• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    The prediction device (1) comprises a set of data input (2) for entering into said device (1), both data from the attacker relating to at least attack models, and data of the defender relating to at least one area to be defended and available defenses, a modeling unit (4) to generate a game tree evaluated from data entered, as part of the game theory, a unit of resolution (6) to define a game balance within the framework of game theory, an equilibrium defining a pair of attacking strategy and defender strategy, an interpretation unit (7) to determine, from the equilibrium of game, an optimal attack solution, as well as an optimum defense solution best suited to this optimal attack solution, and an information transmission unit (9) for transmitting this information.
    公开号:FR3063554A1
    申请号:FR1700216
    申请日:2017-03-03
    公开日:2018-09-07
    发明作者:Ronan Fruit；David VIGOUROUX；Stephane Le Menec；Charlotte Touchard；Alexandre Kotenfoff；Mathias Formoso
    申请人:MBDA France SAS；
    IPC主号:
    专利说明:

    ® FRENCH REPUBLIC
    NATIONAL INSTITUTE OF INDUSTRIAL PROPERTY © Publication number: 3,063,554 (to be used only for reproduction orders)
    ©) National registration number: 17 00216
    COURBEVOIE © Int Cl 8 : G 06 F15 / 02 (2017.01), G 06 F 19/28
    A1 PATENT APPLICATION
    (© Date of filing: 03.03.17. © Applicant (s): MBDA FRANCE— FR. ©) Priority: @ Inventor (s): FRUIT RONAN, VIGOUROUX DAVID, LE MENEC STEPHANE, TOUCHARD CHARLOTTE, (43) Date of public availability of the KOTENFOFF ALEXANDRE and FORMOSO Mathias. request: 07.09.18 Bulletin 18/36. ©) List of documents cited in the report preliminary research: Refer to end of present booklet (© References to other national documents ® Holder (s): MBDA FRANCE. related: ©) Extension request (s): ©) Agent (s): GEVERS & ORES Société anonyme.
    METHOD AND DEVICE FOR PREDICTING OPTIMAL ATTACK AND DEFENSE SOLUTIONS IN A SCENARIO OF MILITARY CONFLICT.
    FR 3 063 554 - A1
    - The prediction device (1) comprises a data entry set (2) for entering said device (1), both attacker data relating at least to attack models, and data of the defender relating to at least one area to be defended and to the means of defense available, a modeling unit (4) for generating a game tree evaluated from input data, within the framework of game theory, a unit of resolution (6) to define a game balance within the framework of game theory, a balance defining a couple of attacking strategy and defensive strategy, an interpretation unit (7) to determine, from the balance of game, an optimal attack solution, as well as an optimal defense solution best suited to this optimal attack solution, and an information transmission unit (9) for transmitting this information.

    The present invention relates to a method and a device for predicting optimal attack and defense solutions in a military conflict scenario.
    The present invention applies to the military field, and more particularly although not exclusively in situations involving command and coordination systems (“Command & Control (C2)” in English) which manage, in particular, an allocation of weapons to threats. In the military field, the defense of a given area against an enemy offensive aims to allocate w the best available resources to deal with threats by maximizing the survival of the strategic points defended. Depending on the threats, the C2 system determines an engagement solution taking into account operational and technical criteria, in particular within the framework of an anti-aircraft weapons system.
    Anti-aircraft weapon systems are known to consist of sensors, launchers, and a type C2 operational command center, in which operators can interact with and supervise the system. The operational command center predicts the trajectories of the attackers, and uses this prediction to plan the defense of the attackers' potential objectives. It seeks to find the best possible defense for the system against an attack scenario.
    Predicting the trajectories of attackers is made particularly complex, due to three factors;
    - high uncertainty affects the attacker's subsequent behavior, and the further we project into the future, the more this actual behavior is unknown;
    - most attackers (fighter planes, missiles, ...) have significant maneuvering capabilities and can therefore change course quickly (change of course, altitude, ...);
    - the attack trajectories, and in particular the terminal maneuvers to rally the objectives, are chosen to defeat the defense system, and are therefore difficult to predict by the operational command center.
    The information available to predict the trajectories comes on the one hand from sensors, in particular radars, which provide the past trajectory of the attackers with uncertainties on the measurements, and on the other hand from the knowledge that the system and the operator have of type of attacker and attack missions that may be encountered.
    The present invention seeks to predict a defense solution implemented by the defender to best protect themselves from the attack and / or, by placing themselves on the attacker's side, an optimal attack solution implemented by the 'attacker.
    The present invention relates more particularly to a method of predicting an optimal attack solution and a corresponding optimal defense solution, in a scenario of military conflict (at least potential) between an attacker and a defender.
    According to the invention, said prediction method comprises:
    a data entry step, implemented by a data entry set, consisting in entering, at least initially before the implementation of the method, both attacker's data relating at least to attack models, and defender data relating to at least one area to be defended and available defenses;
    a scenario modeling step, implemented by a modeling unit, consisting in generating a game tree evaluated from said data entered in the framework of game theory;
    a resolution step, implemented by a resolution unit, consisting in defining a game balance within the framework of game theory, a balance defining a couple of attacking strategy and defending strategy;
    - an interpretation step, implemented by an interpretation unit, consisting in determining, from the game balance, an optimal attack solution and an optimal defense solution, best suited to this solution 'optimal attack; and
    an information transmission step, implemented by an information transmission unit, consisting in transmitting to an operator or a user system, at least said optimal attack solution and said optimal defense solution.
    The method according to the invention thus makes it possible to predict optimal solutions making it possible in particular to define, if one takes the side of the defender, an optimal defense solution, and if one takes the side of the attacker, a solution of optimal attack.
