCTCN2020108774-appb-000016
探测系统模型由火控雷达的最大搜索距离、火控雷达的最大搜索方位角、空空导弹的最大攻击距离、空空导弹的最小攻击距离、空 空导弹的最大搜索方位角、最大不可逃逸距离、空空导弹的最小不可逃逸距离和空空导弹的最大不可逃逸角构成。
导弹模型包括导弹动力学与运动学模型和制导模型,其中,导弹动力学与运动学模型如公式(8)所示:
Figure PCTCN2020108774-appb-000017
其中,
Figure PCTCN2020108774-appb-000018
为第j架导弹位置信息,
Figure PCTCN2020108774-appb-000019
为第j架导弹的飞行速度、弹道倾角和弹道偏角,
Figure PCTCN2020108774-appb-000020
Figure PCTCN2020108774-appb-000021
分别代表导弹三个轴向的过载。
制导模型如公式(9)所示:
Figure PCTCN2020108774-appb-000022
其中,
Figure PCTCN2020108774-appb-000023
为第j架导弹速度矢量方向的变化率,q j为视线角速率,K为导引系数。
在本实施例中,无人机的状态信息包括:无人机的位置、速度、姿态角和控制输入信息;控制输入信息包括无人机的纵向过载、法向过载和滚转角。
所述态势评估模块,配置为基于所述第一信息,通过预设多种预设类别的评估方法获取当前方各无人机的态势评估信息。
在本实施例中,基于博弈双方多无人机的状态信息,提取影响对抗能力的双方要素,评估当前方态势。其中态势评估信息包括机动能力评估信息和视野能力评估信息;视野能力评估信息包括距离评估值、方位角评估值和进入角评估值。其中,距离评估值其计算方法如式(10)所示:
Figure PCTCN2020108774-appb-000024
其中,T i D为距离评估值,D i、D Rmax、D Mmax、D Mmin、D MKmax、D MKmin分别表示当前方第i架无人机与对抗方任一无人机的相对距离、火控雷达的最大搜索距离、空空导弹的最大攻击距离、空空导弹的最小攻击距离、空空导弹的最大不可逃逸距离、空空导弹的最小不可逃逸距离。
方位角评估值其计算方法如式(11)所示:
Figure PCTCN2020108774-appb-000025
其中,Φ i、Φ Rmax、Φ Mmax、Φ MKmax分别表示当前方第i架无人机与对抗方任一无人机的方位角、火控雷达的最大搜索方位角、空空导弹的最大搜索方位角、空空导弹的最大不可逃逸角,T i Φ为方位角评估值。
进入角评估值其计算方法如式(12)所示:
Figure PCTCN2020108774-appb-000026
其中,p i
Figure PCTCN2020108774-appb-000027
分别表示当前方第i架无人机与对抗方任一无人机的进入角和进入角阈值,T i p为进入角评估值。
机动能力评估信息包括能量评估值和空战性能评估值。
其中,能量评估值其计算方法如式(13)所示:
Figure PCTCN2020108774-appb-000028
其中,
Figure PCTCN2020108774-appb-000029
表示第i架无人机的能量值,H i为第i架无人机的飞行高度,V i为第i架无人机的飞行速度,g为重力加速度系数,E T为对抗方任一无人机的能量值,T i E为第i架无人机的能量评估值。
空战性能评估值其计算方法如式(14)所示:
Figure PCTCN2020108774-appb-000030
其中,B i、A i、D i分别表示第i架无人机的机动性参数、火力衡量参数和探测能力衡量参数,
Figure PCTCN2020108774-appb-000031
Figure PCTCN2020108774-appb-000032
分别表示第i架无人机的操纵效能系数、生存力系数、航程系数,T i c为第i架无人机的空战性能评估值。
所述决策模块,配置为基于获取的博弈双方各无人机的数量,统计各方无人机所有的机动策略,并结合所述态势评估信息,获取当前方各无人机相对于对抗方各无人机的最优态势优势值,构建当前方无人机编队相对于对抗方无人机编队的态势矩阵;基于所述态势矩阵,通过预设的策略选取方法获取当前方的对抗策略。
在本实施例中,基于态势评估信息决策对抗策略,生成本方多无人机系统的总任务指令。具体处理如下:
参与博弈对抗的当前方和对抗方无人机数量分别为n和m,其中,当前方每架无人机有N A种机动策略(机动策略分为无人机横向机动策略和纵向机动策略,其中,横向机动策略决定无人机航迹偏角变化 值,纵向机动策略决定无人机高度变化值),对抗方每架无人机有M A种机动策略,构造当前方第i架无人机相对于对抗方第j架无人机的态势矩阵S ij,如式(15)所示:
Figure PCTCN2020108774-appb-000033
其中,
Figure PCTCN2020108774-appb-000034
为对抗方第j架无人机执行第b种机动策略,当前方第i架无人机选择第a种机动策略与之进行对抗时,当前方第i架无人机的态势优势值。
基于S ij,根据min-max原则(即获取第一态势矩阵中每一行的最小值,然后在得到的各行的最小值里取最大值),可以得出当前方第i架无人机相对于对抗方第j架无人机的最优态势优势值S ij(o),并构建当前方无人机编队相对于对抗方无人机编队的态势矩阵S,如式(16)所示:
