Method for obtaining optimal strategy through mutual attack and defense game
Through geometric methods, the barrier plane is constructed in the Cartesian coordinate system, and the winning areas and draw areas of the target defense game are divided, which solves the problem of strategy acquisition in complex game scenarios in multi-player systems, and realizes the optimal strategy provision at any initial position , expanding the application of target defense games.
Patent Information
- Application Number
- CN202410949619.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-16
- Publication Date
- 2024-11-08
AI Technical Summary
In the target defense game of multi-player systems, it is difficult for existing technology to effectively divide state space and obtain optimal strategies, especially when analyzing challenging integral recursive path equations and building barrier planes in complex game scenarios.
The barrier plane is constructed using geometric methods, and the winning area and the draw area are divided in the Cartesian coordinate system through the vertical bisector of the central connection line of the two targets and the initial position line, giving the optimal strategy for different regions.
It provides the optimal strategy for constant-speed gamers at any initial position, expands the application possibility of target defense games, and solves the problem of strategy acquisition in complex game scenarios in the existing technology.
Smart Images

Figure CN118915437A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of target defense game, and in particular to a method for obtaining an optimal strategy in a mutual attack and defense game. Background Art
[0002] In recent decades, confrontation issues in aerospace, automatic control, intelligent robots and other fields have been frequently raised and gradually solved. Pursuit and escape game is a popular method to model the conflict scenario between multiple players. Among the various pursuit and escape game studies, the target defense game is considered to be a popular branch, which involves two teams: the defense team is tasked with protecting the target, and the invasion team is tasked with invading the target. The other two variants are the perimeter defense game and the reconnaissance game. In the former problem, the defender patrols along the border, and in the latter problem, the invader scouts the target area and then returns to the safe zone.
[0003] There are two important concepts when dealing with pursuit-escape games: class games and degree games. The former indicates the outcome of the game, and the latter gives the actual strategy of the corresponding class game. A common class game solution is by determining the barrier plane. The classic barrier plane construction method starts from a point on the boundary of the available part of the target set and then integrates the recursive path equation. However, for multi-player systems, the solution of the integrated recursive path equation is often analytically challenging. Therefore, the following methods are proposed to solve these complex problems: first, divide the complex game into several sub-games and then analyze the optimal behavior of each player in each sub-game; second, use explicit strategy methods; third, numerical approximation methods for solving the HJI equation.
[0004] Existing research progress has basically solved most of the existing problems. Therefore, a new scenario is proposed here, a variant of the target defense game, namely the mutual attack and defense game. Different from the traditional method of constructing the natural barrier plane, the barrier plane is constructed here using a geometric method. Summary of the invention
[0005] Based on this, in order to address the shortcomings of the prior art, the present invention discloses a method for obtaining the optimal strategy for a mutual attack and defense game, which divides the game into winning areas and draw confrontation areas for both parties according to their initial states, and gives the optimal strategies for both parties when they belong to different areas.
[0006] In order to achieve the above object, the technical solution adopted by the present invention is:
[0007] A method for obtaining an optimal strategy in a mutual attack and defense game, wherein two players in the mutual attack and defense game are set to be player 1 and player 2, the target to be protected by player 1 is target 1, and the target to be protected by player 2 is target 2. The positions of the two targets in the mutual attack and defense game, the target radius, and the initial positions of the two players are all known, and the two targets do not intersect; the method comprises:
[0008] A Cartesian coordinate system is established with the center line of the two targets as the X-axis and the perpendicular bisector of the center line of the two targets as the Y-axis, and the initial positions of the two players are respectively located on both sides of the Y-axis. With the Y-axis as the boundary, the side where the initial position of each player is located is the capture advantage area of the player.
[0009] A perpendicular bisector is made of the line connecting the initial positions of the two players in the Cartesian coordinate system, a barrier plane is constructed according to the perpendicular bisector of the line connecting the initial positions of the two players to divide the entire state space into different areas, and the optimal strategies of the two players are obtained for the different areas divided by the barrier plane.