    The prediction of these optimal (attack and defense) solutions in the conflict scenario is based on solving a game theory problem, as detailed below. Game theory makes it possible to analyze situations in which the optimal action for an agent depends on the anticipations it forms on the decision of another agent. These two agents are the attacker and the defender in the context of the present invention.
    In game theory, a game balance, notably a so-called Nash balance, is a situation in a game where no player has an interest in changing strategy. In this situation, the set of choices made by several players, knowing their reciprocal strategies, has become stable due to the fact that no one can modify his strategy alone without weakening his personal position.
    Advantageously, the attacker's data includes at least some of the following data:
    - models of attack procedures, for example attack trajectories;
    - types of threats and associated characteristics;
    - at least one detected position from the appearance of a threat;
    - at least one position assumed to appear as a threat; and
    - attacker's preferences.
    In addition, advantageously, the defender's data includes at least some of the following data:
    - possible missions;
    - data from detection means;
    - characteristics of detection means;
    - positions of potential targets in the area to be defended;
    - defense capabilities of the area to be defended; and
    - Defender preferences.
    Regarding the attacker's data and the defender's data, they may be known and certain data, or else supposed data, which may not be completely accurate, in particular depending on who enters the attacker or the defender. prediction. So if you use the prediction method to plan an attack, you know exactly as the attacker the attacker's data and parameters, but the defender's data available is uncertain. Conversely, if one uses the prediction process to plan a defense, one knows exactly as defender the data and the parameters of the defender. However, the attacker's data is not or hardly accessible. These data are indeed hypotheses, expert opinions, elements from military intelligence, etc. The data therefore include presumed beliefs of the opposing party, presumed by the user of the invention.
    In one embodiment, in particular in the case of the protection of a site during an air attack, the step of entering data also consists in entering, during the implementation of the method, current data at least from the attacker, detected for example for one or more radars.
    Furthermore, advantageously, the modeling step comprises:
    - a first sub-step consisting in generating a set of possible strategies using some of the data entered;
    - a second sub-step consisting in generating a game tree within the framework of game theory, from said set of strategies;
    - a third sub-step consisting in evaluating said game tree, using data from some of the data entered.
    Preferably, the third sub-step consists, from pairs of strategies, of evaluating at least one pair of attacking and defending strategies, and assigning a value to the attacker and a value to the defender.
    In addition, advantageously, the resolution step includes:
    - a first sub-step consisting in pruning the received game tree, from data relating to threats, so as to form a reduced game tree; and
    - a second sub-step consisting in determining the game balance, from this reduced game tree, to deduce a couple of attacking and defending strategies.
    Preferably, the first sub-step also uses detected current data, entered during the implementation of the method, to prune the received game tree.
    Furthermore, advantageously, the interpretation step includes:
    - a first substep consisting in interpreting the optimal attack solution; and
    - a second sub-step consisting in interpreting the optimal defense solution best suited to this optimal attack solution making it possible to define rules of engagement for the defender.
    In addition, advantageously, the interpretation step also includes:
    - a sub-step to assess the dangerousness of the threat; and or
    - a sub-step to assess the probability of success of the optimal attack solution and the probability of success of the optimal defense solution.
    The process according to the invention can be used in various applications in the military field.
    In a preferred application of the prediction method, the defender's data relates to a ground site to be protected provided with defense capabilities, and the defender's data relates to an aerial attack on said ground site to be protected and comprising current data detected.
    Advantageously, the method can be applied to at least one of the following situations, relating to a scenario of military conflict:
    - anti-aircraft defense;
    - air, defensive or offensive combat;
    - mission planning for striking targets.
    The present invention also relates to a device for predicting an optimal attack solution and a corresponding optimal defense solution, in a scenario of at least potential military conflict between an attacker and a defender.
    According to the invention, said device comprises:
    a set of data entry configured to enter, at least initially before the implementation of the method, both data of the attacker relating to at least attack models, and data of the defender relating to at least to an area to be defended and to the means of defense available;
    - a modeling unit configured to generate a game tree evaluated from said data entered within the framework of game theory;
    - a resolution unit configured to define a game balance within the framework of game theory, a balance defining a pair of attacking and defending strategies;
    - an interpretation unit configured to determine, from the game balance, an optimal attack solution and an optimal defense solution, best suited to this optimal attack solution; and
    - an information transmission unit configured to transmit to an operator or a user system, at least said optimal attack solution and said optimal defense solution.
    In a particular embodiment, the data entry set includes at least some of the following:
    - an input element allowing an operator to enter data to be entered;
    - a data loading element associated with a memory and configured to load data into said memory.
    In addition, advantageously, the device comprises at least one detector, for example a radar, capable of detecting current data relating to a means of attack of an attacker, and the data entry assembly comprises a transmission link. data allowing automatic entry of current data detected by the detector.
    A preferred application of the present invention is to predict the defense tactics of a site comprising areas to be protected such as buildings and defenses forming part of an anti-aircraft weapon system and comprising, for example anti-missiles -missiles or other means of air defense, during an aerial attack on the site by an attacker.
    The figures of the appended drawing will make it clear how the invention can be implemented. In these figures, identical references designate similar elements.
    Figure 1 is the block diagram of a prediction device according to the invention.
    Figures 2 and 3 are block diagrams of data processing units of the prediction device of Figure 1.
    FIG. 4 schematically shows the steps of a method implemented by the prediction device of FIG. 1.
    Figure 5 illustrates a simplified attack scenario.
    Figure 6 illustrates a game tree associated with the simplified attack scenario of Figure 5.