Figure PCTCN2020108774-appb-000035
基于态势矩阵S,决策当前方无人机编队的对抗策略,即本方多无人机系统的总任务指令。在本发明实施例中,选取态势矩阵S最大值(即最大元素值)记为S T1,然后去掉S T1所在行和列的元素,再取态势矩阵S剩余最大元素值记为S T2,依此类推,得到S T3、…、S Tn,n为自然数,表示最大元素值的数量,构建态势优势值集合,并进行求和,得到当前方无人机编队相对于对抗方无人机编队的整体态势优势值S T,如式(17)所示:
Figure PCTCN2020108774-appb-000036
Figure PCTCN2020108774-appb-000037
时,本方无人机编队的对抗策略为迎头攻击战术; 当
Figure PCTCN2020108774-appb-000038
时,本方无人机编队的对抗策略为诱饵战术;当S TS时,本方无人机编队的对抗策略为防御分合战术。其中, S
Figure PCTCN2020108774-appb-000039
代表无人机编队整体态势优势值的两个阈值,当整体态势优势值大于
Figure PCTCN2020108774-appb-000040
时,表示该无人机编队态势占优,当整体态势优势值小于 S时,表示该无人机编队态势处劣。
所述协同任务分配模块,配置为根据所述态势矩阵,获取当前方无人机所应对的对抗目标,并结合所述对抗策略、所述最优态势评估值,生成当前方各无人机的控制指令,发送所述无人机编队模块。
在本实施例中,根据S Ti所在态势矩阵S中的行数i和列数j,(表明当前方第i架无人机对对抗方第j架无人机的优势最大),得出当前方第i架无人机分配应对对抗方第j架无人机,再根据当前方第i架无人机相对于对抗方第j架无人机的态势矩阵S ij,得到当前方第i架无人机的机动策略集合。结合对抗策略,选择最优机动策略,生成无人机的任务指令,发送至所述无人机编队模块执行相应任务。
所述视景显示模块,配置为获取所述第一状态信息及博弈对抗图像进行输出显示。
在本实施例中,实时输出多无人机协同博弈对抗图像及各无人机的状态信息。如图2所示:ID为无人机的编号,Longitude、Latitude为经纬度,Course angle为航向角,Height为飞行高度,Velocity为飞行速度,ACC为飞行加速度,“显示路径”和“关闭路径”按钮可控制显示与否无人机飞行轨迹,“保存轨迹”和“导入轨迹”按钮可将无人机飞行轨迹保存至指定目录或从指定目录导入无人机飞行轨迹文件,“读任务”和“写任务”按钮用于读写无人机编队的任务指令,“二维模拟动画”按钮用于动态显示双方无人机编队的博弈对抗过程,“添加航点”和“删除航点”用于手动设置无人机的位置信息,“开始”、“暂停”、“0.5倍速”、“2倍速”和“4倍速”按钮用于设置仿真软件的运行参数信息。
另外,本发明提出的基于多无人机协同博弈对抗的控制系统的执行流程如图3所示:系统初始化后,本方无人机编队以固定队形执行巡航任务,当本方编队发现对抗方目标时,开始执行协同博弈对抗任务,然后针对双方无人机编队的状态信息进行态势评估,并基于态势评估信息决策本方编队的对抗策略,进而生成编队内单架无人机任务指令,当对抗方无人机处于当前方导弹不可逃逸区时,启动当前方导弹系统,根据导弹模型发射导弹对敌打击。
需要说明的是,上述实施例提供的基于多无人机协同博弈对抗的控制系统,仅以上述各功能模块的划分进行举例说明,在实际应用中,可以根据需要而将上述功能分配由不同的功能模块来完成,即将本发明实施例中的模块或者步骤再分解或者组合,例如,上述实施例的模块可以合并为一个模块,也可以进一步拆分成多个子模块,以完成以上描述的全部或者部分功能。对于本发明实施例中涉及的模块、步骤的名称,仅仅是为了区分各个模块或者步骤,不视为对本发明的不当限定。
本领域技术人员应该能够意识到,结合本文中所公开的实施例描述的各示例的模块、方法步骤,能够以电子硬件、计算机软件或者二者的结合来实现,软件模块、方法步骤对应的程序可以置于随机存储器(RAM)、内存、只读存储器(ROM)、电可编程ROM、电可擦除可编程ROM、寄存器、硬盘、可移动磁盘、CD-ROM、或技术领域内所公知的任意其它形式的存储介质中。为了清楚地说明电子硬件和软件的可互换性,在上述说明中已经按照功能一般性地描述了各示例的组成及步骤。这些功能究竟以电子硬件还是软件方式来执行,取决于技术方案的特定应用和设计约束条件。本领域技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能,但是这种实现不应认为超出本发明的范围。
术语“第一”、“第二”等是用于区别类似的对象,而不 是用于描述或表示特定的顺序或先后次序。
术语“包括”或者任何其它类似用语旨在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、物品或者设备/装置不仅包括那些要素,而且还包括没有明确列出的其它要素,或者还包括这些过程、方法、物品或者设备/装置所固有的要素。
至此,已经结合附图所示的优选实施方式描述了本发明的技术方案,但是,本领域技术人员容易理解的是,本发明的保护范围显然不局限于这些具体实施方式。在不偏离本发明的原理的前提下,本领域技术人员可以对相关技术特征作出等同的更改或替换,这些更改或替换之后的技术方案都将落入本发明的保护范围之内。