[0010] In some embodiments, the steps of constructing a barrier plane according to the perpendicular bisector of the line connecting the initial positions of the two players to divide the entire state space into different regions, and obtaining the optimal strategies of the two players for the different regions divided by the barrier plane are specifically:
[0011] The distance between the perpendicular bisector of the line connecting the initial positions of the two players and the centers of the two targets is obtained through coordinates, and a barrier plane is constructed according to the distance between the perpendicular bisector of the line connecting the initial positions of the two players and the centers of the two targets and the target radius to divide the entire state space into winning areas and draw areas for the two players, and the best aiming points for the two players are given for the different areas divided by the barrier plane, thereby obtaining the optimal strategies for the two players.
[0012] In some of the embodiments, when the perpendicular bisector of the line connecting the initial positions of the two players is less than the target radius from the center of target 1 and greater than the target radius from the center of target 2, this corresponds to the winning area of player 2.
[0013] In some of the embodiments, when the perpendicular bisector of the line connecting the initial positions of the two players is less than the target radius from the center of target 2 and greater than the target radius from the center of target 1, this corresponds to the winning area of player 1.
[0014] In some of the embodiments, when the distances from the perpendicular bisector of the line connecting the initial positions of two players to the centers of target 1 and target 2 are both smaller than the target radius, the distances from the perpendicular bisector of the line connecting the initial positions of the two players to the centers of target 1 and target 2 are compared. If it is closer to target 1, it corresponds to the winning area of player 2; if it is closer to target 2, it corresponds to the winning area of player 1.
[0015] In some of the embodiments, when the distance between the perpendicular bisector of the line connecting the initial positions of the two players and the centers of target 1 and target 2 is greater than the target radius, if it is determined that the intersection of the perpendicular bisector of the line connecting the initial positions of the two players and the X-axis is located between the centers of target 1 and target 2, then this corresponds to the draw area of player 1 and player 2.
[0016] In some of the embodiments, if it is determined that the perpendicular bisector of the line connecting the initial positions of the two players and the intersection of the X-axis are located outside the center of target 1, then it corresponds to the winning area of player 2.
[0017] In some of the embodiments, if it is determined that the perpendicular bisector of the line connecting the initial positions of the two players and the intersection of the X-axis are located outside the center of the target 2, then it corresponds to the winning area of the player 1.
[0018] In some of these embodiments, the optimal heading angles for the two players are constant.
[0019] In some of these embodiments, the speeds of the two players are equal.
[0020] Compared with the prior art, the present invention has at least one of the following beneficial effects:
[0021] The present invention provides a method for obtaining the optimal strategy in a mutual attack and defense game. For two players with equal speeds at any initial position, a barrier plane can be constructed by a geometric method according to the initial positions of the two players, so as to divide the entire state space into winning areas and draw areas for the two players, and the optimal strategies of the two players corresponding to different areas under any initial state are given in combination with the barrier plane. This new scenario has not been mentioned in existing work, thereby expanding more possibilities for the implementation of this field in reality. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] The present invention and its features, appearance and advantages will become more apparent by reading the detailed description of non-limiting embodiments with reference to the following drawings. The same reference numerals indicate the same parts in all the drawings. The drawings are not drawn to scale, and emphasis is placed on illustrating the subject matter of the present invention.
[0023] Figure 1 A schematic diagram of a scene constructed by a Cartesian coordinate system according to an embodiment of the present invention;
[0024] Figure 2 A two-dimensional schematic diagram of a cutting plane of the method for obtaining an optimal strategy for a mutual attack and defense game according to the present invention;
[0025] Figure 3 One of the examples of the method for obtaining the optimal strategy for the mutual attack and defense game according to the present invention;
[0026] Figure 4 One of the examples of the method for obtaining the optimal strategy for the mutual attack and defense game according to the present invention;
[0027] Figure 5 One of the examples of the method for obtaining the optimal strategy for the mutual attack and defense game according to the present invention;
[0028] Figure 6 One of the examples of the method for obtaining the optimal strategy for the mutual attack and defense game according to the present invention;
[0029] Figure 7 One of the examples of the method for obtaining the optimal strategy for the mutual attack and defense game according to the present invention;
[0030] Figure 8 These are different terminals of the method for obtaining the optimal strategy for the mutual attack and defense game according to the present invention. DETAILED DESCRIPTION
[0031] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0032] Reference Figure 1-8 The present invention discloses a method for obtaining an optimal strategy in a mutual attack and defense game. Considering a mutual attack and defense game on a plane, the mutual attack and defense game has two players (the players are intelligent agents), and the roles of the two players as invaders or defenders are not clearly distinguished. The two players in the mutual attack and defense game are set as player 1 and player 2, respectively, denoted as A1 and A2. The target protected by player 1 is target 1, denoted as T1, and the target protected by player 2 is target 2, denoted as T2. It is assumed that both targets are circular areas with a radius of r, and the distance between the centers of the two targets is 2d. The positions of the two targets in the mutual attack and defense game, the target radius and the initial positions of the two players are all known, and the two targets do not intersect, wherein the two players have all the information of the opponent and know the state of the other party at any time. Compared with the existing work, the present invention deals with a more practical scenario, in which each player must perform two tasks at the same time, namely protecting his own target and invading the opponent's target.