    The device 1 making it possible to illustrate the invention and represented diagrammatically in FIG. 1 is intended to predict at least one optimal attack solution and a corresponding optimal defense solution, in a scenario of military conflict (at least potential, namely either effective or envisaged) between at least one attacker and at least one defender.
    According to the invention, said device 1 comprises:
    a set of data entry 2 configured to enter said device 1, as specified below, both data of the attacker relating at least to attack models (for example weapon trajectories of 'attack), and data from the defender relating to at least one area to be defended and to the means of defense available in that area;
    - a central unit 3 comprising:
    • a modeling unit 4 configured to generate an evaluated game tree, using at least part of the data entered by means of the data input set 2, said game tree being defined in the context game theory, as detailed below;
    A resolution unit 6 connected via a link 5 to the modeling unit 4 and configured to define a game balance within the framework of game theory, a balance defining a couple of attacking strategy and defender strategy;
    An interpretation unit 7 connected via a link 8 to the resolution unit 6 and configured to determine, from this balance of play, an optimal attack solution and an optimal defense solution ( that is, the one best suited to this optimal attack solution); and
    - an information transmission unit 9 configured to transmit at least said optimal attack solution and said optimal defense solution determined by the central unit 3, to an operator (via, for example, display means 19 which are connected by a link 10 to the central unit 3 and which display this information on a screen) or to a user system (via for example a link 11).
    In a particular embodiment, the data entry set 2 comprises:
    - an input element 12 allowing an operator to enter data, in particular manually, into the central unit 3 via a link 13. This input element 12 can comprise a keyboard, a mouse, a touchpad, etc. , or any other usual means, associated for example with a screen, which allows an operator to enter data into said device 1. This input element 12 can form with the display means 19 a man / machine interface; and
    a data loading element 14 associated with a memory 15 and configured to load data into said memory 15, via a link
    16. This memory 15 can be integrated into the central unit 3, as in the example shown in FIG. 1, or be external to the central unit 3 and linked to the latter.
    In addition, the device 1 also comprises a set 17 of detectors D1 to DN, for example radars, N being an integer greater than 1. These detectors D1 to DN are capable of detecting current data relating to an attack means ( for example a missile) of an attacker. In this case, the data input assembly 2 comprises a data transmission link 18 making it possible to automatically enter into the central unit 3 current data detected (or measured or determined) by at least one detector D1 to DN of set 17.
    The assembly 17 includes at least one detector, such as a radar for example, which monitors the environment of the area to be protected and which is capable of detecting threats and transmitting corresponding information, in particular the position and the kinematics of the threat.
    The attacker's data, entered using data entry set 2, includes at least some of the following data:
    - attack models including in particular the most probable attack procedures. An attack procedure can correspond to a sequence of concrete actions to carry out an attack. A particular attack procedure may include an attack trajectory;
    - types of threats and associated characteristics (classification, weapons, ...);
    - at least one detected position from the appearance of a threat;
    - at least one position assumed to appear as a threat; and
    - the attacker's preferences, such as the cost of using each potential threat (ammunition) or the cost in the event of interception by the defender.
    In addition, defender data, also entered using data entry set 2, includes at least some of the following data:
    - possible missions (an attack on targets and which targets (building, structure, ...), or a reconnaissance mission for example);
    - data from detectors D1 to DN or other detection means, and in particular radars;
    - characteristics of the detection means, for example the frequency used by a radar;
    - positions of potential targets in the area to be defended (buildings, radars, launchers, ammunition, ...);
    - defense capabilities of the area to be defended (characteristics of launchers and available ammunition); and
    - the defender's preferences, such as the cost of firing each munition or the prioritization of the elements of the area to be defended in order of importance.
    The data entry set 2 is configured to enter most of the data before the prediction is implemented by the central unit 3. However, in a particular embodiment, in particular in the case of protection of a site during an air attack, the data entry set 2 can also enter current data during the implementation of the prediction method. It can in particular be current data relating to the attacker, detected by at least one detector D1 to DN of the assembly 17, for example a radar, such as the current trajectory followed by a missile, a drone or an aircraft. of hunting.
    Furthermore, the modeling unit 4 comprises, as shown in FIG. 2:
    - a sub-unit 20 configured to generate a set of possible strategies, using some of the data entered and received via a link
    21. In particular, this may include at least some of the following data:
    • data from detection means;
    • the characteristics of the detection means;
    • the positions of appearance of threats;
    • potential missions and their characteristics;
    • the procedural models, in particular trajectories, to be used; and • types of threats and their characteristics;
    a sub-unit 22 linked to the sub-unit 21 via a link 23 and configured to generate a game tree within the framework of game theory, from said set of strategies generated by the sub- unit 20, and other data such as the potential missions and the characteristics of the detection means; and
    a sub-unit 24 linked to the sub-unit 22 via a link 25 and configured to evaluate said game tree, using some of the data entered, such as the characteristics of launchers and ammunition .
    The sub-unit 24 comprises, as shown in FIG. 2:
    - An element 26 configured to, from pairs of strategies, evaluate at least one pair of attacking strategy and defending strategy;
    - an element 27 linked by a link 28 to element 26 and containing a utility function for assigning a value to the attacker; and
    - an element 29 linked by a link 30 to element 26 and containing a utility function for assigning a value to the defender.
    In addition, the resolution unit 6 comprises, as shown in FIG. 1:
    a sub-unit 31 configured to prune the game tree received from the modeling unit 4, from additional data such as threats, as well as current data, entered by an operator or received in real time from the 'set 17 of detectors and specifying for example the current threat positions. By this pruning (or reduction), the subunit 31 forms a reduced play tree. The sub-unit 31 can therefore also use detected current data, entered during the implementation of the prediction method, to prune the received game tree; and
    a sub-unit 32 linked to the sub-unit 31 via a link 33 and configured to determine the game balance, from this reduced game tree received from the sub-unit 31, in order to d '' deduce a couple of attacking strategy and defending strategy. The game balance can be the Nash balance or some other common balance.