Claims (13)

  1. 一种基于多无人机协同博弈对抗的控制系统,其特征在于,该系统包括管理模块以及博弈双方的无人机编队模块、态势评估模块、决策模块、协同任务分配模块;
    所述管理模块,配置为存储所述无人机编队模块发送的第一信息并发送至所述态势评估模块;
    所述无人机编队模块,配置为当巡航过程中发现对抗方的无人机时,获取当前方、对抗方无人机编队中的各无人机的状态信息,作为第一信息并发送至所述管理模块;还配置为控制当前方各无人机执行所述协同任务分配模块发送的控制指令;
    所述态势评估模块,配置为基于所述第一信息,通过预设多种预设类别的评估方法获取当前方各无人机的态势评估信息;
    所述决策模块,配置为基于获取的博弈双方各无人机的数量,统计各方无人机所有的机动策略,并结合所述态势评估信息,获取当前方各无人机相对于对抗方各无人机的最优态势优势值,构建当前方无人机编队相对于对抗方无人机编队的态势矩阵;基于所述态势矩阵,通过预设的策略选取方法获取当前方的对抗策略;
    所述协同任务分配模块,配置为根据所述态势矩阵,获取当前方无人机所应对的对抗目标,并结合所述对抗策略、所述最优态势评估值,生成当前方各无人机的控制指令,发送所述无人机编队模块。
  2. 根据权利要求1所述的基于多无人机协同博弈对抗的控制系统,其特征在于,该系统还包括视景显示模块;
    所述视景显示模块,配置为获取所述第一状态信息及博弈对抗图像进行输出显示。
  3. 根据权利要求2所述的基于多无人机协同博弈对抗的控制系统,其特征在于,
    所述管理模块,还配置为进行系统设置和博弈对抗性能分析;
    所述系统设置包括仿真调度管理、博弈双方多无人机系统初始化和离散事件触发管理;其中,所述仿真调度管理,包括系统仿真时长、步长、仿真开始时间、仿真结束时间的设置管理;所述博弈双方多无人机系统初始化,配置为对博弈双方多无人机的飞行状态进行初始设置;所述离散事件触发管理,配置为进行人工离散事件设置,以中止多无人机双方的博弈进程;
    所述博弈对抗性能分析包括胜率结果分析、过程趋势分析和对抗时间分析;其中,所述胜率结果分析,配置为以图表形式显示博弈双方胜率;所述过程趋势分析,配置为以曲线形式显示博弈双方态势变化趋势;所述对抗时间分析,配置为以图表形式显示双方博弈时间。
  4. 根据权利要求3述的基于多无人机协同博弈对抗的控制系统,其特征在于,所述状态信息包括无人机的位置、速度、姿态角和控制输入信息;所述控制输入信息包括无人机的纵向过载、法向过载和滚转角。
  5. 根据权利要求3所述的基于多无人机协同博弈对抗的控制系统,其特征在于,所述态势评估信息包括机动能力评估信息和视野能力评估信息;所述视野能力评估信息包括距离评估值、方位角评估值和进入角评估值;所述机动能力评估信息包括能量评估值和空战性能评估值。
  6. 根据权利要求5所述的基于多无人机协同博弈对抗的控制系统,其特征在于,所述距离评估值其计算方法为:
    Figure PCTCN2020108774-appb-100001
    其中,T i D为距离评估值,D i、D Rmax、D Mmax、D Mmin、D MKmax、D MKmin分别表示当前方第i架无人机与对抗方任一无人机的相对距离、火控雷达的最大搜索距离、空空导弹的最大攻击距离、空空导弹的最小攻击距离、空空导弹的最大不可逃逸距离、空空导弹的最小不可逃逸距离。
  7. 根据权利要求5所述的基于多无人机协同博弈对抗的控制系统,其特征在于,所述方位角评估值其计算方法为:
    Figure PCTCN2020108774-appb-100002
    其中,Φ i、Φ Rmax、Φ Mmax、Φ MKmax分别表示当前方第i架无人机与对抗方任一无人机的方位角、火控雷达的最大搜索方位角、空空导弹的最大搜索方位角、空空导弹的最大不可逃逸角,T i Φ为方位角评估值。