[0033] The construction of the barrier surface of the mutual attack and defense game is related to the initial state. The barrier surface is a classic way to solve similar game problems to divide the entire space into winning spaces for different players. There are many ways to construct it. The present invention uses a known initial state (i.e., initial position) to solve the barrier surface of the mutual attack and defense game by a geometric method; different initial states correspond to different optimal aiming points, and then the optimal strategies of the two players are obtained respectively;
[0034] Specifically, the method for obtaining the optimal strategy in the above-mentioned mutual attack and defense game includes:
[0035] Step S1, connect the centers of the two targets, make the perpendicular bisector of the center line of the two targets as the Y axis, take the center line of the two targets as the X axis, and establish a Cartesian coordinate system based on this, and the initial positions of the two players are located on both sides of the Y axis respectively; when the speeds of the two gamers are equal, if the capture advantage is not specified, it is impossible to determine the winner of the game when a collision occurs. The basic concept of the capture advantage area is that the game result of the collision in the player's area points to the victory of the player. Therefore, the perpendicular bisector of the center line of the two targets is used as the dividing line of the capture advantage area of both parties (that is, the Y axis is the dividing line of the capture advantage area of the two players), and the two sides of the Y axis belong to the capture advantage area of player 1 and player 2 respectively; that is, with the Y axis as the boundary, the side where the initial position of each player is located belongs to the capture advantage area of the player.
[0036] Specifically, a perpendicular bisector of the center line of the two targets is drawn to divide the plane into two parts, which are recorded as S1 and S2. Without loss of generality, this scene is placed in a Cartesian coordinate system, the centers of the two targets are fixed on the X-axis, and the corresponding perpendicular bisector is fixed on the Y-axis.
[0037] Step S2, draw a perpendicular bisector of the line connecting the initial positions of the two players in the Cartesian coordinate system, construct a barrier plane based on the perpendicular bisector of the line connecting the initial positions of the two players to divide the entire state space into different areas, and obtain the optimal strategies of the two players for the different areas divided by the barrier plane.
[0038] In the above step S2, the step of constructing a barrier plane according to the perpendicular bisector of the line connecting the initial positions of the two players to divide the entire state space into different areas, and obtaining the optimal strategies of the two players for the different areas divided by the barrier plane is specifically as follows: obtaining the distance from the perpendicular bisector of the line connecting the initial positions of the two players to the centers of the two targets through coordinates, and constructing a barrier plane according to the distance from the perpendicular bisector of the line connecting the initial positions of the two players to the centers of the two targets and the radius of the targets to divide the entire state space into winning areas and draw areas for the two players, and giving the best aiming points for the two players for the different areas divided by the barrier plane, thereby obtaining the optimal strategies for the two players.
[0039] Specifically, when the distance from the perpendicular bisector of the line connecting the initial positions of the two players to the center of target 1 is less than the target radius, and the distance from the center of target 2 is greater than the target radius, then this corresponds to the winning area of player 2; when the distance from the perpendicular bisector of the line connecting the initial positions of the two players to the center of target 2 is less than the target radius, and the distance from the center of target 1 is greater than the target radius, then this corresponds to the winning area of player 1; when the distance from the perpendicular bisector of the line connecting the initial positions of the two players to the centers of target 1 and target 2 are both less than the target radius, compare the distances from the perpendicular bisector of the line connecting the initial positions of the two players to the centers of target 1 and target 2. If it is closer to target 1, then it corresponds to The winning area of player 2; if it is closer to target 2, it corresponds to the winning area of player 1; when the perpendicular bisector of the line connecting the initial positions of the two players is greater than the target radius from the center of target 1 and target 2, if the intersection of the perpendicular bisector of the line connecting the initial positions of the two players and the X-axis is judged to be between the centers of target 1 and target 2, then this corresponds to the tie area of player 1 and player 2; if the intersection of the perpendicular bisector of the line connecting the initial positions of the two players and the X-axis is judged to be outside the center of target 1, then it corresponds to the winning area of player 2; if the intersection of the perpendicular bisector of the line connecting the initial positions of the two players and the X-axis is judged to be outside the center of target 2, then it corresponds to the winning area of player 1.