    Furthermore, the interpretation unit 7 comprises, as shown in FIG. 3:
    a subunit 34 configured to generate the optimal attack solution, from the game balance received from the resolution unit 6. This optimal attack solution defines probabilized missions, and types of threats and their probable characteristics; and
    - a subunit 35 linked to the subunit 34 via a link 36 and configured to assess the dangerousness of the threat.
    In addition, the interpretation unit 7 also includes:
    a sub-unit 37 configured to generate, from the game balance received from the resolution unit 6, the optimal defense solution best suited to said optimal attack solution, making it possible to define rules of advocate engagement; and
    - a sub-unit 38 for evaluating the probability of success of said optimal attack solution and the probability of success of said optimal defense solution.
    The prediction of these optimal solutions (attack and defense) in the conflict scenario is therefore based on the resolution of a game theory problem.
    Game theory can be defined as the theoretical framework modeling the situations in which the optimal (i.e. preferred) action for one agent depends on the expectations it forms on the decision of the other agent. These two agents are the attacker and the defender in the context of the present invention.
    In game theory, a game balance such as a Nash balance is a situation in a game where no player has an interest in changing strategy. In this situation the set of choices made by several players, knowing their reciprocal strategies, has become stable due to the fact that no one can modify his strategy alone without weakening his personal position
    Game theory is a theory of interacting decision. She studies situations where individuals make decisions, each being aware that the outcome of their own choice depends on that of others.
    A preferred application of the device 1, as described above, aims to define the best possible defense strategy for a site to be protected, that is to say the optimal defense solution, during an attack aerial of this site and in particular after the detection of the launch of a potential threat, such as the detection by a radar of a missile launch likely to reach the site or of an approach of a drone, an airplane fighter or bomber.
    Confronted with such an attack, the command center of the site to be protected is therefore faced with a technical problem, namely which one or which of its means of defense, such as an anti-missile missile, to use against the missile which attacks them. The aim of this technical problem is to defend at least the potential targets of the site, according to its means of defense and what is known about the attack: isolated attack and directed towards which target, other possible attack, simple mission reconnaissance, for example in the event of detection of an enemy aircraft. This prediction must also take into account any costs: cost of the ammunition used, cost of potential destruction, etc. The device 1 makes it possible to receive the data to be processed, both before the implementation of the prediction method and during this implementation. artwork.
    To determine the optimal defense solution, an operator therefore first enters into the device 1 (using the data entry set 2) all of the known data relating to the attacker or potential attackers and the relative data. to the defender (that is to say relating in particular to the characteristics of the site to be protected and to the weapons systems present and capable of being used to protect the site). Current data regarding real-time detections, such as a new missile launch, or an updated position of the previously detected missile, can also be entered (using data entry set 2) as soon as they are known. The prediction device 1 determines in this preferred application an optimal defense solution consisting in defining a commitment proposal.
    An engagement proposal may include an optimal allocation plan which specifies an allocation of weapons from the site to be defended and provides times or dates for firing these weapons to destroy threats, in order to best allocate available resources to deal with threats by maximizing the survival of the strategic points defended.
    The device 1, as described above, is intended to implement a method of predicting an optimal attack solution and a corresponding optimal defense solution, in a scenario of at least potential military conflict between a attacker and defender.
    As shown in FIG. 4, this prediction method comprises:
    a step E1 of data entry, implemented by the data entry set 2, consisting in entering the central unit 3 of the device 1, at least initially before the implementation of the method and possibly at the during the process, both attacker data relating at least to attack models, and defender data relating at least to an area to be defended and to the means of defense available;
    - a scenario modeling step E2, implemented by the modeling unit 4, consisting in generating a game tree evaluated from said data entered in the framework of game theory;
    - a resolution step E3, implemented by the resolution unit 6, consisting in defining a game balance within the framework of game theory, for example using entered current data;
    an interpretation step E4, implemented by the interpretation unit 7, consisting in determining, from the game balance defined in step E3, at least one optimal attack solution, as well as '' an optimal defense solution, best suited to this optimal attack solution; and
    a step E5 of information transmission, implemented by the information transmission unit 9, consisting in transmitting to an operator and / or a user system, at least said optimal attack solution and said solution optimal defense.
    The method according to the invention thus makes it possible to predict an optimal defense solution if one takes the side of the defender, and an optimal attack solution if one takes the side of the attacker.
    In the context of the present invention, the decision-makers (attacker and defender) are assumed to be VNM-rational (for "Von-Neumann and Morgenstern"). We thus place ourselves within the framework of the theory of expected utility where we know how to express the preferences of a decision maker in the presence of random events. We know how to define a function (called utility) which, with an alternative for the decision maker, associates a real number and gives in this way levels of preference between the different alternatives. The goals of the attacker and the defender are therefore to maximize their respective utility functions.
    The problem is then modeled as a sequential game (the decisions are linked sequentially) to two players, the attacker (the threats are represented as a single player which corresponds to the decision / planning level of the attack tactic used ) and the defender (a C2 system for example). The game is modeled as:
    - Bayesian: players do not know all the parameters of the game they are playing. In particular, the defender does not know the missions of the attackers, and the attackers do not necessarily know the configuration of the defense system;
    - with imperfect information: the defender does not have information about the decisions made by the attacker during mission preparation;
    - perfect memory: players have a history of their past decisions; and
    - unlimited rationality: players are not limited in terms of calculation complexity. The attacker's decisions are not influenced by information received during the attack, i.e. the attack is considered to be planned in advance during mission preparation and no longer changes.