  8. 根据权利要求5所述的基于多无人机协同博弈对抗的控制系统,其特征在于,所述进入角评估值其计算方法为:
    Figure PCTCN2020108774-appb-100003
    其中,p i
    Figure PCTCN2020108774-appb-100004
    分别表示当前方第i架无人机与对抗方任一无人机的进入角和进入角阈值,T i p为进入角评估值。
  9. 根据权利要求5所述的基于多无人机协同博弈对抗的控制系统,其特征在于,所述能量评估值其计算方法为:
    Figure PCTCN2020108774-appb-100005
    其中,
    Figure PCTCN2020108774-appb-100006
    表示第i架无人机的能量值,H i为第i架无人机的飞行高度,V i为第i架无人机的飞行速度,g为重力加速度系数,E T为对抗方任一无人机的能量值,T i E为第i架无人机的能量评估值。
  10. 根据权利要求5所述的基于多无人机协同博弈对抗的控制系统,其特征在于,所述空战性能评估值其计算方法为:
    Figure PCTCN2020108774-appb-100007
    其中,B i、A i、D i分别表示第i架无人机的机动性参数、火力衡量参数和探测能力衡量参数,
    Figure PCTCN2020108774-appb-100008
    Figure PCTCN2020108774-appb-100009
    分别表示第i架无人机的操纵效能系数、生存力系数、航程系数,T i c为第i架无人机的空战性能评估值。
  11. 根据权利要求3所述的基于多无人机协同博弈对抗的控制系统, 其特征在于,所述决策模块中“获取当前方各无人机相对于对抗方各无人机的最优态势优势值,构建当前方无人机编队相对于对抗方无人机编队的态势矩阵”,其方法为:
    基于当前方、对抗方各无人机的机动策略,结合所述态势评估信息,构建当前方各无人机相对于对抗方各无人机的第一态势矩阵;所述机动策略包括无人机横向机动策略和纵向机动策略,其中,横向机动策略决定无人机航迹偏角变化值,纵向机动策略决定无人机高度变化值;
    基于所述第一态势矩阵中的态势优势值,通过min-max原则,得到当前方各无人机相对于对抗方各无人机的最优态势优势值,并构建当前方无人机编队相对于对抗方无人机编队的态势矩阵。
  12. 根据权利要求3所述的基于多无人机协同博弈对抗的控制系统,其特征在于,所述决策模块中“基于所述态势矩阵,通过预设的策略选取方法获取当前方的对抗策略”,其方法为:
    依次选取所述态势矩阵中最大值,并删除所述态势矩阵中最大值对应行和列的所有元素;
    对选取的最大值进行累加,得到当前方无人机编队相对于对抗方无人机编队的整体态势优势值;
    基于所述整体态势优势值,将该值与设定的最小最大态势优势阈值进行比对,获取当前方无人机编队的对抗策略。
  13. 根据权利要求12所述的基于多无人机协同博弈对抗的控制系统,其特征在于,所述协同任务分配模块中“根据所述态势矩阵,获取当前方无人机所应对的对抗目标”,其方法为:根据所述态势矩阵依次选取的最大值对应的行列坐标,获取当前方无人机所应对的对抗方的无人机。
PCT/CN2020/108774 2020-03-03 2020-08-13 基于多无人机协同博弈对抗的控制系统 WO2021174765A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/261,036 US11669110B2 (en) 2020-03-03 2020-08-13 Control system based on multi-unmanned aerial vehicle cooperative strategic confrontation

