Method and device for many-to-many pursuit game of unmanned underwater vehicle

Abstract

The invention provides a many-to-many pursuit and escape game method and device for unmanned underwater vehicles, and relates to the technical field of cluster control, and the method comprises the steps: obtaining the positions of a plurality of vehicles, the vehicles comprising pursuit parties and escape parties; respectively constructing motion models of the aircraft based on the position of the aircraft; based on cooperative detection, the pursuing party locates the escape party and determines a target escape object, and the escape party locates the pursuing party and determines a target pursuing object; based on the target pursuit object and the target escape object, establishing a first optimization algorithm between the pursuit party and the target escape object and a second optimization algorithm between the escape party and the target pursuit object; and jointly solving the first optimization algorithm and the second optimization algorithm, and determining the behavior action of the pursuit party and the behavior action of the escape party. According to the invention, many-to-many pursuit game control of the unmanned underwater vehicle can be realized.

Description

Technical Field

[0001] The present application relates to the technical field of cluster control, and in particular to a multi-to-multi pursuit and escape game method and device for unmanned underwater vehicles. Background Art

[0002] As a classic research topic in the field of control, the pursuit-escape problem mainly studies the optimal strategies of the pursuer and the escapee in specific scenarios and their mutual game relationship. By constructing the dynamic motion model and optimization objectives of both parties, this problem can analyze how the pursuer can most effectively shorten the distance with the escapee, and how the escapee can maximize the distance through strategies to escape the pursuit. The research of this type of problem is not only of great theoretical significance, but also widely exists in practical applications, such as in scenarios such as spacecraft docking, aircraft interception, and cooperative control of unmanned systems, helping to optimize path planning and task execution strategies in complex environments.

[0003] At present, the hot research objects of pursuit and escape problems are concentrated on unmanned systems such as drones, unmanned boats and on-orbit spacecraft. These unmanned systems have a common feature, that is, they can use a variety of detection and communication means such as radio, radar and relay satellites, have a wide detection range and long-distance communication capabilities, so that the position, speed and other information of the pursuit and escape parties are more transparent, which helps both parties to formulate more accurate game strategies. However, unmanned underwater vehicles (UUVs) are one of the important directions of unmanned system development. Therefore, a method is needed to realize the multi-to-multi pursuit and escape game control of unmanned underwater vehicles. Summary of the invention

[0004] The present application provides a method and device for a multi-to-multi pursuit and escape game of unmanned underwater vehicles, which can realize multi-to-multi pursuit and escape game control of unmanned underwater vehicles.

[0005] In a first aspect of the present application, a multi-to-multi pursuit and escape game method for unmanned underwater vehicles is provided, the method comprising:

[0006] Acquiring positions of a plurality of aircraft, including the pursuing aircraft and the escaping aircraft;

[0007] constructing motion models of the aircraft respectively based on the positions of the aircraft;

[0008] Based on the collaborative detection, the pursuer locates the fleeing party and determines the target fleeing object, the fleeing party locates the pursuer and determines the target fleeing object, the target fleeing object is the fleeing party that meets the preset conditions among the multiple fleeing parties, and the target pursuing object is the pursuer that meets the preset conditions among the multiple pursuing parties;

[0009] Based on the target pursuit object and the target escape object, establishing a first optimization algorithm between the pursuit party and the target escape object, and a second optimization algorithm between the escape party and the target pursuit object;

[0010] The first optimization algorithm and the second optimization algorithm are jointly solved to determine the behavior actions of the pursuer and the behavior actions of the escaper.

[0011] Optionally, constructing the motion models of the aircraft based on the positions of the aircraft respectively includes:

[0012] According to the relative motion state between the pursuer and the escaper, the motion model is constructed, and the motion model is expressed as follows:

[0013]

[0014]

[0015] Among them, x ij =[(x ij y ij z ij )] T is the (3×1) relative position vector of the ith pursuer relative to the jth escaper, u ij =[(u ij v ij w ij )] T is the (3×1) relative velocity vector of the ith pursuer relative to the jth escaper, a i =[(a xi a yi a zi )] T is the (3×1) acceleration vector of the ith pursuer relative to the inertial system, a j =[(a xj a yj a zj )] T is the (3×1) acceleration vector of the jth escaping party relative to the inertial system;

[0016] Based on the motion model, the following state space equation is obtained:

[0017]

[0018]

[0019]

[0020]

[0021] Among them, y ij is the relative state vector of the ith pursuer relative to the jth escaper (6×1), F is the state coefficient matrix (6×6), G is the control input coefficient matrix (6×3), a ij is the control input made by the i-th pursuer relative to the j-th escaper, a ji is the control input made by the j-th escaper to the i-th pursuer.

[0022] Optionally, based on the target pursuit object and the target escape object, a first optimization algorithm between the pursuit party and the target escape object and a second optimization algorithm between the escape party and the target pursuit object are established, specifically including:

[0023] determining a first leader among the plurality of said pursuit parties;

[0024] A first relative motion model between the first navigator and the target escape object is established. The first relative motion model is expressed as follows:

[0025]

[0026] Among them, y p1-e# is the relative state vector of the first navigator relative to the target escape object, F is the state coefficient matrix (6×6), G is the control input coefficient matrix (6×3), a p1-e# is the control input made by the first navigator relative to the target escape object, a e#-p1 a control input made by the target escape object relative to the first navigator;

[0027] Based on the first relative motion model, a first cost function is determined, where the first cost function is expressed as follows:

[0028]

[0029] Among them, J1 is the first cost function, Q p1-e# is a symmetric semi-positive definite weighting matrix, y T p1-e# is the vector transpose of the relative state vector of the first navigator relative to the target escape object, y p1-e# is the relative state vector of the first navigator relative to the target escape object;

[0030] Identify the second leader among multiple escape parties;

[0031] A second relative motion model between the second navigator and the target object is established. The second relative motion model is expressed as follows:

[0032]

[0033] Among them, y e1-p# is the relative state vector of the second navigator relative to the target pursuit object, F is the state coefficient matrix (6×6), G is the control input coefficient matrix (6×3), a e1-p# is the control input made by the second navigator relative to the target pursuit object, a p#-e1 a control input made by the target pursuit object relative to the second navigator;

[0034] Based on the second relative motion model, a second cost function is determined, where the second cost function is expressed as follows:

[0035]

[0036] Among them, J2 is the second cost function, Q e1-q# is a symmetric semi-positive definite weighting matrix, y T e1-p# is the vector transpose of the relative state vector of the second navigator relative to the target pursuit object, y e1-p# is the relative state vector of the second navigator relative to the target pursuit object.