[0040] Specifically, the states of the two players are represented by x1=(x1,y1) and x2=(x2,y2). The entire state can be represented as in, Indicates that the state space is four-dimensional. The control quantities of players 1 and 2, i.e. A1 and A2, are their instantaneous direction angles, where u1 = φ and u2 = ψ. The dynamic equation It is represented by a system of differential equations:
[0041]
[0042] Among them, v1 and v2 are the speeds of player 1 and player 2 respectively. are the rates of change of player 1's position, are the rates of change of the position of player 2, that is, the rates of change of player 1 and player 2 on the X-axis and Y-axis respectively. The present invention considers a realistic scenario. If two players collide, the player wreckage belonging to their space can be repaired and put into use again. This can be simplified as follows: the result of the collision depends entirely on the positions of the two players. That is, a collision occurring at S1 can make player 1 win, and a collision occurring at S2 can make player 2 win; the centers of the two targets are located at (-d, 0) and (d, 0) respectively. The two non-intersecting advantage areas are recorded as and The target areas are represented as and in, It means that the state space is two-dimensional.
[0043] The background of the problems considered in this invention is based on the following assumptions:
[0044] Assumption 1: The distance between the centers of two targets is greater than the diameter of each target, that is, d>r;
[0045] Assumption 2: The maximum speed of the two players is the same, that is, v1 = v2 = v max ;
[0046] Assumption 3: The initial states of the two players satisfy x1∈S1, x2∈S2;
[0047] Assumption 4: Each player has complete information about the other player and considers a non-participating information structure.
[0048] The outcome of the game depends on the choice of parameters. Assumption 1 ensures that the two goals do not intersect, Assumptions 2 and 3 ensure the fairness of the game, and Assumption 4 stipulates that neither player knows the opponent's current decision. The unexpected information structure mentioned in Assumption 4 means that players cannot obtain future information when making current decisions.
[0049] Traditional target defense games usually have two outcomes: the invader successfully invades or the defender successfully captures. However, in the mutual attack and defense game, there is an additional terminal outcome - a draw, that is, the game will end under one of the following conditions:
[0050]
[0051] in
[0052]
[0053] They represent the results of A1 breaking through T2 and A2 breaking through T1.
[0054]
[0055] Indicates the result of a tie between A1 and A2.
[0056] in, represents the terminal set of the game, The terminal set representing A1's victory; The terminal set representing A2's victory; The set of terminals that represent a tie between A1 and A2.
[0057] In the game-like problem, the purpose of constructing the barrier plane is usually to divide the plane into different winning areas. In the present invention, the purpose is to determine the winning area and the draw area. When the initial states of A1 and A2 are known, the perpendicular bisector of the line between the two points is referred to as the perpendicular bisector l below. o It can be written as y=mx+n, where
[0058]
[0059] In fact, as the game progresses, the coordinates of x1, x2, y1, and y2 are changing in real time, and m and n are also changing in real time.
[0060] Definition 1: Consider a mutual attack and defense game, where the two players have equal speeds, denoted as A1 and A2. The barrier plane can be composed of the following parts:
[0061]
[0062] Where r represents the target radius; From the center of T1 and T2 to the perpendicular bisector l o Distance. B1(x) and B2(x) are the distance from the center of T1 and T2 to the perpendicular bisector l o The difference in distance from the target radius. B3(x) is the distance from the center of T1 and T2 to the perpendicular bisector l o The difference in distance.
[0063] Lemma 1: The entire state space of the mutual attack and defense game can be divided into three different terminal sets based on B1(x), B2(x) and additional judgments, such as Figure 8 As shown in Table 1.