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010138475.2 2020-03-03
CN202010138475.2A CN111221352B (zh) 2020-03-03 2020-03-03 基于多无人机协同博弈对抗的控制系统

Publications (1)

Publication Number Publication Date
WO2021174765A1 true WO2021174765A1 (zh) 2021-09-10

Family

ID=70831445

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/108774 WO2021174765A1 (zh) 2020-03-03 2020-08-13 基于多无人机协同博弈对抗的控制系统

Country Status (3)

Country Link
US (1) US11669110B2 (zh)
CN (1) CN111221352B (zh)
WO (1) WO2021174765A1 (zh)

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113805569A (zh) * 2021-09-23 2021-12-17 北京理工大学 基于多智能体技术的对抗系统、方法、终端及存储介质
CN113885353A (zh) * 2021-09-29 2022-01-04 北京大翔航空科技有限公司 一种分布式智能无人机集群对抗的硬件在回路仿真系统
CN113893539A (zh) * 2021-12-09 2022-01-07 中国电子科技集团公司第十五研究所 智能体的协同对战方法及装置
CN114167756A (zh) * 2021-12-08 2022-03-11 北京航空航天大学 多无人机协同空战决策自主学习及半实物仿真验证方法
CN114237291A (zh) * 2021-12-15 2022-03-25 西北工业大学 一种柔性约束多无人机编队控制方法
CN114609905A (zh) * 2022-03-07 2022-06-10 大连海事大学 一种船舶编队事件触发控制方法
CN114625119A (zh) * 2021-10-11 2022-06-14 中国舰船研究设计中心 一种面向任务的无人艇自主控制决策系统架构
CN115016526A (zh) * 2022-05-01 2022-09-06 西北工业大学 一种双层防御的集群自组织控制方法
CN115268481A (zh) * 2022-07-06 2022-11-01 中国航空工业集团公司沈阳飞机设计研究所 一种无人机对抗策略决策方法及其系统
CN115310257A (zh) * 2022-06-23 2022-11-08 中国人民解放军军事科学院国防科技创新研究院 基于人工势场的态势估计方法及装置
CN115468454A (zh) * 2022-08-30 2022-12-13 南京理工大学 虚拟力法与引战配合相结合的多弹协同对抗策略
CN116225049A (zh) * 2022-12-21 2023-06-06 中国航空工业集团公司沈阳飞机设计研究所 一种多无人机狼群协同作战攻防决策算法
CN117193344A (zh) * 2023-09-21 2023-12-08 大连海事大学 基于同步制导的机/船协同多元任务事件触发控制方法
CN117572893A (zh) * 2024-01-15 2024-02-20 白杨时代(北京)科技有限公司 基于强化学习的无人机集群对抗策略获取方法及相关设备
CN117633563A (zh) * 2024-01-24 2024-03-01 中国电子科技集团公司第十四研究所 一种基于optics算法的多目标自顶向下式层次化分群方法

