Multi-cluster unmanned ship cluster escorting method and system based on rotation scheduling

Abstract

The invention discloses a multi-cluster unmanned ship cluster escorting method and system based on rotary scheduling. The system comprises a preprocessing module, an interactive modeling module and an escorting processing module. The preprocessing module is used for preprocessing a plurality of given clusters, ensuring that the number of the clusters is an even number, and grouping every two preprocessed clusters; the interactive modeling module is used for carrying out interactive modeling on the information of the two clusters in each group to form an undirected symbol graph with a balanced structure, and establishing a directed communication graph from a navigator to the unmanned surface vehicles in the clusters; the escort processing module constructs a escort scheme based on rotation scheduling according to a preset escort distance, determines a escort vector of each group, constructs escort errors and sliding mode variables based on escort information and in-cluster formation information, and designs an escort controller for each unmanned ship according to dynamics and kinematics models of the unmanned ship. According to the invention, a plurality of clusters are ensured to uniformly carry out rotary escorting around the navigator at different escorting distances, so that an efficient and reliable solution is provided for the application of the unmanned ship cluster in the escorting task.

Description

Technical Field

[0001] The present invention relates to the field of unmanned boat swarm decision-making planning and coordinated control, and in particular to a multi-cluster unmanned boat swarm escort method and system based on round-robin scheduling. Background Art

[0002] The widespread use of unmanned boats helps maintain maritime order, ensure the smooth flow of key sea lanes, and provide solid protection for maritime security. With the increase in maritime activities and the diversification of security threats, the safety of unmanned boat swarms in maritime operations is becoming increasingly important. In this context, the escort mission of unmanned boat swarms is particularly important. Through swarm collaboration, unmanned boat swarms can efficiently perform escort missions and improve the safety and reliability of maritime operations.

[0003] In the multi-cluster unmanned boat swarm escort mission, in order to effectively protect the navigator, a reasonable escort scheme is to make multiple clusters symmetrically escort on both sides of the navigator. However, how to keep multiple clusters in pairs on both sides of the navigator by relying only on local information interaction is still a problem to be solved. To this end, the present invention proposes a multi-cluster unmanned boat swarm escort system and device based on round-robin scheduling to solve the above problem.

[0004] Existing research on the problem of unmanned boat swarm escort, such as an improved zero-space formation maneuvering control method for unmanned boat escort missions (Liao Yulei, Ge Yu, Song Zhibin, Yu Wei, Wang Bo, Li Ye, Shi Changting, Liu Haibo. An improved zero-space formation maneuvering control method for unmanned boat escort missions, CN116203958A), mainly focuses on the formation escort of a single cluster and obstacle avoidance during the formation process, without considering that the unmanned boat swarm may be presented in the form of multiple clusters in reality; or a method and system for predicting the interception point of maritime threat targets for unmanned boat swarm escort (Xiao Zhenyu, Liao Yulei, Ren Zijia, Yu Wei, Shi Changting, Liu Haibo. A method and system for predicting the interception point of maritime threat targets for unmanned boat swarm escort, CN117173934A), mainly considers the interception of threat targets during navigation. However, in reality, the unmanned boat swarm may escort the navigator in the form of clusters, and the unmanned boats transmit escort and formation information in the form of local information interaction, rather than centrally allocating and executing interception tasks. Summary of the invention

[0005] Purpose of the invention: The present invention provides a method and system for escorting multiple clusters of unmanned boats based on rotation scheduling, which ensures that multiple clusters evenly rotate around a leader at different escort distances to escort, thereby providing an efficient and reliable solution for the application of unmanned boat clusters in escort missions.