[0037] Optionally, before jointly solving the first optimization algorithm and the second optimization algorithm to determine the behavior of the pursuer and the behavior of the escaper, the method further includes:

[0038] generating a plurality of first control inputs for the first navigator;

[0039] Based on each of the first control inputs, respectively determine a first output result of the first cost function;

[0040] generating a plurality of second control inputs for the second navigator;

[0041] Based on each of the second control inputs, respectively determine a second output result of the second cost function;

[0042] Determine a first target input among the plurality of first control inputs, and a second target input among the plurality of second control inputs, when a first output result corresponding to the first target input is minimum, and a second output result corresponding to the second target input is maximum;

[0043] A first behavior action of the first navigator is determined based on the first target input, and a second behavior action of the second navigator is determined based on the second target input.

[0044] Optionally, the jointly solving the first optimization algorithm and the second optimization algorithm to determine the behavior of the pursuer and the behavior of the escaper specifically includes:

[0045] According to the first behavior action, updating the behavior action of the pursuer, wherein when updating the behavior action of the pursuer, the position of the pursuer satisfies the communication constraint condition;

[0046] According to the second behavior, the behavior of the escaped party is updated, wherein when the behavior of the escaped party is updated, the position of the escaped party satisfies a communication constraint condition, and the communication constraint condition is as follows:

[0047]

[0048] Among them, (x L (t), y L (t), z L (t)) is the position of the first navigator, (x F (t), y F (t), z F (t)) is the position of the first follower, the first follower is a pursuer other than the first navigator among the plurality of pursuers, d L is the effective communication distance of the first navigator, d F is the effective communication distance of the first follower; or,

[0049] (x L (t), y L (t), z L (t)) is the position of the second navigator, (x F (t), y F (t), z F (t)) is the position of the second follower, the second follower is a runaway party among the plurality of runaway parties except the second leader, d L is the effective communication distance of the second navigator, d F is the effective communication distance of the second follower.

[0050] Optionally, before the pursuing party locates the fleeing party based on the collaborative detection to determine the target fleeing object, and the fleeing party locates the pursuing party to determine the target pursuing object, the method further includes:

[0051] The plurality of pursuit parties respectively determine their own first positions;

[0052] Each of the pursuing parties determines a relative direction to any one of the fleeing parties;

[0053] The second position of any one of the escaped parties is determined according to the plurality of the first positions and the plurality of the relative directions, specifically by calculating the following formula group:

[0054]

[0055] Among them, (x i ,y i , z i ) is the i-th first position, i = 1, 2, ..., k, (x, y, z) is the second position of any escapee, (A i , B i , C i ) is the vector of the i-th relative direction, i = 1, 2,…, k.

[0056] Optionally, based on the collaborative detection, the pursuing party locates the fleeing party to determine the target fleeing object, and the fleeing party locates the pursuing party to determine the target pursuing object, specifically including:

[0057] Determining the distance between any one of the pursuing party and the escaping party's aircraft according to each of the first positions and each of the second positions;

[0058] sorting the plurality of aircraft distances from small to large, determining the smallest aircraft distance among the plurality of aircraft distances, and obtaining the target aircraft distance;

[0059] The target pursuit object corresponding to the distance of the target aircraft among the plurality of the pursuing parties is determined, and the target escape object corresponding to the distance of the target aircraft among the plurality of the escaping parties is determined.

[0060] In a second aspect of the present application, a multi-to-multi pursuit and escape game device for unmanned underwater vehicles is provided, the device comprising an acquisition module, a construction module, a positioning module and a processing module, wherein:

[0061] The acquisition module is used to acquire the positions of multiple aircraft, including the pursuing aircraft and the escaping aircraft;

[0062] The construction module is used to respectively construct the motion model of the aircraft based on the position of the aircraft;

[0063] The positioning module is used for, based on the collaborative detection, the pursuing party to locate the fleeing party and determine the target fleeing object, the fleeing party to locate the pursuing party and determine the target fleeing object, the target fleeing object is the fleeing party that meets the preset conditions among the multiple fleeing parties, and the target pursuing object is the pursuing party that meets the preset conditions among the multiple pursuing parties;

[0064] The building module is used to establish a first optimization algorithm between the pursuit party and the target escaping object, and a second optimization algorithm between the escaping party and the target pursuit object based on the target pursuit object and the target escaping object;

[0065] The processing module is used to jointly solve the first optimization algorithm and the second optimization algorithm to determine the behavior actions of the pursuer and the behavior actions of the escaper.

[0066] In the third aspect of the present application, an electronic device is provided, including a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes any one of the methods described above.

[0067] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores instructions, and when the instructions are executed, any of the methods described above is executed.

[0068] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0069] 1. This application obtains the position information of multiple unmanned underwater vehicles, constructs their motion models respectively, and uses collaborative detection technology to locate the pursuer and the escaper, thereby determining the target pursuit and target escape objects. Based on these objects, the first and second optimization algorithms are established respectively to describe the relative motion of the pursuer and the escaper and their optimization goals. By jointly solving the two optimization algorithms, the optimal behavior of each can be determined, and then the strategic game control of the pursuer and the escaper can be realized, ensuring the effective completion of the multi-to-multi pursuit and escape game process in a complex underwater environment.

[0070] 2. Furthermore, the present application constructs motion models of the pursuer and the fleeer, locates the target object based on collaborative detection, and establishes optimization algorithms respectively, which can effectively analyze and predict the motion behaviors of both parties. By jointly solving the optimization algorithm, the optimal control input can be generated for each navigator, and the optimal behavior strategy can be determined based on the optimal input. This method ensures the real-time and accuracy in many-to-many games, and improves the efficiency and success rate of unmanned vehicle pursuit tasks in underwater environments, especially in communication-restricted and complex scenarios.

[0071] 3. This application uses collaborative detection and mutual positioning of multiple aircraft to accurately determine the relative positions and target objects of the pursuer and the fleeing party. The first positions and relative directions of multiple pursuers are used to accurately calculate the position of the fleeing party, and the optimal pursuit object and fleeing object are selected by sorting the aircraft distances. This method improves the accuracy and efficiency of positioning, ensuring that the most threatening fleeing party and the most capable pursuer can be quickly identified in complex environments, thereby optimizing the pursuit strategy and action decision-making of unmanned aircraft and improving the success rate of missions. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 It is a flowchart of a multi-to-multi pursuit and escape game method of an unmanned underwater vehicle disclosed in an embodiment of the present application;

[0073] Figure 2 It is a module schematic diagram of a multi-to-multi pursuit and escape game device for unmanned underwater vehicles disclosed in an embodiment of the present application;

[0074] Figure 3 It is a structural schematic diagram of an electronic device disclosed in an embodiment of the present application.