Families Citing this family (31)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111240353B (zh) * 2020-01-07 2021-06-15 南京航空航天大学 基于遗传模糊树的无人机协同空战决策方法
CN111221352B (zh) 2020-03-03 2021-01-29 中国科学院自动化研究所 基于多无人机协同博弈对抗的控制系统
CN111707148B (zh) * 2020-06-28 2021-02-09 北京航空航天大学 一种时变时延条件下多导弹协同制导方法及系统
US11797004B2 (en) * 2020-07-31 2023-10-24 Aurora Flight Sciences Corporation, a subsidiary of The Boeing Company Causing a robot to execute a mission using a task graph and a task library
CN111814915B (zh) * 2020-08-26 2020-12-25 中国科学院自动化研究所 多智能体时空特征提取方法及系统、行为决策方法及系统
CN112612300B (zh) * 2020-11-27 2023-06-09 合肥工业大学 多平台智能决策的多目标博弈方法及装置
CN112612298B (zh) * 2020-11-27 2023-06-09 合肥工业大学 对抗环境下多无人机战术决策的多目标博弈方法及装置
CN112529110B (zh) * 2020-12-29 2023-04-07 中国科学院自动化研究所 对手策略反演方法、系统、装置
CN112742028B (zh) * 2021-01-22 2024-05-21 中国人民解放军国防科技大学 一种对抗博弈的编队决策方法、系统、介质及设备
CN112947545B (zh) * 2021-01-25 2022-10-21 中国人民解放军国防科技大学 基于条件触发机制的无人机集群任务适变架构设计方法
CN113377119B (zh) * 2021-04-21 2023-05-09 北京控制工程研究所 一种基于能力模型的航天器任务推演方法和系统
CN113867178B (zh) * 2021-10-26 2022-05-31 哈尔滨工业大学 面向多机器人对抗的虚实迁移训练系统
CN114019828A (zh) * 2021-11-29 2022-02-08 中国人民解放军国防科技大学 一种无人机集群多模态虚实交互仿真系统及方法
CN114239392B (zh) * 2021-12-09 2023-03-24 南通大学 无人机决策模型训练方法、使用方法、设备及介质
CN114167899B (zh) * 2021-12-27 2023-05-26 北京联合大学 一种无人机蜂群协同对抗决策方法及系统
CN114326818B (zh) * 2022-01-10 2023-06-02 大连海事大学 一种基于动态事件触发的无人机输出反馈协同控制系统
CN114397911B (zh) * 2022-01-18 2024-04-09 中科视拓(北京)科技有限公司 一种基于多智能体的无人机集群对抗决策方法
CN114355987A (zh) * 2022-01-19 2022-04-15 北京航空航天大学 一种基于弹性重要度的无人机编队重构方法
CN114662999B (zh) * 2022-05-24 2022-08-30 深圳联和智慧科技有限公司 一种无人机集群调动处理方法、系统及云平台
CN115016533B (zh) * 2022-05-31 2023-03-24 中国航空工业集团公司沈阳飞机设计研究所 一种无人机多机协同任务分配控制系统及其方法
CN115113642B (zh) * 2022-06-02 2023-06-20 中国航空工业集团公司沈阳飞机设计研究所 一种多无人机时空关键特征自学习协同对抗决策方法
CN114911269B (zh) * 2022-06-17 2023-05-16 电子科技大学 一种基于无人机群的组网雷达干扰策略生成方法
CN115310345B (zh) * 2022-06-23 2023-10-03 中国人民解放军军事科学院国防科技创新研究院 基于人工势场的智能算法评估方法及装置
CN114859939A (zh) * 2022-07-05 2022-08-05 中国人民解放军战略支援部队航天工程大学 一种基于智能移动的节点机器人路径规划方法
CN115357842B (zh) * 2022-10-18 2023-01-17 北京理工大学 基于自适应评价函数的机动突防策略优化方法
CN115755955B (zh) * 2022-11-02 2023-06-02 哈尔滨工业大学 基于截获概率的空空导弹协同探测方法
CN115993842B (zh) * 2023-03-23 2023-05-30 西北工业大学深圳研究院 一种用于群体智能系统的编队协同控制方法
CN116663239B (zh) * 2023-04-24 2024-01-05 四川大学 一种基于黄金分割法的导弹不可逃逸区攻击距离计算方法
CN116680542B (zh) * 2023-06-27 2024-01-09 北京五木恒润科技有限公司 一种主干分支态势与旁路分支态势生成方法及系统
CN116893690A (zh) * 2023-07-25 2023-10-17 西安爱生技术集团有限公司 一种基于强化学习的无人机躲避攻击输入数据计算方法
CN117075496A (zh) * 2023-09-18 2023-11-17 广东昊一航空科技有限公司 一种无人机方阵运行仿真系统

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105204512A (zh) * 2015-09-14 2015-12-30 北京航空航天大学 一种基于简化模型机器博弈的六自由度无人作战飞机近距格斗方法
CN105427032A (zh) * 2015-11-09 2016-03-23 河海大学 一种无人机对抗决策评估方法
CN107463094A (zh) * 2017-07-13 2017-12-12 江西洪都航空工业集团有限责任公司 一种不确定信息下多无人机空战动态博弈方法
KR20180051135A (ko) * 2016-11-08 2018-05-16 주식회사 블루젠드론 무인 비행기용 마일즈 시스템
CN110058608A (zh) * 2019-04-08 2019-07-26 合肥工业大学 多无人机协同对抗的控制方法、系统及存储介质
CN111221352A (zh) * 2020-03-03 2020-06-02 中国科学院自动化研究所 基于多无人机协同博弈对抗的控制系统