    We are looking for a game solution, that is to say a couple of probabilistic strategies for each player. The game is solved in particular by calculating a “semi-proper quasi-perfect equilibrium” type equilibrium which is a refinement of the Nash equilibrium suitable for modeling.
    Procedures, for example trajectories, possible attacks, as well as missions and classifications of plausible threats can be extracted from the attacker's strategy. The game solution found also makes it possible to obtain one or more potential defense strategies for the C2 system, i.e. one or more shooting solutions adapted to the attack scenario.
    By way of illustration, there is shown in FIG. 5, a possible and very simplified graphic representation of a game as it can be taken into account by the device 1, comprising the following hypotheses.
    It is assumed that the attacker has the choice between different trajectories. These trajectories are discretized. Many other potential characteristics of the attack can also be represented: the choice of mission (target building, reconnaissance mission, ...) or the classification of threats (drone, fighter plane, ...) are some of them. non-exhaustive examples. The defender's actions are also discretized. At regular time intervals, the defender receives measurements from the detectors, for example from the assembly 17 of the device 1. He then has the possibility of engaging an attacker by initiating a shot before receiving a new measurement or he has the opportunity to wait.
    In the example of FIG. 5, it is considered that there is a single track (that is to say a single attacker), a single launcher (a single decision to fire possible, but on several possible dates) and a single objective (or target) 39.
    As for trajectories T1, T2, T3 and T4, they are considered as possible for the attacker to reach objective 39. In FIG. 5, we have represented:
    - a position PO at a time / 0, not allowing to discriminate the trajectories T1 to T4;
    - two positions P1A and P1B at a time tO + At making it possible to discriminate on the one hand the pair of paths T1 and T2 and on the other hand the pair of paths T3 and T4; and
    - four positions P1A, P1B, P2C and P2D at a time ίΟ + 2Δζ when approaching the objective (or target) 39, forming part of the trajectories T1, T2, T3 and T4 respectively.
    The game associated with this scenario in FIG. 5 is represented in the form of a game tree 40 in FIG. 6, to which a time scale 41 has been associated.
    On this game tree 40, the large circles 42 represent the decision nodes of the attacker and the defender. The small circles 43 represent the terminal nodes. The attacker has the choice between four trajectories T1, T2, T3 and T4. The defender does not have the information of these trajectories but receives the measurements from the detectors, which allows him to conjecture these trajectories. It makes the decision whether or not to fire each measurement reception (s) from a detector.
    The horizontal dotted lines 44 indicate that the linked decision nodes cannot be distinguished by the defender. At tO, he does not know what the attacker's trajectory is. So, from his point of view, the decision is the same for the four nodes. He has no information to distinguish them.
    At tO + At, after receiving a measurement, the defender is able to make an initial selection on the possible trajectories (between T1 and T2 on the one hand, and T3 and T4 on the other hand).
    At the terminal nodes, the players have made all their decisions, and we can therefore calculate the utility functions of the attacker and the defender.
    The method implemented by the device 1 takes into account many elements to predict the optimal attack solution of the attacker, and to deduce therefrom the most suitable optimal defense solution. Beyond the trajectories of threats, it also assesses their characteristics.
    Applying game theory to the defense problem of a C2 system takes into account all the information available to defense. In particular, the knowledge of the operator is integrated into the procedures, in particular the trajectories, studied to model the attack and in the probabilities associated with these procedures or trajectories upstream of the resolution of the game (belief a priori). Prediction is thus refined by considering the operator's expertise on the mission and the classification of the threat, since the prediction process makes it possible to distinguish the potential targets of the attack according to their importance. It also gives the operator the possibility of entering the probable characteristics of the threat considered. It also allows the operator to enter information of which he may otherwise be aware and which allows him to favor certain hypotheses.
    The integration of this knowledge into the algorithms used by the central unit 3 allows very good interaction of the operator with the device 1.
    The prediction process makes it possible to represent attack strategies in order to better exploit them. It can manage attack scenarios of an area by several simultaneous threats, instead of considering each threat individually. Assuming that a single command center ordered the attack, the threat procedures (trajectories, for example) are not independent, and were chosen voluntarily. The attacks of these threats are coordinated with each other. This approach predicts the procedures (trajectories for example) of these threats simultaneously, taking into account this dependence.
    The prediction process also exploits the notion of position of appearance (i.e. the position at which each threat is detected the first time). During an attack, the position of appearance of a threat is not trivial. The central unit 3 interprets the fact that the attacker chose a position rather than another, which makes the prediction more realistic, since this choice gives information on the attacker's objective.
    The prediction process not only makes it possible to predict the trajectory of a threat, but it also associates probabilities with other elements. To best represent the problem, the different possible missions and classifications of potential threats are finely modeled. By associating them with the information available to the defender, the device 1 deduces the future trajectory of a threat, and also assesses the target of the attack, as well as the characteristics of the threat such as its nature and its weaponry.
    The device 1 and the prediction method, as described above, are preferably implemented to protect an area from an air attack as indicated above, but they can be extended to a deployment aid. We can indeed compare the effectiveness of two deployments against a type of air attack by comparing the values obtained for the usefulness of the defender in each deployment.
    In addition, this device 1 and this method can be used as an aid to mission planning. It is enough, in this case, to place oneself from the point of view of the attacker, by implementing the aforementioned operations, and to take into account the known or supposed data available to the attacker on the two protagonists.