[0006] Technical solution: The present invention discloses a multi-cluster unmanned boat swarm escort method based on rotation scheduling, comprising the following steps:

[0007] Step 1: Preprocess the given multiple clusters to ensure that the number of clusters is even, and group the preprocessed clusters into pairs;

[0008] Step 2: Model the information interaction between the two clusters in each group as a structurally balanced undirected signed graph, and establish a directed communication graph from the navigator to the unmanned boats in the cluster;

[0009] Step 3: According to the preset escort distance, a round-robin-based escort scheme is constructed, and the escort vector of each group is determined;

[0010] Step 4: Construct escort error and sliding mode variables based on escort information and intra-cluster formation information;

[0011] Step 5: Design an escort controller for each unmanned boat based on the dynamics and kinematics model of the unmanned boat.

[0012] Furthermore, in step 1, preprocessing is performed on the given multiple clusters to ensure that the number of clusters is an even number, and the preprocessed clusters are grouped in pairs, which specifically includes the following steps:

[0013] Step 11: The combat effectiveness of each cluster is recorded as the sum of the combat effectiveness of all unmanned boats in the cluster. The combat effectiveness of each cluster is recorded. If the number of initial clusters is an odd number, the two clusters with the smallest combat effectiveness are merged to ensure that the number of clusters is an even number.

[0014] Step 12: The preprocessed clusters are labeled 1,...,N, where N is an even number, and the combat power of each cluster is c i , c i , i=1,...,N are sorted from large to small. According to the order of combat effectiveness from large to small, every two clusters are grouped into a group, then the number of groups is N / 2, and the i-th group is recorded as in The unmanned boat sets corresponding to the two clusters, n i is the total number of unmanned boats in the i-th group, n i1 ,n i -n i1 They are the number of unmanned boats in the two clusters within the group. Note that the smaller the index, the greater the combat effectiveness of the cluster corresponding to the group.

[0015] Furthermore, in step 2, the information interaction between the two clusters in each group is modeled as a structurally balanced undirected signed graph, and a directed communication graph from the navigator to the unmanned boats in the cluster is established, which specifically includes the following steps:

[0016] Step 21: For the i-th group The information interaction of all unmanned boats in the group is modeled as an undirected connected graph, using the adjacency matrix Describe the information interaction of unmanned boats within the group. and unmanned boats communicate with each other and are in the same cluster, then If no boat and unmanned boats communicate with each other, but in different clusters, then otherwise, This modeling method ensures that the information interaction graph between the two clusters in each group is a structurally balanced undirected signed graph;

[0017] Step 22: For the i-th group Establish a directed communication graph from the navigator to the unmanned boats in the cluster to ensure that there are two unmanned boats in the group. These two unmanned boats come from different clusters and communicate with each other. In addition, these two unmanned boats can receive the information from the navigator. If the unmanned boat If you receive a message from the navigator, otherwise

[0018] Furthermore, in step 3, according to the preset escort distance, a round-robin-based escort scheme is constructed, and the escort vector of each group is determined, which specifically includes the following steps:

[0019] Step 31: The navigator sets the escort distances ρ1 and ρ2, where ρ1<ρ2. Draw two circles with the navigator as the center and ρ1 and ρ2 as the radius. The groups are allocated on the inner circle, that is, the clusters with greater combat effectiveness are allocated on the inner circle, where the symbol Indicates rounding down; the remaining groups are allocated on the outer circle, that is, the clusters with poor combat effectiveness are allocated on the outer circle to play the role of detecting threats;

[0020] Step 32: To determine the escort vector of each formation, let N2=N / 2-N1, calculate the angle The escort scheme based on round-robin scheduling is designed as follows:

[0021]

[0022] where χ i is the escort vector of the i-th group, i = 1, ..., N / 2, θ2, θ1 are the angular velocities of the rotation of the unmanned boats on the inner and outer circles. Based on this escort scheme, it is ensured that the clusters on the inner and outer circles rotate around the leader evenly for escort.