[0064] Proof: First, satisfying B1(x)>0 and B2(x)<0 simultaneously means that the perpendicular bisector l o The distance from the center of target 2 is less than the target radius, and the distance from the center of target 1 is greater than the target radius. At this time, the perpendicular bisector l o It only intersects with T2, which means that there is at least one point on T2 that both players can reach at the same time, which corresponds to the winning area of player 1, such as Figure 3 Conversely, when B1(x)<0 and B2(x)>0, it means that the perpendicular bisector l o The distance from the center of target 1 is less than the target radius, and the distance from the center of target 2 is greater than the target radius; we can get the exact opposite conclusion. At this time, the perpendicular bisector l o It only intersects with T1, which means that there is at least one point on T1 that both players can reach at the same time, which corresponds to the winning area of player 1, such as Figure 4 shown.
[0065] Secondly, satisfying both B1(x)<0 and B2(x)<0 means that the perpendicular bisector l o The distances from the centers of target 1 and target 2 are both greater than the target radius, i.e. the perpendicular bisector l o Neither intersects with T1 nor intersects with T2. In this special case, an auxiliary condition is required to determine the terminal set. The parameter here is In fact, it represents the perpendicular bisector l o The intersection with the x-axis. When , it indicates the perpendicular bisector l o The intersection with the x-axis is between the two target centers, i.e. the perpendicular bisector l o The intersection with the X-axis is between the centers of target 1 and target 2. At this time, the perpendicular bisector l o There is no strategy that guarantees victory for either player, so this initial state corresponds to a special termination condition, namely a draw, which corresponds to the draw area of player 1 and player 2, and the corresponding terminal state corresponds to like Figure 5 As shown. When , it means that the perpendicular bisector l o Outside the two target areas; specifically, the perpendicular bisector l o The intersection with the X-axis is located outside the center of target 1 (i.e., to the left of the center of target 1), and T1 is completely within the advantage area of A2. Therefore, there is more than one point that can make A2 win. This corresponds to the winning area of player 2. The corresponding terminal state corresponds to like Figure 7 As shown. Vice versa; that is, the perpendicular bisector l o The intersection with the X-axis is located outside the center of target 2 (that is, to the right of the center of target 2), and T2 is completely within the advantage area of A1. Therefore, there is more than one point that can make A1 win. This corresponds to the winning area of player 1. The corresponding terminal state corresponds to
[0066] Finally, satisfying both B1(x)>0 and B2(x)>0 means that the perpendicular bisector l o The distances from the centers of target 1 and target 2 are both less than the target radius, indicating that the perpendicular bisector l o It intersects both T1 and T2. Here, B3(x) is also needed as an auxiliary judgment condition to analyze the possible results (compare the perpendicular bisector l o The distance from the center of target 1 and target 2). In fact, in this case, when B3(x)<0, it means that the perpendicular bisector l ois closer to the center of T2. Therefore, there is a point on the boundary of T2 that is closer to the perpendicular bisector l than any point on the boundary of T1. o . Relatively close points on the boundary of T2 become easier to reach, i.e. the perpendicular bisector l o Closer to target 2, it corresponds to the winning area of player 1, such as Figure 6 When B3(x)>0, it means the perpendicular bisector l o is closer to the center of T1. Therefore, there is a point on the boundary of T1 that is closer to the perpendicular bisector l than any point on the boundary of T2. o Points relatively close to the boundary of T1 become easier to reach, i.e., the perpendicular bisector l o The closer to target 1, the corresponding winning area for player 2.
[0067] Theorem 1 and Lemma 1 consider y1≠y2. In fact, we need to consider another special case, that is, y1=y2, in which case (5) does not exist. In this case, the difference lies in the definition of d1 and d2, which can be restated as d′1 and d′2, where Represents the center of T1 and T2 to the perpendicular bisector l o The perpendicular bisector l o Based on the relationship y1=y2, it can be updated as The further analysis has many similarities with Lemma 1 and is omitted here.
[0068] Lemma 2: First consider the case where y1≠y2. When B1(x)>0 and B2(x<0, the best aiming points for both player 1 and player 2 are
[0069]
[0070] When B1(x)<0 and B2(x)>0, the best aiming points for both Player 1 and Player 2 are
[0071]
[0072] When B1(x)<0 and B2(x)<0, according to The relationship between and d further determines the optimal aiming point. The best aiming point is
[0073] x * =0,y * =n. (9)
[0074] when , the optional aiming points for players 1 and 2 are
[0075]
[0076] when , the optional aiming points for players 1 and 2 are
[0077]
[0078] When B1(x)>0, B2(x)>0 and B3(x)<0, the optimal aiming points of Player 1 and Player 2 are the same as those in Formula (11). When B1(x)>0, B2(x)>0 and B3(x)>0, the optimal aiming points of Player 1 and Player 2 are the same as those in Formula (10).