[0075] Explanation of the reference numerals: 201, acquisition module; 202, construction module; 203, positioning module; 204, processing module; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. DETAILED DESCRIPTION

[0076] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0077] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0078] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0079] The pursuit and escape problem is a classic topic in the field of control. It studies the optimal strategies and games of the pursuer and the escapee in specific scenarios. By building a dynamic motion model and optimizing goals, the action strategies of both parties are analyzed. This research has wide applications in the fields of spacecraft docking, aircraft interception, and cooperative control of unmanned systems, and optimizes the execution of tasks in complex environments. At present, research hotspots are concentrated on unmanned systems such as drones, unmanned boats, and on-orbit spacecraft. They use a variety of detection methods to enhance information transparency and facilitate accurate strategy formulation. As a development focus, unmanned underwater vehicles also require many-to-many pursuit and escape game methods to cope with complex environments.

[0080] This embodiment discloses a multi-to-multi pursuit and escape game method for unmanned underwater vehicles. Figure 1 , including the following steps S110-S150:

[0081] S110, obtaining positions of multiple aircraft.

[0082] The detection and communication of unmanned underwater vehicles depends on the sonar they are equipped with. A single passive sonar can only detect the direction of the target, but cannot detect the location of the target. Although active sonar can detect the target location, the detection angle range is small, and the use of active sonar will increase its own noise, thereby increasing the risk of exposure, so it cannot be used frequently. Therefore, the underwater pursuit and escape game of unmanned underwater vehicles is constrained by their own limited detection capabilities. One way to make up for the underwater detection capabilities of unmanned underwater vehicles is to use three or more unmanned underwater vehicles to perform passive sonar detection on the same target at the same time, thereby locating the target location.

[0083] The pursuit problem can be classified into one-to-one problem, many-to-one problem, one-to-many problem, and many-to-many problem according to the number of parties involved. In the pursuit problem, it is crucial for any party to know the location of the other party. At present, under the condition that unmanned underwater vehicles generally use passive sonar detection, using multiple unmanned underwater vehicles to coordinate detection and positioning of targets is a feasible way. Under this premise, the unmanned underwater vehicle pursuit problem is a many-to-many pursuit problem.

[0084] The embodiment of the present application discloses a multi-to-multi pursuit and escape game method for unmanned underwater vehicles, where the two parties in the game are a pursuit cluster and an escape cluster. The algorithm includes three main functional modules: a collaborative detection module, a navigator module, and a pursuit and escape game module, each of which has the following advantages:

[0085] 1. Collaborative detection module: By setting up multiple UUVs for collaborative detection and positioning, the deficiency of a single UUV passive detection that cannot locate the target position is made up for, so that the two parties in the game can detect each other's position information, providing sufficient information conditions for solving the pursuit and escape game problem.

[0086] 2. Navigator module: constrains followers to always follow the navigator, maintains the integrity of the internal communication link of the cluster, and realizes information sharing within the cluster. As the navigator module of the pursuer, it can identify the easiest target to pursue in real time and only pursue the easiest target. As the navigator module of the escaper, it can identify the most dangerous pursuer in real time and only consider the pursuit of the most dangerous pursuer. This reasonably simplifies the complex many-to-many pursuit problem and significantly reduces the complexity of calculation.

[0087] 3. Pursuit and escape game module: In the pursuit and escape game process, a pair of relatively contradictory optimization objective functions are established and solved. Since the navigator module constantly changes the easiest to pursue object and the most dangerous pursuer, the specific optimization objective function is constantly changing, but the structure of the optimization objective function is simple, which makes it easy to use common optimization algorithms to quickly solve the pursuit and escape game strategy.

[0088] S120, constructing motion models of the aircraft respectively based on the positions of the aircraft.

[0089] Define the game problem of vehicle pursuit and escape in underwater three-dimensional space: the pursuing vehicles form a cluster {p1, p2, ..., p m},m≥3; the escape party's spacecraft forms a cluster {e1, e2, ..., e n}, n≥3. Each of the two clusters has a leader and the rest are followers. Without loss of generality, let the leader of the pursuer be p1 and the leader of the escaper be e1. The detection and communication ranges of the aircraft are both spherical ranges centered on the aircraft. Usually, the effective communication range is smaller than the detection range.

[0090] First, we need to build a motion model based on the relative position and speed of the pursuer and the escaper. The relative position vector x ij This is the relative position of the ith pursuer relative to the jth escaper, denoted by x ij =[(x ij ,y ij , z ij )] T , which is the position difference between the pursuer and the escaper in three-dimensional space. Relative velocity vector uij This is the relative speed of the ith pursuer relative to the jth escaper, denoted as u ij =[(u ij ,y ij , z ij )] T , which is the speed difference between the pursuer and the escaper in three-dimensional space. The acceleration ai of the pursuer represents the acceleration of the i-th pursuer in the inertial system, expressed as a i =[(a xi , a yi , a zi )] T , the acceleration aj of the escaped party represents the acceleration of the jth escaped party in the inertial system, denoted by a j =[(a xj , a yj , a zj )] T Based on the above variables, the relative motion state of the spacecraft can be expressed by the following differential equation:

[0091]

[0092]

[0093] Among them, x ij =[(x ij y ij z ij )] T is the (3×1) relative position vector of the ith pursuer relative to the jth escaper, u ij =[(u ij v ij w ij )] T is the (3×1) relative velocity vector of the ith pursuer relative to the jth escaper, a i =[(a xi a yi a zi )] T is the (3×1) acceleration vector of the ith pursuer relative to the inertial system, a j =[(a xj a yj a zj )] T is the (3×1) acceleration vector of the jth escaping party relative to the inertial system;

[0094] Next, the above motion model is converted into state space equations. This is done by combining the relative position and velocity vectors into a 6-dimensional state vector y ij The relative state vector yij Contains relative position and relative velocity, recorded as:

[0095]

[0096] This is a 6×1 vector that contains all the components of relative position and velocity.

[0097] Furthermore, according to the motion model, we further combine and write it in state space form to obtain the following state space equation:

[0098]

[0099] Among them, y ij is the relative state vector of the ith pursuer relative to the jth escaper (6×1), F is the state coefficient matrix (6×6), G is the control input coefficient matrix (6×3), a ij is the control input made by the i-th pursuer relative to the j-th escaper, a ji is the control input made by the j-th escaper to the i-th pursuer.