    While determining the best solutions for attacking the opponent, the device 1 calculates the best defense solution (s), in the form of a sequence of decisions. The properties of equilibrium mean that the attack and defense solutions thus calculated are the best possible for both players. This approach can be generalized to many decision support systems in the military field, in particular the systems for planning and conducting missions used in weapons systems or even in staffs, or systems of preparation of military operations, at the tactical or even strategic level.
    权利要求:
    Claims (16)
    [1" id="c-fr-0001]
    1. Device for predicting an optimal attack solution and a corresponding optimal defense solution, in a scenario of at least potential military conflict between an attacker and a defender, characterized in that it comprises:
    - a data entry set (2) configured to enter said device (1), both attacker data and defender data, the defender data being relating to a ground site to be protected provided defense capabilities, and the attacker's data relating to an aerial attack on said site on the ground to be protected and comprising detected current data, the attacker's data comprising at least some of the following data:
    • models of attack procedures;
    • types of threats and associated characteristics;
    • at least one detected position from the appearance of a threat;
    • at least one position assumed to be a threat; and • the attacker's preferences, and the defender's data comprising at least some of the following data:
    • possible missions;
    • data from detection means;
    • characteristics of detection means;
    • positions of potential targets in the area to be defended;
    • defense capabilities of the area to be defended; and • defender preferences;
    - a central unit (3) comprising:
    • a modeling unit (4) configured to generate an evaluated game tree, from said input data, within the framework of game theory;
    • a resolution unit (6) configured to define a game balance within the framework of game theory, a balance defining a pair of attacking strategy and defending strategy;
    • an interpretation unit (7) configured to determine, from the game balance, an optimal attack solution and an optimal defense solution, best suited to this optimal attack solution; and
    - an information transmission unit (9) configured to transmit to an operator or a user system, at least said optimal attack solution and said optimal defense solution.
    [2" id="c-fr-0002]
    2. Device according to claim 1, characterized in that the data input assembly (2) comprises at least some of the following elements:
    - an input element (12) allowing an operator to enter data to be entered;
    - a data loading element (14) associated with a memory (15) and configured to load data into said memory (15).
    [3" id="c-fr-0003]
    3. Device according to one of claims 1 and 2, characterized in that it comprises at least one detector (D1, DN) capable of detecting current data relating to an attacking means of an attacker, and in that that the data entry assembly (2) comprises a data transmission link (18) allowing automatic entry of current data detected by the detector (D1, DN).
    [4" id="c-fr-0004]
    4. Device according to any one of claims 1 to 3, characterized in that the information transmission unit (9) comprises:
    - display means (19) which are connected via a link (10) to the central unit (3); or
    - A link (11) for transmitting, at least said optimal attack solution and said optimal defense solution, to a user system.
    [5" id="c-fr-0005]
    5. Device according to any one of claims 1 to 4, characterized in that the optimal defense solution consists in defining a commitment proposal comprising an optimal allocation plan which specifies an arms allocation of the site to be defended and provides firing dates for these weapons to destroy threats, with the aim of optimally allocating the resources available to deal with threats by maximizing the survival of strategic points defended.
    [6" id="c-fr-0006]
    6. Method for predicting an optimal attack solution and a corresponding optimal defense solution, in a scenario of at least potential military conflict between an attacker and a defender, characterized in that it comprises:
    a step (E1) of data entry, implemented by a set of data entry (2), consisting in entering, at least initially before the implementation of the method, both data of the attacker and defender data, the defender data relating to a ground site to be protected provided with defense capabilities, and the attacker's data being relating to an air attack of said ground site to be protected and comprising current data detected, the attacker's data comprising at least some of the following data:
    • models of attack procedures;
    • types of threats and associated characteristics;
    • at least one detected position from the appearance of a threat;
    • at least one position assumed to be a threat; and • the attacker's preferences, and the defender's data comprising at least some of the following data:
    • possible missions;
    • data from detection means;
    • characteristics of detection means;
    • positions of potential targets in the area to be defended;
    • defense capabilities of the area to be defended; and • defender preferences;
    - a scenario modeling step (E2), implemented by a modeling unit (4), consisting in generating a game tree evaluated from said data entered within the framework of game theory;
    - a resolution step (E3), implemented by a resolution unit (6), consisting in defining a game equilibrium within the framework of game theory, an equilibrium defining a pair of attacking and defending strategies;
    an interpretation step (E4), implemented by an interpretation unit (7), consisting in determining, from the game balance, an optimal attack solution and an optimal defense solution, the better suited to this optimal attack solution; and
    - a step (E5) of information transmission, implemented by an information transmission unit (9), consisting in transmitting to an operator or a user system, at least said optimal attack solution and said solution optimal defense.
    [7" id="c-fr-0007]
    7. Method according to claim 6, characterized in that it applies to at least one of the following situations, relating to a scenario of military conflict:
    - anti-aircraft defense;
    - air, defensive or offensive combat;
    5 - mission planning for hitting targets.
    [8" id="c-fr-0008]
    8. Method according to one of claims 6 and 7, characterized in that the optimal defense solution consists in defining a commitment proposal comprising an optimal allocation plan which specifies an arms allocation of the site to be defended and provides dates for firing these weapons to destroy threats, in order to best allocate the resources available to deal with threats by maximizing the survival of strategic points defended.
    [9" id="c-fr-0009]
    9. Method according to any one of claims 6 to 8, characterized in that the step (E4) of data entry also consists of entering
    15 of the current data detected, at least by the attacker, during the implementation of the method.
    [10" id="c-fr-0010]
    10. Method according to any one of claims 6 to 9, characterized in that the modeling step (E2) comprises:
    - a first sub-step consisting in generating a set of possible strategies
    20 using some of the data entered;
    - a second sub-step consisting in generating a game tree within the framework of game theory, from said set of strategies;
    - a third sub-step consisting in evaluating said game tree, using data from some of the data entered.