[0079] Then consider the case where y1=y2. When B1(x)>0 and B2(x)<0, the best aiming points for both player 1 and player 2 are
[0080]
[0081] When B1(x)<0 and B2(x)>0, the best aiming points for both Player 1 and Player 2 are
[0082]
[0083] When B1(x)<0 and B2(x)<0, it is necessary to further determine the relationship between 2d and d′1 and d′2. When 2d=d′2-d′1, the best aiming points of player 1 and player 2 are the same as those in equation (10). When 2d=d′1-d′2, the best aiming points of player 1 and player 2 are the same as those in equation (11).
[0084] Proof: When the optimal aiming point represents a point on the barrier plane, it is necessary to ensure that the advantages of both parties remain unchanged in the game. The solution to the optimal aiming point is related to the semi-permeability of the barrier surface, that is, with the help of B1(x) and B2(x), the plane can be divided into areas where one party may win. When only B1(x)>0 or B2(x)>0 is satisfied, it means that player 1 or player 2 may win. When both B1(x)>0 and B2(x)>0 are satisfied, it is necessary to further determine which party is the real winner through B3(x).
[0085] First, consider the case where y1≠y2. Based on the above analysis, when B1(x)>0 and B2(x<0, that is, the perpendicular bisector l o When it intersects only with T2, the best aiming point can be found. On the contrary, when B1(x)<0 and B2(x)>0, that is, the perpendicular bisector l o When it intersects only with T1, the best aiming point can be found. When B1(x)<0 and B2(x)<0, according to The relationship with d again determines the best aiming point. When the best aiming point is on the perpendicular bisector l o The intersection with the y-axis. There is more than one goal point where player 2 can win; in fact, player 2 has a time advantage when moving to any point on T1, in which case the strategy for both players can be specified as moving toward the center of T1 without having to find an equilibrium point on the barrier plane. When B1(x)>0 and B2(x)>0, further determine B3(x). When B3(x)<0, the best aiming point is the same as (11), and when B3(x)>0, the best aiming point is the same as (10).
[0086] Then consider the case where y1=y2. When B1(x)>0 and B2(x)<0, consider the sign of y1. When B1(x)<0 and B2(x)>0, consider the sign of y2. When B1(x)<0 and B2(x)<0, further determine the optimal aiming point through the relationship between 2d and d′1 and d′2. When 2d=d′1-d′2, the optimal aiming point is the same as equation (11), and when 2d=d′2-d′1, the optimal aiming point is the same as equation (10).
[0087] Theorem 1: Under the optimal gameplay, the optimal heading angles of the two players are constant, that is, the trajectories of the two players are straight lines.
[0088] Proof: The optimal control input for both players can be easily obtained, where the Hamiltonian is
[0089]
[0090] And the common state is Furthermore, the dynamics of the common state is Therefore, all common states and optimal heading angles are constant. Therefore, the optimal trajectory for the player is a straight line.
[0091] Theorem 2: Consider a mutual attack and defense game with two players A1 and A2 of the same speed. The optimal strategy (i.e., the optimal state feedback strategy) for the two players is
[0092]
[0093] where x * and * In different cases, more specific conclusions are derived by considering different terminal sets obtained in Table 1 and different optimal aiming points described in Lemma 2, respectively.
[0094] Proof: Lemma 1 gives how to divide the terminal set in the entire state space, and Lemma 2 gives different optimal aiming points at different positions when the space is divided by the barrier plane. Theorem 1 further proves that the optimal trajectory is a straight line. Therefore, in Theorem 2, the optimal strategies of A1 and A2 should both point to the optimal aiming point (x * ,y * ) is located.
[0095] Note 1: When y1≠y2, the solution of the barrier plane can be solved by the method in other references. By fixing the position of A2, the barrier plane corresponding to A1 can be realized in two-dimensional space, such as Figure 2 As shown. And assumption 3 needs to satisfy x1∈S1, so the curves in the figure are all on the left side of the y-axis.
[0096] The number of devices and processing scales described here are used to simplify the description of the present invention, and the application, modification and variation of the present invention will be obvious to those skilled in the art.