[0100] The state coefficient matrix F represents the structure of the system and is used to describe the relationship between relative position and velocity. The matrix is:

[0101]

[0102] The upper left part of the matrix is ​​a zero matrix, and the upper right part is a unit matrix, which represents the relationship between position and velocity. The control input coefficient matrix G is used to describe the influence of acceleration on the system state. The matrix is:

[0103]

[0104] Its structure shows that acceleration acts directly on velocity and indirectly on position. ij and a ji are the control inputs of the pursuer and the escaper, representing their respective accelerations. The relative motion of the spacecraft is controlled by inputting accelerations.

[0105] S130, based on the collaborative detection, the pursuing party locates the fleeing party and determines the target fleeing object, and the fleeing party locates the pursuing party and determines the target pursuing object.

[0106] The main function of the cooperative detection module is to cooperatively detect and locate the target position. Cooperative positioning principle: Assume that k (k ≥ 3) vehicles in one party are located at points A1 (x1, y1, z1), A2 (x2, y2, z3), ..., A k (x k ,y k , z k), the target spacecraft is detected at point B(x B ,y B , z B Passive sonar detection reveals the vector (A 1B ), (A 2B )…(A kB ) direction, then we can let the vector (A 1B ), (A 2B )…(A kB ) are (a1, b1, c1), (a2, b2, c2), …, (a k , b k , c k ). Thus we can get the straight line A iB The equation is:

[0107]

[0108] Based on the above principle, for the pursuer, since there are multiple pursuers participating in the collaborative detection, the system can solve the position of the escapee by combining the linear equations of multiple pursuers. Specifically, the system can solve the second position (x B ,y B , z B ). The system of simultaneous equations is as follows:

[0109]

[0110] Among them, (x i ,y i , z i ) is the i-th first position, i = 1, 2, ..., k, (x, y, z) is the second position of any escapee, (A i , B i , C i ) is the vector of the i-th relative direction, i = 1, 2,…, k.

[0111] The detection data of each pursuer are integrated to construct the above equations. A numerical algorithm (such as the least squares method or the iteration method) is used to solve the simultaneous equations, and finally the three-dimensional position coordinates of the escaper are obtained. For the escaper, the position of any pursuer is also calculated using the above method, and the specific calculation process will not be further described here.

[0112] In one possible implementation, based on collaborative detection, the pursuing party locates the fleeing party and determines the target fleeing object, and the fleeing party locates the pursuing party and determines the target pursuing object, specifically including: determining the aircraft distance between any pursuing party and the fleeing party based on each first position and each second position respectively; sorting multiple aircraft distances from small to large, determining the smallest aircraft distance among the multiple aircraft distances, and obtaining the target aircraft distance; determining the target pursuing object corresponding to the target aircraft distance among multiple pursuing parties, and determining the target fleeing object corresponding to the target aircraft distance among multiple fleeing parties.

[0113] Specifically, first, the current positions of multiple pursuers and escapers need to be obtained, that is, the first position of the pursuer and the second position of the escaper. For any pursuer and any escaper, the Euclidean distance between the two needs to be calculated, that is, the straight-line distance in three-dimensional space. The calculation formula for this distance is as follows:

[0114]

[0115] Among them, D ij Denotes the distance between the i-th pursuer and the j-th escaper, i = 1, 2, ..., n, j = 1, 2, ..., m. In this way, the system calculates the distance between each pursuer and each escaper to form a distance matrix D, the size of which is n × m, where n is the number of pursuers and m is the number of escapers.

[0116] After calculating the distances between all the pursuers and the escapers, the system will sort these distances. You can use a standard sorting algorithm (such as quick sort or heap sort) to sort the distances between all the vehicles in ascending order. After the sorting is completed, the system will select the smallest distance from the sorting result, that is, the first distance value after sorting. According to the calculated minimum distance, the index of the corresponding pursuer and escaper is determined, that is, the pursuer and escaper associated with the minimum distance are determined respectively, so as to obtain the target pursuer and the target escaper.

[0117] S140, based on the target pursuit object and the target escape object, establish a first optimization algorithm between the pursuit party and the target escape object, and a second optimization algorithm between the escape party and the target pursuit object.

[0118] Regardless of the pursuit cluster or the escape cluster, one of the functions of the leader module is to constrain the followers to always follow the leader, maintain the integrity of the internal communication link of the cluster, and realize information sharing within the cluster. As the pursuit party's leader module, its other function is to identify the easiest target to pursue in real time. The pursuit party may detect multiple targets of the escape party at the same time, and the pursuit party always selects the closest escape party (the target escape object, marked as e# ) as the easiest target to be pursued. Therefore, from the perspective of the pursuer, the two parties in the pursuit-escape game are the pursuer's leader and the easiest target to be pursued.

[0119] As the leader module of the escapee, its other function is to identify the most dangerous pursuer in real time. The escapee may detect multiple pursuers at the same time, and the escapee always selects the closest pursuer (the target pursuer, marked as p # ) as the most dangerous pursuer. Therefore, from the perspective of the fleeing party, the two parties in the pursuit-escape game are the most dangerous pursuer and the leader of the fleeing party.

[0120] Among the multiple aircrafts of the pursuer, a first navigator p1 is first determined, which is the control center of the optimization algorithm. The navigator is usually the aircraft closest to the escapee, and the other pursuers act in coordination through the communication system and follow the first navigator to pursue. The location information of each pursuer can be shared through the wireless network, and the optimal navigator can be determined after comprehensive analysis. # A relative motion model is established, the first relative motion model, and the first relative motion model is expressed as follows:

[0121]

[0122] Among them, y p1-e# is the relative state vector of the first navigator relative to the target escape object, F is the state coefficient matrix (6×6), G is the control input coefficient matrix (6×3), a p1-e# is the control input made by the first navigator relative to the target escape object, a e#-p1 The control input made by the target escape object relative to the first navigator.

[0123] This is the differential equation describing the relative motion between the first navigator of the pursuit party and the target escape object, expressing the rate of change of relative state. p1-e# is the relative state vector between the first navigator and the target escape object, which includes the relative position and relative velocity. This vector is usually a 6-dimensional vector, in which the first 3 dimensions represent the relative position and the last 3 dimensions represent the relative velocity. The state coefficient matrix is ​​a 6×6 constant matrix used to describe the dynamic characteristics of the system. Usually, the structure of the matrix determines the dynamic evolution relationship of the state variables, which is divided into position and velocity parts. The control input coefficient matrix is ​​a 6×3 matrix that represents the effect of acceleration on the relative state change. This formula describes how the relative position and velocity of the pursuit party navigator and the escape object change over time at time t. By adjusting a p1-e# and a e#-p1 , you can control how the pursuer approaches the fleeing party, or how the fleeing party moves away from the pursuer.