    25
    [11" id="c-fr-0011]
    11. The method as claimed in claim 10, characterized in that the third sub-step consists, from pairs of strategies, of evaluating at least one pair of attacking strategy and defending strategy, and assigning a value to the attacker and worth to the defender.
    [12" id="c-fr-0012]
    12. Method according to any one of claims 6 to 11,
    30 characterized in that the resolution step (E3) comprises:
    - a first sub-step consisting in pruning the received game tree, from data relating to threats, so as to form a reduced game tree; and
    - a second sub-step consisting in determining the game balance, from this reduced game tree, to deduce a couple of attacking and defending strategies.
    [13" id="c-fr-0013]
    13. The method of claim 12, characterized in that the first sub-step also uses detected current data, entered during the implementation of the method, to prune the received game tree.
    [14" id="c-fr-0014]
    14. Method according to any one of claims 6 to 13, characterized in that the interpretation step (E4) comprises:
    - a first substep consisting in interpreting the optimal attack solution; and
    - a second sub-step consisting in interpreting the optimal defense solution best suited to this optimal attack solution making it possible to define rules of engagement for the defender.
    [15" id="c-fr-0015]
    15. Method according to claim 14, characterized in that the interpretation step (E4) also includes a substage for evaluating the dangerousness of the threat.
    [16" id="c-fr-0016]
    16. Method according to one of claims 14 and 15, characterized in that the interpretation step (E4) also comprises a sub-step for evaluating the probability of success of the optimal attack solution and the probability of success of the optimal defense solution.
    类似技术:
    公开号 | 公开日 | 专利标题
    Thomas2020|Kosovo and the current myth of information superiority
    Lieber et al.2017|The new era of counterforce: Technological change and the future of nuclear deterrence
    EP3370219A1|2018-09-05|Method and device for predicting optimal attack and defence solutions in the scenario of a military conflict
    Gilli et al.2016|The diffusion of drone warfare? Industrial, organizational, and infrastructural constraints
    Johnson et al.1995|Dominant battlespace knowledge: The winning edge
    Plaw et al.2015|The drone debate: A primer on the us use of unmanned aircraft outside conventional battlefields
    Yoo2017|Embracing the machines: rationalist war and new weapons technologies
    Østensen et al.2018|Russian use of private military and security companies-the implications for European and Norwegian Security
    Krause2008|The last good chance: A reassessment of US operations at Tora Bora
    Libicki1999|Illuminating tomorrow's war
    Phelps2021|On Killing Remotely: The Psychology of Killing with Drones
    Grimal et al.2018|Combat drones: Hives, swarms, and autonomous action?
    Jenks2016|The gathering swarm: the path to increasingly autonomous weapons systems
    Puckett2004|In this era of smart weapons, is a state under an international legal obligation to use precision-guided technology in armed conflict
    Gormley2009|The Path to Deep Nuclear Reductions. Dealing with American Conventional Superiority
    Marton2017|Evolution in military affairs in the battlespace of Syria and Iraq
    Rios2009|Sun Tzu was a hacker: An examination of the tactics and operations from a real world cyber attack
    Giachetti2015|System of systems capability needs analysis via a stochastic network model
    Shultz Jr2020|Transforming US Intelligence for Irregular War: Task Force 714 in Iraq
    Valeriano et al.2020|De-Escalation Pathways and Disruptive Technology: Cyber Operations as Off-Ramps to War
    US11276324B2|2022-03-15|Method and device for predicting optimum attack and defence solutions in a military conflict scenario
    Hacohen2019|West Bank's Area C: Israel's Eastern Line of Defense
    Gilli et al.2013|The Diffusion of Drone Warfare? Industrial, Infrastructural and Organizational Constraints
    Jones2014|Long-Range Precision-Strike Cruise Missiles in NATO Operations
    Machingo2015|CLOSE AIR SUPPORT | FOR COUNTERINSURGENCY | AND THE UPWARD TRAJECTORY OF UNMANNED AIRCRAFT SYSTEMS |: NAVIGATING THE UNDISCOVERED DOMAIN
    同族专利:
    公开号 | 公开日
    FR3063554B1|2021-04-02|
    US20190385473A1|2019-12-19|
    RU2726394C1|2020-07-13|
    WO2018158510A1|2018-09-07|
    IL269011D0|2019-10-31|
    SG11201907915UA|2019-09-27|
    EP3370219A1|2018-09-05|
    CN110574091A|2019-12-13|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    FR2767770A1|1997-09-01|1999-03-05|Alsthom Cge Alcatel|Method of resolution of conflicts in rail network using computer|
    US20050183569A1|2002-04-22|2005-08-25|Neal Solomon|System, methods and apparatus for managing a weapon system|
    US7194353B1|2004-12-03|2007-03-20|Gestalt, Llc|Method and system for route planning of aircraft using rule-based expert system and threat assessment|
    US20150054826A1|2009-03-19|2015-02-26|Real Time Companies|Augmented reality system for identifying force capability and occluded terrain|