[0097] Although the embodiments of the present invention have been disclosed as above, they are not limited to the applications listed in the specification and implementation modes. They can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily implemented. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to the specific details and the illustrations shown and described herein.
Claims
1. A method for obtaining an optimal strategy in a mutual attack and defense game, wherein two players in the mutual attack and defense game are set to be player 1 and player 2, the target to be protected by player 1 is target 1, and the target to be protected by player 2 is target 2. The positions of the two targets in the mutual attack and defense game, the target radius, and the initial positions of the two players are all known, and the two targets do not intersect; characterized in that: The method comprises: A Cartesian coordinate system is established by taking the center line of the two targets as the X-axis and the perpendicular bisector of the center line of the two targets as the Y-axis, and the initial positions of the two players are located on both sides of the Y-axis respectively, and the side where the initial position of each player is located is the capture advantage area of the player with the Y-axis as the boundary; A perpendicular bisector is made of the line connecting the initial positions of the two players in the Cartesian coordinate system, a barrier plane is constructed according to the perpendicular bisector of the line connecting the initial positions of the two players to divide the entire state space into different areas, and the optimal strategies of the two players are obtained for the different areas divided by the barrier plane.
2. The method for obtaining the optimal strategy through mutual attack and defense game as claimed in claim 1, characterized in that: The steps of constructing a barrier plane according to the perpendicular bisector of the line connecting the initial positions of the two players to divide the entire state space into different areas, and obtaining the optimal strategies of the two players for the different areas divided by the barrier plane are specifically as follows: The distance between the perpendicular bisector of the line connecting the initial positions of the two players and the centers of the two targets is obtained through coordinates, and a barrier plane is constructed according to the distance between the perpendicular bisector of the line connecting the initial positions of the two players and the centers of the two targets and the target radius to divide the entire state space into winning areas and draw areas for the two players, and the best aiming points for the two players are given for the different areas divided by the barrier plane, thereby obtaining the optimal strategies for the two players.
3. The method for obtaining the optimal strategy of mutual attack and defense game as claimed in claim 2, characterized in that: When the perpendicular bisector of the line connecting the initial positions of the two players is less than the target radius from the center of target 1, and greater than the target radius from the center of target 2, this corresponds to the winning area of player 2.
4. The method for obtaining the optimal strategy of mutual attack and defense game as claimed in claim 2, characterized in that: When the perpendicular bisector of the line connecting the initial positions of the two players is less than the target radius from the center of target 2, and greater than the target radius from the center of target 1, this corresponds to the winning area of player 1.
5. The method for obtaining the optimal strategy through mutual attack and defense game as claimed in claim 2, characterized in that: When the distance between the perpendicular bisector of the line connecting the initial positions of the two players and the centers of target 1 and target 2 is less than the target radius, compare the distance between the perpendicular bisector of the line connecting the initial positions of the two players and the centers of target 1 and target 2. If it is closer to target 1, it corresponds to the winning area of player 2; if it is closer to target 2, it corresponds to the winning area of player 1.
6. The method for obtaining the optimal strategy through mutual attack and defense game as claimed in claim 2, characterized in that: When the distance between the perpendicular bisector of the line connecting the initial positions of the two players and the center of target 1 and target 2 is greater than the target radius, if the intersection of the perpendicular bisector of the line connecting the initial positions of the two players and the X-axis is determined to be between the center of target 1 and the center of target 2, then this corresponds to the draw area of player 1 and player 2.
7. The method for obtaining the optimal strategy through mutual attack and defense game as claimed in claim 6, characterized in that: If the intersection of the perpendicular bisector of the line connecting the initial positions of the two players and the X-axis is determined to be outside the center of target 1, it corresponds to the winning area of player 2.
8. The method for obtaining the optimal strategy through mutual attack and defense game as claimed in claim 6, characterized in that: If the intersection of the perpendicular bisector of the line connecting the initial positions of the two players and the X-axis is determined to be outside the center of target 2, it corresponds to the winning area of player 1.
9. The method for obtaining the optimal strategy through mutual attack and defense game as claimed in claim 1, characterized in that: The optimal heading angles of the two players are constant.
10. The method for obtaining the optimal strategy through mutual attack and defense game as claimed in claim 1, characterized in that: Both players have equal speed.