[0124] Based on the first relative motion model, a first cost function J1 of the pursuer is defined:

[0125]

[0126] Among them, J1 is the first cost function, Q p1-e# is a symmetric semi-positive definite weighting matrix, y T p1-e# is the vector transpose of the relative state vector of the first navigator relative to the target escape object, y p1-e# is the relative state vector of the first navigator relative to the target escape object.

[0127] The first cost function is used to measure the efficiency of the pursuit process. The goal of minimizing the cost function is to optimize the strategy of the pursuer to make the pursuit more effective. The transpose of the relative state vector represents the relative position and velocity of the two. The symmetric semi-positive definite weighted matrix is ​​used to assign different weights to different variables in the relative state. By designing this matrix, the importance of different states can be emphasized, such as whether to pay more attention to the optimization of position or speed. By weighting and integrating the relative state, the first cost function describes the comprehensive error of relative position and velocity during the pursuit process. The pursuer hopes to reduce the distance to the fleeing object by minimizing this error. The goal of this formula is to minimize J1, that is, the pursuer finds the best strategy to minimize the relative state error and shorten the pursuit time and distance.

[0128] Similarly, it is necessary to determine a second navigator e1 in the escape party, that is, the aircraft with the highest decision-making power in the escape party, and the determination method is similar to that of the pursuit party. Assume that the navigator has been determined to be e1 in the escape party, and the target pursuit object is p in the pursuit party. # Next, a second relative motion model between the second navigator and the target object is established. The second relative motion model is expressed as follows:

[0129]

[0130] Among them, y e1-p# is the relative state vector of the second navigator relative to the target pursuit object, F is the state coefficient matrix (6×6), G is the control input coefficient matrix (6×3), a e1-p# The control input made by the second navigator relative to the target pursuit object, a p#-e1 Control inputs made by the target pursuit object relative to the second navigator;

[0131] Based on the second relative motion model, a second cost function is determined, and the second cost function is expressed as follows:

[0132]

[0133] Among them, J2 is the second cost function, Q e1-q# is a symmetric semi-positive definite weighting matrix, y T e1-p# is the vector transpose of the relative state vector of the second navigator relative to the target pursuit object, y e1-p# is the relative state vector of the second navigator relative to the target pursuit object.

[0134] For the pursuer, the cost function J1 is minimized and the control input is adjusted to achieve the best pursuit effect. For the escaper, the cost function J2 is maximized and the control input is adjusted to achieve the best escape effect.

[0135] S150, jointly solving the first optimization algorithm and the second optimization algorithm to determine the behavior of the pursuer and the behavior of the escaper.

[0136] In a possible implementation, before jointly solving the first optimization algorithm and the second optimization algorithm to determine the behavior of the pursuer and the behavior of the escapee, the method also includes: generating multiple first control inputs for the first navigator; determining the first output result of the first cost function based on each first control input; generating multiple second control inputs for the second navigator; determining the second output result of the second cost function based on each second control input; determining a first target input among the multiple first control inputs and a second target input among the multiple second control inputs, at which time the first output result corresponding to the first target input is the smallest and the second output result corresponding to the second target input is the largest; determining the first behavior of the first navigator based on the first target input, and determining the second behavior of the second navigator based on the second target input.

[0137] Specifically, the first navigator p1 of the pursuer requires multiple different control input options, which represent different action strategies that it may take. These control inputs usually include different accelerations and directional adjustments. Similarly, the second navigator e1 of the escaper also generates multiple control inputs. These inputs are various action plans that the escaper may take in order to maximize the distance between the escaper and the pursuer. Usually, these control inputs can be obtained through certain predefined rules, algorithms or random generation methods to ensure that enough strategy space is covered for subsequent optimization.

[0138] Based on each first control input, the first output result of the first cost function is determined respectively. For each control input of the pursuer, it is substituted into the first cost function for solution to obtain its corresponding output result. As mentioned earlier, the first cost function measures the effectiveness of the pursuer's control strategy in the pursuit process. Its goal is to minimize the error of the pursuer, that is, the pursuer tries to get as close to the fleeing party as possible. Each control input corresponds to the output of a cost function. The smaller this value is, the more effective the strategy is and the better the pursuit effect is.

[0139] Similarly, based on each second control input, the second output result of the second cost function is determined respectively: for each control input of the escapee, it is also substituted into the second cost function for solution to obtain the output result. As mentioned earlier, the second cost function measures the effectiveness of the escapee's control strategy during the escape process. The escapee hopes to maximize the relative error by optimizing the control input, that is, to increase the distance from the pursuer. Each control input has a corresponding cost function output result. The larger the value, the more effective the escape strategy.

[0140] Finally, among all the generated control inputs for the pursuer, find the control input that minimizes the output of the first cost function as the first target input. This represents the optimal action strategy for the pursuer. Similarly, for the escaper, find the control input that maximizes the output of the second cost function as the second target input. This input represents the optimal escape strategy for the escaper. This step can usually be achieved through optimization algorithms, such as gradient descent, genetic algorithms, or other numerical optimization techniques. By traversing or iterating different control inputs, find the optimal solution for minimizing the error for the pursuer and maximizing the error for the escaper.

[0141] Based on the target control input of the first navigator of the pursuing party, the specific action is determined. This may be to adjust the acceleration, change the direction, or change the speed to more effectively pursue the fleeing party. For the second navigator of the fleeing party, based on its target control input, its specific action is determined to maximize the distance with the pursuing party and escape.

[0142] Furthermore, after determining the optimal control input of the pursuer, the first navigator of the pursuer converts its control input into actual action. The first navigator's behavior includes adjusting the position, speed, and acceleration to more effectively pursue the fleeing party. When updating the behavior of the pursuer, it is necessary to ensure that all pursuers maintain an effective communication distance. The pursuit system is usually composed of multiple unmanned devices, with the first navigator in charge and other pursuers as followers, and all members need to maintain the stability of the communication link. In order to ensure the overall coordination of the pursuers, the communication distance must meet the following constraints:

[0143]

[0144] Among them, (x L (t), y L (t), z L (t)) is the position of the first navigator, (x F (t), y F (t), z F (t)) is the position of the first follower, the first follower is a pursuer other than the first navigator among multiple pursuers, d L is the effective communication distance of the first navigator, d F is the effective communication distance of the first follower; or,

[0145] (x L (t), y L (t), z L (t)) is the position of the second pilot, (x F (t), y F (t), z F (t)) is the position of the second follower, the second follower is the escapee among the multiple escapees except the second leader, d L is the effective communication distance of the second pilot, d F is the effective communication distance of the second follower.