[0023] Furthermore, in step 4, based on the escort information and the intra-cluster formation information, constructing the escort error and sliding mode variable specifically includes the following steps:

[0024] Step 41: For the i-th group definition And define Assume that there are no unmanned boats in this formation The position and heading angle vector are The corresponding speed is Right now Unmanned boat The escort error is defined as

[0025]

[0026] in Describes the time-varying formation configuration of the unmanned boats in the corresponding cluster. is corresponding to The time derivative of , η0 is the position and heading angle vector of the navigator, ζ0 is the velocity vector corresponding to the navigator, that is

[0027] Step 42: construct auxiliary escort and formation error variables according to the escort information and formation information. The specific construction method is as follows:

[0028]

[0029] in It's an unmanned boat In the collection V l i The communication neighbor set in , l = 1, 2, in particular, if Then l = 1, otherwise, l = 2;

[0030] Step 43: Construct escort sliding mode variables

[0031]

[0032] Furthermore, in step 5, according to the dynamics and kinematics model of the unmanned boat, designing an escort controller for each unmanned boat specifically includes the following steps:

[0033] Step 51: Unmanned Boat The dynamic model of is expressed in the geodetic coordinate system as:

[0034]

[0035] in Unmanned Boat location, represents the heading angle, Unmanned Boat The velocity vector in the geodetic coordinate system, is the rotation matrix, is the time derivative of the rotation matrix, M ij is the inertia matrix of the unmanned boat, is the system torque, is the control input;

[0036] Step 52: Unmanned Boat The escort controller is designed as follows:

[0037]

[0038] where α is a sufficiently large control gain, i = 1,...,N / 2, j = 1,..,n i .

[0039] Accordingly, a multi-cluster unmanned boat swarm escort system based on round-robin scheduling includes: a preprocessing module, an interactive modeling module and an escort processing module; the preprocessing module preprocesses a given plurality of clusters to ensure that the number of clusters is an even number, and groups the preprocessed clusters in pairs; the interactive modeling module models the information interaction between the two clusters in each grouping as a structurally balanced undirected symbolic graph, and establishes a directed communication graph from the navigator to the unmanned boats in the cluster; the escort processing module constructs an escort scheme based on round-robin scheduling according to a preset escort distance, and determines the escort vector of each group, constructs escort errors and sliding mode variables based on escort information and formation information within the cluster, and designs an escort controller for each unmanned boat according to the dynamic and kinematic models of the unmanned boat.

[0040] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages: (1) The present invention utilizes a structurally balanced symbolic graph to model the information interaction between unmanned boats, thereby realizing that the clusters formed in pairs can run symmetrically on both sides of the leader, thereby achieving the purpose of escort; (2) The present invention takes into account the clustered escort of the cluster and designs a rotating escort scheme, so that multiple clusters can evenly rotate around the leader at different escort distances for escort, thereby achieving multi-directional escort for the leader. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a schematic diagram of the method flow of the present invention.

[0042] Figure 2 It is a schematic diagram of the escort scheme based on round-robin scheduling of the present invention.

[0043] Figure 3 An information interaction diagram modeling the symbolic diagram of the present invention.

[0044] Figure 4 This is a schematic diagram of the motion trajectory of an unmanned boat generated by the multi-cluster unmanned boat swarm escort controller of the present invention.

[0045] Figure 5A schematic diagram of the escort position error generated by the multi-cluster unmanned boat swarm escort controller of the present invention.

[0046] Figure 6 A schematic diagram of the escort speed error generated by the multi-cluster unmanned boat swarm escort controller of the present invention. DETAILED DESCRIPTION

[0047] like Figure 1 As shown, a multi-cluster unmanned boat swarm escort method based on rotation scheduling includes the following steps:

[0048] Step 1: Preprocess the given multiple clusters to ensure that the number of clusters is even, and group the preprocessed clusters into pairs;

[0049] Step 2: Model the information interaction between the two clusters in each group as a structurally balanced undirected signed graph, and establish a directed communication graph from the navigator to the unmanned boats in the cluster;

[0050] Step 3: According to the preset escort distance, a round-robin-based escort scheme is constructed, and the escort vector of each group is determined;

[0051] Step 4: Construct escort error and sliding mode variables based on escort information and intra-cluster formation information;

[0052] Step 5: Design an escort controller for each unmanned boat based on the dynamics and kinematics model of the unmanned boat.