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9812020B2 (en) * 2015-08-13 2017-11-07 Hon Hai Precision Industry Co., Ltd. Electronic device and unmanned aerial vehicle control method
US20170131726A1 (en) * 2015-11-06 2017-05-11 SySense Incorporated Automated operation of aircraft systems in inverted-v formations
CN105700555B (zh) * 2016-03-14 2018-04-27 北京航空航天大学 一种基于势博弈的多无人机协同搜索方法
US11069254B2 (en) * 2017-04-05 2021-07-20 The Boeing Company Method for simulating live aircraft infrared seeker obscuration during live, virtual, constructive (LVC) exercises
CN108153332B (zh) * 2018-01-09 2020-05-19 中国科学院自动化研究所 基于大包线博弈策略的轨迹模拟系统
CN111712773A (zh) * 2019-07-09 2020-09-25 深圳市大疆创新科技有限公司 一种无人机协同作业的控制方法、电子设备及系统
US11482128B2 (en) * 2019-08-21 2022-10-25 The Boeing Company Simulating aircraft data for tactics training

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105204512A (zh) * 2015-09-14 2015-12-30 北京航空航天大学 一种基于简化模型机器博弈的六自由度无人作战飞机近距格斗方法
CN105427032A (zh) * 2015-11-09 2016-03-23 河海大学 一种无人机对抗决策评估方法
KR20180051135A (ko) * 2016-11-08 2018-05-16 주식회사 블루젠드론 무인 비행기용 마일즈 시스템
CN107463094A (zh) * 2017-07-13 2017-12-12 江西洪都航空工业集团有限责任公司 一种不确定信息下多无人机空战动态博弈方法
CN110058608A (zh) * 2019-04-08 2019-07-26 合肥工业大学 多无人机协同对抗的控制方法、系统及存储介质
CN111221352A (zh) * 2020-03-03 2020-06-02 中国科学院自动化研究所 基于多无人机协同博弈对抗的控制系统

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
KANG YANGMING; LIU ZHEN; PU ZHIQIANG; YI JIANQIANG; ZU WEI: "Beyond-Visual-Range Tactical Game Strategy for Multiple UAVs", 2019 CHINESE AUTOMATION CONGRESS (CAC), 22 November 2019 (2019-11-22), pages 5231 - 5236, XP033711542, DOI: 10.1109/CAC48633.2019.8996232 *
LUO DE-LIN, ZHANG, HAI-YANG; XIE, RONG-ZENG; WU, SHUN-XIANG: "Unmanned aerial vehicles swarm conflict based on multi-agent system", KONGZHI LILUN YU YINGYONG - CONTROL THEORY & APPLICATIONS, vol. 32, no. 11, 1 November 2015 (2015-11-01), pages 1498 - 1504, XP055843174, ISSN: 1000-8152, DOI: 10.7641/CTA.2015.50366 *

Cited By (27)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113805569A (zh) * 2021-09-23 2021-12-17 北京理工大学 基于多智能体技术的对抗系统、方法、终端及存储介质
CN113805569B (zh) * 2021-09-23 2024-03-26 北京理工大学 基于多智能体技术的对抗系统、方法、终端及存储介质
CN113885353A (zh) * 2021-09-29 2022-01-04 北京大翔航空科技有限公司 一种分布式智能无人机集群对抗的硬件在回路仿真系统
CN113885353B (zh) * 2021-09-29 2024-01-30 北京大翔航空科技有限公司 一种分布式智能无人机集群对抗的硬件在回路仿真系统
CN114625119A (zh) * 2021-10-11 2022-06-14 中国舰船研究设计中心 一种面向任务的无人艇自主控制决策系统架构
CN114625119B (zh) * 2021-10-11 2024-03-29 中国舰船研究设计中心 一种面向任务的无人艇自主控制决策系统架构
CN114167756B (zh) * 2021-12-08 2023-06-02 北京航空航天大学 多无人机协同空战决策自主学习及半实物仿真验证方法
CN114167756A (zh) * 2021-12-08 2022-03-11 北京航空航天大学 多无人机协同空战决策自主学习及半实物仿真验证方法
CN113893539A (zh) * 2021-12-09 2022-01-07 中国电子科技集团公司第十五研究所 智能体的协同对战方法及装置
CN113893539B (zh) * 2021-12-09 2022-03-25 中国电子科技集团公司第十五研究所 智能体的协同对战方法及装置
CN114237291A (zh) * 2021-12-15 2022-03-25 西北工业大学 一种柔性约束多无人机编队控制方法
CN114237291B (zh) * 2021-12-15 2023-06-30 西北工业大学 一种柔性约束多无人机编队控制方法
CN114609905A (zh) * 2022-03-07 2022-06-10 大连海事大学 一种船舶编队事件触发控制方法
CN114609905B (zh) * 2022-03-07 2024-04-05 大连海事大学 一种船舶编队事件触发控制方法
CN115016526B (zh) * 2022-05-01 2024-04-30 西北工业大学 一种双层防御的集群自组织控制方法
CN115016526A (zh) * 2022-05-01 2022-09-06 西北工业大学 一种双层防御的集群自组织控制方法
CN115310257A (zh) * 2022-06-23 2022-11-08 中国人民解放军军事科学院国防科技创新研究院 基于人工势场的态势估计方法及装置
CN115268481A (zh) * 2022-07-06 2022-11-01 中国航空工业集团公司沈阳飞机设计研究所 一种无人机对抗策略决策方法及其系统
CN115268481B (zh) * 2022-07-06 2023-06-20 中国航空工业集团公司沈阳飞机设计研究所 一种无人机对抗策略决策方法及其系统
CN115468454A (zh) * 2022-08-30 2022-12-13 南京理工大学 虚拟力法与引战配合相结合的多弹协同对抗策略
CN116225049A (zh) * 2022-12-21 2023-06-06 中国航空工业集团公司沈阳飞机设计研究所 一种多无人机狼群协同作战攻防决策算法
CN117193344A (zh) * 2023-09-21 2023-12-08 大连海事大学 基于同步制导的机/船协同多元任务事件触发控制方法
CN117193344B (zh) * 2023-09-21 2024-03-15 大连海事大学 基于同步制导的机/船协同多元任务事件触发控制方法
CN117572893B (zh) * 2024-01-15 2024-03-19 白杨时代(北京)科技有限公司 基于强化学习的无人机集群对抗策略获取方法及相关设备
CN117572893A (zh) * 2024-01-15 2024-02-20 白杨时代(北京)科技有限公司 基于强化学习的无人机集群对抗策略获取方法及相关设备
CN117633563A (zh) * 2024-01-24 2024-03-01 中国电子科技集团公司第十四研究所 一种基于optics算法的多目标自顶向下式层次化分群方法
CN117633563B (zh) * 2024-01-24 2024-05-10 中国电子科技集团公司第十四研究所 一种基于optics算法的多目标自顶向下式层次化分群方法