[0146] After the introduction of communication constraints, the pursuers and escapers must not only consider how to pursue or escape, but also ensure the effectiveness of communication within their respective teams. Therefore, when choosing the movement path and speed, it is necessary to comprehensively consider the optimal path and communication constraints, making the game more complex and multi-dimensional.

[0147] This formula indicates that the distance between the leader and the follower cannot exceed the smaller communication range of the two to ensure that they can maintain communication with each other. This constraint is very important in cooperative control systems because if communication fails, the cooperative strategy will not be able to be effectively executed. Similarly, the second leader of the escape party takes corresponding behavioral actions based on its optimal control input, including adjusting position, acceleration or speed, in an attempt to stay away from the pursuer. When updating the behavior of the escape party, it is also necessary to ensure that the escape parties can maintain communication. The communication distance between the escape party's leader and other escape parties also needs to meet similar constraints. The escape party's leader and follower also need to keep the distance within their respective communication ranges. By maintaining this distance, the escape party can implement an effective cooperative escape strategy, allowing the entire escape party system to act in unison and increase the chance of successful escape.

[0148] By adopting the technical solution of the present application, the position information of multiple unmanned underwater vehicles is obtained, their motion models are constructed respectively, and the pursuit party and the escape party are located using collaborative detection technology, thereby determining the target pursuit and target escape objects. Based on these objects, the first and second optimization algorithms are established respectively to describe the relative motion of the pursuit party and the escape party and their optimization goals. By jointly solving the two optimization algorithms, the optimal behavior actions of each can be determined, and then the strategic game control of the pursuit party and the escape party can be realized, ensuring that the multi-to-multi pursuit and escape game process is effectively completed in a complex underwater environment.

[0149] This embodiment also discloses a multi-to-multi pursuit and escape game device for unmanned underwater vehicles, referring to Figure 2 The device includes an acquisition module 201, a construction module 202, a positioning module 203 and a processing module 204, wherein:

[0150] The acquisition module 201 is used to acquire the positions of multiple aircraft, including the pursuing aircraft and the escaping aircraft.

[0151] The construction module 202 is used to construct the motion model of each aircraft based on the position of the aircraft.

[0152] The positioning module 203 is used to locate the fleeing party based on collaborative detection, and determine the target fleeing object. The fleeing party locates the pursuing party and determines the target pursuing object. The target fleeing object is the fleeing party that meets the preset conditions among multiple fleeing parties, and the target pursuing object is the pursuing party that meets the preset conditions among multiple pursuing parties.

[0153] The construction module 202 is used to establish a first optimization algorithm between the pursuer and the target escaping object, and a second optimization algorithm between the escaping party and the target pursuer based on the target pursuer and the target escaping object.

[0154] The processing module 204 is used to jointly solve the first optimization algorithm and the second optimization algorithm to determine the behavior of the pursuer and the behavior of the escaper.

[0155] In a possible implementation, the construction module 202 is used to construct a motion model according to the relative motion state between the pursuer and the escaper. The motion model is expressed as follows:

[0156]

[0157]

[0158] Among them, x ij =[(x ij y ij z ij )] Tis the (3×1) relative position vector of the ith pursuer relative to the jth escaper, u ij =[(u ij v ij w ij )] T is the (3×1) relative velocity vector of the ith pursuer relative to the jth escaper, a i =[(a xi a yi a zi )] T is the (3×1) acceleration vector of the ith pursuer relative to the inertial system, a j =[(a xj a yj a zj )] T is the (3×1) acceleration vector of the j-th escaping party relative to the inertial system.

[0159] The processing module 204 is used to obtain the following state space equation based on the motion model:

[0160]

[0161]

[0162]

[0163]

[0164] Among them, y ij is the relative state vector of the ith pursuer relative to the jth escaper (6×1), F is the state coefficient matrix (6×6), G is the control input coefficient matrix (6×3), a ij is the control input made by the i-th pursuer relative to the j-th escaper, a ji is the control input made by the j-th escaper to the i-th pursuer.

[0165] In a possible implementation, the processing module 204 is configured to determine a first leader among a plurality of pursuers.

[0166] The construction module 202 is used to establish a first relative motion model between the first navigator and the target escape object. The first relative motion model is expressed as follows:

[0167]

[0168] Among them, y p1-e# is the relative state vector of the first navigator relative to the target escape object, F is the state coefficient matrix (6×6), G is the control input coefficient matrix (6×3), ap1-e# is the control input made by the first navigator relative to the target escape object, a e#-p1 The control input made by the target escape object relative to the first navigator.

[0169] The processing module 204 is used to determine a first cost function based on the first relative motion model. The first cost function is expressed as follows:

[0170]

[0171] Among them, J1 is the first cost function, Q p1-e# is a symmetric semi-positive definite weighting matrix, y T p1-e# is the vector transpose of the relative state vector of the first navigator relative to the target escape object, y p1-e# is the relative state vector of the first navigator relative to the target escape object.

[0172] The processing module 204 is used to determine a second leader among the multiple escape parties.

[0173] The construction module 202 is used to establish a second relative motion model between the second navigator and the target object. The second relative motion model is expressed as follows:

[0174]

[0175] Among them, y e1-p# is the relative state vector of the second navigator relative to the target pursuit object, F is the state coefficient matrix (6×6), G is the control input coefficient matrix (6×3), a e1-p# The control input made by the second navigator relative to the target pursuit object, a p#-e1 Control inputs made by the target pursuit object relative to the second pilot.

[0176] The processing module 204 is used to determine a second cost function based on the second relative motion model. The second cost function is expressed as follows:

[0177]

[0178] Among them, J2 is the second cost function, Q e1-q# is a symmetric semi-positive definite weighting matrix, y T e1-p# is the vector transpose of the relative state vector of the second navigator relative to the target pursuit object, y e1-p# is the relative state vector of the second navigator relative to the target pursuit object.

[0179] In a possible implementation, the processing module 204 is configured to generate a plurality of first control inputs for the first navigator.

[0180] The processing module 204 is used to determine first output results of the first cost function based on each first control input.

[0181] The processing module 204 is configured to generate a plurality of second control inputs for the second navigator.

[0182] The processing module 204 is used to determine the second output results of the second cost function based on each second control input.

[0183] The processing module 204 is used to determine a first target input among the multiple first control inputs and a second target input among the multiple second control inputs, when the first output result corresponding to the first target input is the smallest and the second output result corresponding to the second target input is the largest.