[0053] Furthermore, in step 1, in order to escort the target, the number of clusters must be an even number, so when the number of initial clusters is an odd number, the clusters need to be preprocessed. The details are as follows:

[0054] Step 11: The combat effectiveness of each cluster is recorded as the sum of the combat effectiveness of all unmanned boats in the cluster. The combat effectiveness of each cluster is recorded; if the number of initial clusters is an odd number, the two clusters with the smallest combat effectiveness are merged to ensure that the number of clusters is an even number.

[0055] Step 12: The preprocessed clusters are labeled 1,...,N, where N is an even number, and the combat power of each cluster is c i , c i , i=1,...,N are sorted from large to small, and every two clusters are grouped in the order of combat effectiveness from large to small, so the number of groups is N / 2. The i-th group is recorded as in The unmanned boat sets corresponding to the two clusters, n i is the total number of unmanned boats in the i-th group, n i1 ,n i -n i1are the number of unmanned boats in the two clusters in the group. Note that the smaller the index, the greater the combat power of the cluster corresponding to the group.

[0056] In step 2, the information interaction between the two clusters in each group is modeled as a structurally balanced undirected signed graph, and a directed communication graph from the navigator to the unmanned boats in the cluster is established. The modeling process is as follows:

[0057] Step 21: For the i-th group The information interaction of all unmanned boats in the group is modeled as an undirected connected graph, using the adjacency matrix Describe the information interaction of unmanned boats within the group. and unmanned boats can communicate with each other and are in the same cluster, then If no boat and unmanned boats can communicate with each other, but in different clusters. otherwise, This modeling method can ensure that the information interaction graph between the two clusters in each group is a structurally balanced undirected signed graph.

[0058] Step 22: For the i-th group Establish a directed communication graph from the navigator to the unmanned boats in the cluster to ensure that there are two unmanned boats in the group. These two unmanned boats come from different clusters and can communicate with each other. In addition, these two unmanned boats can receive the information from the navigator. l i If you can receive the information sent by the navigator, otherwise

[0059] In step 3, according to the preset escort distance, a round-robin-based escort scheme is constructed, and the escort vector of each group is determined. The escort scheme and escort vector are designed as follows:

[0060] Step 31: The navigator sets the escort distances ρ1 and ρ2, where ρ1<ρ2. Draw two circles with the navigator as the center and ρ1 and ρ2 as the radius. The groups are allocated on the inner circle, that is, the clusters with greater combat effectiveness are allocated on the inner circle, where the symbol It means rounding down; the remaining groups are allocated on the outer circle, that is, the clusters with poor combat effectiveness are allocated on the outer circle to play the role of detecting threats.

[0061] Step 32: To determine the escort vector of each formation, let N2=N / 2-N1, calculate the angle The escort scheme based on round-robin scheduling is designed as follows:

[0062]

[0063] where χ i is the escort vector of the i-th group, i=1,...,N / 2, θ2,θ1 are the angular velocities of the unmanned boats on the inner and outer circles. The schematic diagram of the escort scheme based on rotation scheduling is as follows Figure 2 Based on this escort scheme, it can be ensured that the clusters on the inner and outer circles evenly rotate around the leader for escort.

[0064] In step 4, based on the escort information and the intra-cluster formation information, the escort error and sliding mode variables are constructed. The specific construction is as follows:

[0065] Step 41: For the i-th group definition And define Assume that there are no unmanned boats in this formation The position and heading angle vector are The corresponding speed is Right now Unmanned boat The escort error is defined as

[0066]

[0067] in Describes the time-varying formation configuration of the unmanned boats in the corresponding cluster. is corresponding to The time derivative of , η0 is the position and heading angle vector of the navigator, ζ0 is the velocity vector corresponding to the navigator, that is

[0068] Step 42: construct auxiliary escort and formation error variables according to the escort information and formation information. The specific construction method is as follows:

[0069]

[0070] in It's an unmanned boat In the collection V l i The communication neighbors set in , l = 1,2. In particular, if Then l=1, otherwise, l=2.