    US9177259B1|2010-11-29|2015-11-03|Aptima Inc.|Systems and methods for recognizing and reacting to spatiotemporal patterns|
    FR2975784A1|2011-05-27|2012-11-30|Thales Raytheon Systems Company Sas|Evaluation device for anti-aircraft defense system, has calculation unit calculating final instant of impact of defending point in absence of destruction of enemy aerodyne and notice time for path, and determination unit connected to unit|
    FR2987108A1|2012-02-22|2013-08-23|Gdi Simulation|COMBAT SHOOTING SIMULATOR FOR WEAPONS ASSOCIATED WITH ABSOLUTE OPERATION AND RELATED FIELD OPERATION|
    EP2765545A1|2013-02-08|2014-08-13|MBDA France|Method and device for multi objective optimisation|
    EP2765544A1|2013-02-08|2014-08-13|MBDA France|Method and device for optimizing a resource allocation plan|
    US9524358B1|2013-06-10|2016-12-20|The United States Of America As Represented By The Secretary Of The Navy|Core-modular interoperability architecture for modeling and simulation|
    FR3015023A1|2013-12-18|2015-06-19|Dassault Aviat|SYSTEM FOR AIDING THE PREPARATION OF A MISSION OF ONE OR MORE AIRCRAFT OR AIRCRAFT GROUPS AND CORRESPONDING METHOD|
    EP2887277A1|2013-12-18|2015-06-24|Thales Nederland B.V.|Method, system and computer program for decision support|
    EP2899491A1|2014-01-22|2015-07-29|MBDA France|Device for marking a target and target-processing system comprising such a target-marking device|
    US20170309198A1|2016-04-26|2017-10-26|Visual Awareness Technologies & Consulting, Inc.|Systems and methods for determining mission readiness|RU2734144C1|2019-11-27|2020-10-13|ФЕДЕРАЛЬНОЕ ГОСУДАРСТВЕННОЕ КАЗЕННОЕ ВОЕННОЕ ОБРАЗОВАТЕЛЬНОЕ УЧРЕЖДЕНИЕ ВЫСШЕГО ОБРАЗОВАНИЯ "Военная академия Ракетных войск стратегического назначения имени Петра Великого" МИНИСТЕРСТВА ОБОРОНЫ РОССИЙСКОЙ ФЕДЕРАЦИИ|Device for simulation of process of antiaircraft means operation|US7194153B1|2003-05-26|2007-03-20|Omron Corporation|Optical switch device|
    RU2345421C1|2007-06-05|2009-01-27|Закрытое акционерное общество Главное Управление Научно-Производственное Объединение "Стройтехавтоматика"|Testing simulate training desk "monoblock" for preparation, control, assessment and quality prediction of remote monitoring and blocking of critical infrastructure, equipped with operators' smart support devices|
    CN102111384A|2009-12-28|2011-06-29|北京安码科技有限公司|Game tree security automatic response method and device adopting same|
    US8610041B1|2011-05-23|2013-12-17|Lockheed Martin Corporation|Missile tracking by GPS reflections|
    US9737988B2|2014-10-31|2017-08-22|Intelligent Fusion Technology, Inc|Methods and devices for demonstrating three-player pursuit-evasion game|
    CN106169264B|2016-08-23|2018-08-17|济南环太机电技术有限公司|Simulated training system is blocked in aircraft carrier ejection|
    CN106355970A|2016-08-31|2017-01-25|上海交通大学|Virtual military training system|
    CN106469317A|2016-09-20|2017-03-01|哈尔滨工业大学深圳研究生院|A kind of method based on carrying out Opponent Modeling in non-perfect information game|
    CN106453379B|2016-10-28|2018-10-16|华中科技大学|The security strategy dynamic acquisition method of Process Control System based on attacking and defending game|CN109658742B|2018-12-28|2020-07-10|北京信息科技大学|Dense flight autonomous conflict resolution method based on preorder flight information|
    CN110411278B|2019-06-20|2021-11-23|合肥工业大学|Nash equilibrium game intelligent decision method and system for deploying multiple unmanned aerial vehicles|
    法律状态:
    2018-03-29| PLFP| Fee payment|Year of fee payment: 2 |
    2018-09-07| PLSC| Search report ready|Effective date: 20180907 |
    2019-03-29| PLFP| Fee payment|Year of fee payment: 3 |
    2020-03-31| PLFP| Fee payment|Year of fee payment: 4 |
    2021-03-30| PLFP| Fee payment|Year of fee payment: 5 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1700216|2017-03-03|
    FR1700216A|FR3063554B1|2017-03-03|2017-03-03|METHOD AND DEVICE FOR PREDICTING OPTIMAL ATTACK AND DEFENSE SOLUTIONS IN A MILITARY CONFLICT SCENARIO|FR1700216A| FR3063554B1|2017-03-03|2017-03-03|METHOD AND DEVICE FOR PREDICTING OPTIMAL ATTACK AND DEFENSE SOLUTIONS IN A MILITARY CONFLICT SCENARIO|
    SG11201907915UA| SG11201907915UA|2017-03-03|2018-02-19|Method and device for predicting optimum attack and defence solutions in a military conflict scenario|
    US16/489,998| US11276324B2|2017-03-03|2018-02-19|Method and device for predicting optimum attack and defence solutions in a military conflict scenario|
    PCT/FR2018/000031| WO2018158510A1|2017-03-03|2018-02-19|Method and device for predicting optimum attack and defence solutions in a military conflict scenario|
    EP18290011.8A| EP3370219A1|2017-03-03|2018-02-19|Method and device for predicting optimal attack and defence solutions in the scenario of a military conflict|
    RU2019127032A| RU2726394C1|2017-03-03|2018-02-19|Method and device for predicting optimal solutions of attack and defence in scenario of military conflict|
    CN201880014864.5A| CN110574091A|2017-03-03|2018-02-19|Method and apparatus for predicting optimal attack and defense solutions in military conflict scenarios|
    IL26901119A| IL269011D0|2017-03-03|2019-08-29|Method and device for predicting optimum attack and defence solutions in a military conflict scenario|
    [返回顶部]