[0184] The processing module 204 is configured to determine a first behavior action of the first navigator based on the first target input, and determine a second behavior action of the second navigator based on the second target input.

[0185] In a possible implementation, the processing module 204 is configured to update the behavior of the pursuer according to the first behavior, wherein when updating the behavior of the pursuer, the position of the pursuer satisfies the communication constraint condition.

[0186] The positioning module 203 is used to update the behavior of the escaped party according to the second behavior, wherein when updating the behavior of the escaped party, the location of the escaped party satisfies the communication constraint condition, and the communication constraint condition is as follows:

[0187]

[0188] Among them, (x L (t), y L (t), z L (t)) is the position of the first navigator, (x F (t), y F (t), z F (t)) is the position of the first follower, the first follower is a pursuer other than the first navigator among multiple pursuers, d L is the effective communication distance of the first navigator, d F is the effective communication distance of the first follower; or,

[0189] (x L (t), y L (t), z L (t)) is the position of the second pilot, (x F (t), y F (t), z F(t)) is the position of the second follower, the second follower is the escapee among the multiple escapees except the second leader, d L is the effective communication distance of the second pilot, d F is the effective communication distance of the second follower.

[0190] In a possible implementation, the positioning module 203 is used for multiple pursuing parties to respectively determine their own first positions.

[0191] The positioning module 203 is used for each pursuing party to determine the relative direction to any fleeing party.

[0192] The positioning module 203 is used to determine the second position of any escaped party according to the multiple first positions and the multiple relative directions, which is specifically calculated by the following formula group:

[0193]

[0194] Among them, (x i ,y i , z i ) is the i-th first position, i = 1, 2, ..., k, (x, y, z) is the second position of any escapee, (A i , B i , C i ) is the vector of the i-th relative direction, i = 1, 2,…, k.

[0195] In a possible implementation, the processing module 204 is used to determine the distance between any one of the pursuit party's aircraft and the escape party's aircraft based on each of the first positions and each of the second positions.

[0196] The processing module 204 is used to sort the multiple aircraft distances from small to large, determine the smallest aircraft distance among the multiple aircraft distances, and obtain the target aircraft distance.

[0197] The processing module 204 is used to determine the target pursuit object corresponding to the target aircraft distance among multiple pursuit parties, and determine the target escape object corresponding to the target aircraft distance among multiple escape parties.

[0198] It should be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0199] This embodiment also discloses an electronic device, referring to Figure 3 The electronic device may include: at least one processor 301 , at least one communication bus 302 , a user interface 303 , a network interface 304 , and at least one memory 305 .

[0200] The communication bus 302 is used to realize the connection and communication between these components.

[0201] The user interface 303 may include a display screen (Display) and a camera (Camera). The optional user interface 303 may also include a standard wired interface and a wireless interface.

[0202] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0203] Among them, the processor 301 may include one or more processing cores. The processor 301 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 305, and calling data stored in the memory 305. Optionally, the processor 301 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 301 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 301, and it can be implemented separately through a chip.

[0204] Among them, the memory 305 may include a random access memory (RAM) or a read-only memory (Read-Only Memory). Optionally, the memory includes a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 305 may also be at least one storage device located away from the aforementioned processor 301. The memory 305 as a computer storage medium may include an operating system, a network communication module, a user interface 303 module, and an application program for a multi-to-multi pursuit and escape game method of an unmanned underwater vehicle.

[0205] exist Figure 3 In the electronic device shown, the user interface 303 is mainly used to provide an input interface for the user and obtain data input by the user; and the processor 301 can be used to call an application program storing a multi-to-multi pursuit and escape game method of an unmanned underwater vehicle in the memory 305. When executed by one or more processors 301, the electronic device executes one or more methods in the above-mentioned embodiments.

[0206] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the order of the actions described, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0207] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0208] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0209] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0210] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0211] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory 305 and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned memory 305 includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk.

[0212] The present application also discloses a computer-readable storage medium, wherein the computer-readable storage medium stores instructions. When executed by one or more processors 301, the electronic device executes one or more of the methods described in the above embodiments.

[0213] The above is only an exemplary embodiment of the present disclosure, and the scope of the present disclosure cannot be limited thereto. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the truth of practice, it will be easy for those skilled in the art to think of other embodiments of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field that are not recorded in the present disclosure. The description and examples are regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A multi-to-multi pursuit and escape game method for unmanned underwater vehicles, characterized in that: The method comprises: Acquiring positions of a plurality of aircraft, including the pursuing aircraft and the escaping aircraft; constructing motion models of the aircraft respectively based on the positions of the aircraft; Based on the collaborative detection, the pursuer locates the fleeing party and determines the target fleeing object, the fleeing party locates the pursuer and determines the target fleeing object, the target fleeing object is the fleeing party that meets the preset conditions among the multiple fleeing parties, and the target pursuing object is the pursuer that meets the preset conditions among the multiple pursuing parties; Based on the target pursuit object and the target escape object, establishing a first optimization algorithm between the pursuit party and the target escape object, and a second optimization algorithm between the escape party and the target pursuit object; The first optimization algorithm and the second optimization algorithm are jointly solved to determine the behavior actions of the pursuer and the behavior actions of the escaper.

2. The method for many-to-many pursuit and escape game of unmanned underwater vehicles according to claim 1, characterized in that: The step of constructing the motion models of the aircraft based on the positions of the aircraft specifically includes: According to the relative motion state between the pursuer and the escaper, the motion model is constructed, and the motion model is expressed as follows: Among them, x ij =[(x ij y ij z ij )] T is the (3×1) relative position vector of the ith pursuer relative to the jth escaper, u ij =[(u ij v ij w ij )] T is the (3×1) relative velocity vector of the ith pursuer relative to the jth escaper, a i =[(a xi a yi a zi )] T is the (3×1) acceleration vector of the ith pursuer relative to the inertial system, a j =[(a xj a yj a zj )] T is the (3×1) acceleration vector of the jth escaping party relative to the inertial system; Based on the motion model, the following state space equation is obtained: Among them, y ij is the relative state vector of the ith pursuer relative to the jth escaper (6×1), F is the state coefficient matrix (6×6), G is the control input coefficient matrix (6×3), a ij is the control input made by the i-th pursuer relative to the j-th escaper, a ji is the control input made by the j-th escaper to the i-th pursuer.