[0071] Step 43: Construct escort sliding mode variables

[0072]

[0073] In step 5, an escort controller is designed for each unmanned boat based on the dynamics and kinematics model of the unmanned boat. The dynamics and kinematics model analysis and controller design of the unmanned boat are as follows:

[0074] Step 51: Unmanned Boat The dynamic model of is expressed in the geodetic coordinate system as:

[0075]

[0076] in Unmanned Boat location, represents the heading angle, Unmanned Boat The velocity vector in the geodetic coordinate system, is the rotation matrix, is the time derivative of the rotation matrix, M ij is the inertia matrix of the unmanned boat, is the system torque, is the control input;

[0077] Step 52: Unmanned Boat The escort controller is designed as follows:

[0078]

[0079] where α is a sufficiently large control gain, i = 1,...,N / 2, j = 1,..,n i .

[0080] Accordingly, a multi-cluster unmanned boat swarm escort system based on round-robin scheduling includes: a preprocessing module, an interactive modeling module and an escort processing module; the preprocessing module preprocesses a given plurality of clusters to ensure that the number of clusters is an even number, and groups the preprocessed clusters in pairs; the interactive modeling module models the information interaction between the two clusters in each grouping as a structurally balanced undirected symbolic graph, and establishes a directed communication graph from the navigator to the unmanned boats in the cluster; the escort processing module constructs an escort scheme based on round-robin scheduling according to a preset escort distance, and determines the escort vector of each group, constructs escort errors and sliding mode variables based on escort information and formation information within the cluster, and designs an escort controller for each unmanned boat according to the dynamic and kinematic models of the unmanned boat.

[0081] In this embodiment, MATLAB 2022a is used as the simulation calculation software to simulate the multi-cluster unmanned boat swarm escort scenario, and a multi-cluster unmanned boat swarm escort system and device based on round-robin scheduling of the present invention is used for simulation.

[0082] In this embodiment, there are 4 clusters, each of which has 5 unmanned boats, and the first cluster and the second cluster are grouped as The escort distance is set to ρ1 = 30, and the third and fourth clusters are grouped as The escort distance is set to ρ2 = 50, and the escort vectors of the two formations are

[0083]

[0084] The initial state of the leader is η0(0) = [0,0,0] T , the dynamics of the navigator is Where υ0=[1,2,-0.01] T The formation vector of the unmanned boat in the group is

[0085]

[0086]

[0087] Where a=4, b=0.01.

[0088] Figure 3 The information interaction diagram between the navigator and the unmanned boats in the two formations is given.

[0089] Figure 4 The schematic diagram of the unmanned boat motion trajectory generated by the designed multi-cluster unmanned boat swarm escort controller is given. Figure 4 It can be seen that under the designed multi-cluster unmanned boat swarm escort controller, the navigator is at the center of all clusters, effectively achieving the escort goal.

[0090] Figure 5 A schematic diagram of the escort position error generated by the designed multi-cluster unmanned boat swarm escort controller is given. It can be seen that the escort position error eventually tends to zero.

[0091] Figure 6 A schematic diagram of the escort speed error generated by the designed multi-cluster unmanned boat swarm escort controller is given. It can be seen that the escort speed error eventually tends to zero.

[0092] It can be clearly seen from the above simulation results that the multi-cluster unmanned boat swarm escort system and device based on round-robin scheduling proposed in the present invention can achieve effective escort of multiple clusters of unmanned boats.

[0093] To sum up, the multi-cluster unmanned boat swarm escort system and device based on rotation scheduling of the present invention can ensure that multiple clusters evenly rotate around the leader for escort at different escort distances, thereby providing an efficient and reliable solution for the application of unmanned boat swarms in escort missions.