Also Published As

Publication number Publication date
US11669110B2 (en) 2023-06-06
US20210405660A1 (en) 2021-12-30
CN111221352A (zh) 2020-06-02
CN111221352B (zh) 2021-01-29

Similar Documents

Publication Publication Date Title
WO2021174765A1 (zh) 基于多无人机协同博弈对抗的控制系统
CN112947581B (zh) 基于多智能体强化学习的多无人机协同空战机动决策方法
CN113791634A (zh) 一种基于多智能体强化学习的多机空战决策方法
Burgin et al. Rule-based air combat simulation
CN112198892B (zh) 一种多无人机智能协同突防对抗方法
CN114063644B (zh) 基于鸽群反向对抗学习的无人作战飞机空战自主决策方法
Zhang et al. Maneuver decision-making of deep learning for UCAV thorough azimuth angles
Fu et al. The overview for UAV air-combat decision method
Li et al. Autonomous maneuver decision-making for a UCAV in short-range aerial combat based on an MS-DDQN algorithm
CN114492805A (zh) 一种基于模糊推理的空战机动决策设计方法
CN115951709A (zh) 基于td3的多无人机空战策略生成方法
CN113671825A (zh) 一种基于强化学习的机动智能决策规避导弹方法
Qiu et al. One-to-one air-combat maneuver strategy based on improved TD3 algorithm
Duan et al. Autonomous maneuver decision for unmanned aerial vehicle via improved pigeon-inspired optimization
CN115061495A (zh) 一种基于鹰鸽博弈的无人机群组对抗自主控制方法
Wang et al. Research on naval air defense intelligent operations on deep reinforcement learning
CN116679742B (zh) 一种多六自由度飞行器协同作战决策方法
Meng et al. UAV Attack and Defense Optimization Guidance Method Based on Target Trajectory Prediction
Liu et al. Research on Individual Performance Index of Air Cluster Combat Aircraft Based on Differential Game Theory
Hu et al. An Intelligent MAV-UAV Cooperative Combat Planning Method Based on Deep Reinforcement Learning
CN116009592A (zh) 一种基于个体相似度鸽群优化的无人飞行器自主突防方法
CN116068889A (zh) 一种巡飞弹的饱和攻击方法、装置及存储介质
Lei et al. Air Combat Intelligent Assistant Decision-Making Sample Data Cultivation Method
Li et al. An Auction-based Attack-defense Decision-making Method for UAV Air Combat
Lei et al. Autonomous maneuver decision for unmanned aerial vehicle via improved pigeon-inspired optimization

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20922808

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20922808

Country of ref document: EP

Kind code of ref document: A1