3. The method for many-to-many pursuit and escape game of unmanned underwater vehicles according to claim 1, characterized in that: The first optimization algorithm between the pursuit party and the target escaped object and the second optimization algorithm between the escaped party and the target pursuit object are established based on the target pursuit object and the target escaped object, specifically including: determining a first leader among the plurality of said pursuit parties; A first relative motion model between the first navigator and the target escape object is established. The first relative motion model is expressed as follows: Among them, y p1-e# is the relative state vector of the first navigator relative to the target escape object, F is the state coefficient matrix (6×6), G is the control input coefficient matrix (6×3), a p1-e# is the control input made by the first navigator relative to the target escape object, a e#-p1 a control input made by the target escape object relative to the first navigator; Based on the first relative motion model, a first cost function is determined, where the first cost function is expressed as follows: Among them, J1 is the first cost function, Q p1-e# is a symmetric semi-positive definite weighting matrix, y T p1-e# is the vector transpose of the relative state vector of the first navigator relative to the target escape object, y p1-e# is the relative state vector of the first navigator relative to the target escape object; Identify the second leader among multiple escape parties; A second relative motion model between the second navigator and the target object is established. The second relative motion model is expressed as follows: Among them, y e1-p# is the relative state vector of the second navigator relative to the target pursuit object, F is the state coefficient matrix (6×6), G is the control input coefficient matrix (6×3), a e1-p# is the control input made by the second navigator relative to the target pursuit object, a p#-e1 a control input made by the target pursuit object relative to the second navigator; Based on the second relative motion model, a second cost function is determined, where the second cost function is expressed as follows: Among them, J2 is the second cost function, Q e1-q# is a symmetric semi-positive definite weighting matrix, y T e1-p# is the vector transpose of the relative state vector of the second navigator relative to the target pursuit object, y e1-p# is the relative state vector of the second navigator relative to the target pursuit object.

4. The method for many-to-many pursuit and escape game of unmanned underwater vehicles according to claim 3 is characterized in that: Before jointly solving the first optimization algorithm and the second optimization algorithm to determine the behavior of the pursuer and the behavior of the escaper, the method further includes: generating a plurality of first control inputs for the first navigator; Based on each of the first control inputs, respectively determine a first output result of the first cost function; generating a plurality of second control inputs for the second navigator; Based on each of the second control inputs, respectively determine a second output result of the second cost function; Determine a first target input among the plurality of first control inputs, and a second target input among the plurality of second control inputs, when a first output result corresponding to the first target input is minimum, and a second output result corresponding to the second target input is maximum; A first behavior action of the first navigator is determined based on the first target input, and a second behavior action of the second navigator is determined based on the second target input.

5. The method for many-to-many pursuit and escape game of unmanned underwater vehicles according to claim 4, characterized in that: The jointly solving the first optimization algorithm and the second optimization algorithm to determine the behavior of the pursuer and the behavior of the escaper specifically includes: According to the first behavior action, updating the behavior action of the pursuer, wherein when updating the behavior action of the pursuer, the position of the pursuer satisfies the communication constraint condition; According to the second behavior, the behavior of the escaped party is updated, wherein when the behavior of the escaped party is updated, the position of the escaped party satisfies a communication constraint condition, and the communication constraint condition is as follows: Among them, (x L (t), y L (t), z L (t)) is the position of the first navigator, (x F (t), y F (t), z F (t)) is the position of the first follower, the first follower is a pursuer other than the first navigator among the plurality of pursuers, d L is the effective communication distance of the first navigator, d F is the effective communication distance of the first follower; or, (x L (t), y L (t), z L (t)) is the position of the second navigator, (x F (t), y F (t), z F (t)) is the position of the second follower, the second follower is a runaway party among the plurality of runaway parties except the second leader, d L is the effective communication distance of the second navigator, d F is the effective communication distance of the second follower.

6. The method for many-to-many pursuit and escape game of unmanned underwater vehicles according to claim 1, characterized in that: Before the pursuing party locates the fleeing party based on the cooperative detection and determines the target fleeing object, and the fleeing party locates the pursuing party and determines the target pursuing object, the method further includes: The plurality of pursuit parties respectively determine their own first positions; Each of the pursuing parties determines a relative direction to any one of the fleeing parties; The second position of any one of the escaped parties is determined according to the plurality of the first positions and the plurality of the relative directions, specifically by calculating the following formula group: Among them, (x i ,y i , z i ) is the i-th first position, i = 1, 2, ..., k, (x, y, z) is the second position of any escapee, (A i , B i , C i ) is the vector of the i-th relative direction, i = 1, 2, …, k.

7. The method for many-to-many pursuit and escape game of unmanned underwater vehicles according to claim 6, characterized in that: Based on the collaborative detection, the pursuing party locates the fleeing party and determines the target fleeing object, and the fleeing party locates the pursuing party and determines the target pursuing object, specifically including: Determining the distance between any one of the pursuing party and the escaping party's aircraft according to each of the first positions and each of the second positions; sorting the plurality of aircraft distances from small to large, determining the smallest aircraft distance among the plurality of aircraft distances, and obtaining the target aircraft distance; The target pursuit object corresponding to the distance of the target aircraft among the plurality of the pursuing parties is determined, and the target escape object corresponding to the distance of the target aircraft among the plurality of the escaping parties is determined.

8. A multi-to-multi pursuit and escape game device for unmanned underwater vehicles, characterized in that: The device comprises an acquisition module (201), a construction module (202), a positioning module (203) and a processing module (204), wherein: The acquisition module (201) is used to acquire the positions of multiple aircraft, including the pursuing aircraft and the escaping aircraft; The construction module (202) is used to construct the motion model of the aircraft based on the position of the aircraft; The positioning module (203) is used for, based on the collaborative detection, the pursuing party to locate the fleeing party and determine the target fleeing object, the fleeing party to locate the pursuing party and determine the target fleeing object, the target fleeing object is the fleeing party that meets the preset conditions among the multiple fleeing parties, and the target fleeing object is the pursuing party that meets the preset conditions among the multiple pursuing parties; The construction module (202) is used to establish a first optimization algorithm between the pursuit party and the target escaping object, and a second optimization algorithm between the escaping party and the target pursuit object based on the target pursuit object and the target escaping object; The processing module (204) is used to jointly solve the first optimization algorithm and the second optimization algorithm to determine the behavior of the pursuer and the behavior of the escaper.

9. An electronic device, characterized in that: The electronic device comprises a processor (301), a communication bus (302), a memory (305), a user interface (303), a network interface (304) and a memory (305), wherein the memory (305) is used to store instructions, the user interface (303) and the network interface (304) are both used to communicate with other devices, and the processor (301) is used to execute the instructions stored in the memory (305) so that the electronic device executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 7 is performed.

Images