Claims

1. A multi-cluster unmanned boat swarm escort method based on rotation scheduling, characterized in that: The steps include: Step 1: Preprocess the given multiple clusters to ensure that the number of clusters is even, and group the preprocessed clusters into pairs; Step 2: Model the information interaction between the two clusters in each group as a structurally balanced undirected signed graph, and establish a directed communication graph from the navigator to the unmanned boats in the cluster; Step 3: According to the preset escort distance, a round-robin-based escort scheme is constructed, and the escort vector of each group is determined; Step 4: Construct escort error and sliding mode variables based on escort information and intra-cluster formation information; Step 5: Design an escort controller for each unmanned boat based on the dynamics and kinematics model of the unmanned boat.

2. The multi-cluster unmanned boat swarm escort method based on rotation scheduling as claimed in claim 1 is characterized in that: In step 1, preprocessing is performed on the given multiple clusters to ensure that the number of clusters is an even number, and the preprocessed clusters are grouped in pairs. Specifically, the steps include: Step 11: The combat effectiveness of each cluster is recorded as the sum of the combat effectiveness of all unmanned boats in the cluster. The combat effectiveness of each cluster is recorded. If the number of initial clusters is an odd number, the two clusters with the smallest combat effectiveness are merged to ensure that the number of clusters is an even number. Step 12: The preprocessed clusters are labeled 1,...,N, where N is an even number, and the combat effectiveness of each cluster is c i , c i , i=1,...,N are sorted from large to small. According to the order of combat effectiveness from large to small, every two clusters are grouped into a group, then the number of groups is N / 2, and the i-th group is recorded as in The unmanned boat sets corresponding to the two clusters, n i is the total number of unmanned boats in the i-th formation, n i1 ,n i -n i1 They are the number of unmanned boats in the two clusters within the group. Note that the smaller the index, the greater the combat effectiveness of the cluster corresponding to the group.

3. The multi-cluster unmanned boat swarm escort method based on rotation scheduling as claimed in claim 1 is characterized in that: In step 2, the information interaction between the two clusters in each group is modeled as a structurally balanced undirected signed graph, and a directed communication graph from the navigator to the unmanned boats in the cluster is established, which specifically includes the following steps: Step 21: For the i-th group The information interaction of all unmanned boats in the group is modeled as an undirected connected graph, using the adjacency matrix Describe the information interaction of unmanned boats within the group. and unmanned boats communicate with each other and are in the same cluster, then If no boat and unmanned boats communicate with each other, but in different clusters, then otherwise, This modeling method ensures that the information interaction graph between the two clusters in each group is a structurally balanced undirected signed graph; Step 22: For the i-th group Establish a directed communication graph from the navigator to the unmanned boats in the cluster to ensure that there are two unmanned boats in the group. These two unmanned boats come from different clusters and communicate with each other. In addition, these two unmanned boats can receive the information from the navigator. If the unmanned boat If you receive a message from the navigator, otherwise 4. The multi-cluster unmanned boat swarm escort method based on rotation scheduling as claimed in claim 1 is characterized in that: In step 3, according to the preset escort distance, a round-robin-based escort scheme is constructed, and the escort vector of each group is determined. Specifically, the steps include: Step 31: The navigator sets the escort distances ρ1 and ρ2, where ρ1<ρ2. Draw two circles with the navigator as the center and ρ1 and ρ2 as the radius. The groups are allocated on the inner circle, that is, the clusters with greater combat effectiveness are allocated on the inner circle, where the symbol Indicates rounding down; the remaining groups are allocated on the outer circle, that is, the clusters with poor combat effectiveness are allocated on the outer circle to play the role of detecting threats; Step 32: To determine the escort vector of each formation, let N2=N / 2-N1, calculate the angle The escort scheme based on round-robin scheduling is designed as follows: where χ i is the escort vector of the i-th group, i=1,...,N / 2, is the angular velocity of the rotation of the unmanned boats on the inner and outer circles. Based on this escort scheme, it is ensured that the clusters on the inner and outer circles rotate around the leader evenly for escort.

5. The multi-cluster unmanned boat swarm escort method based on rotation scheduling as claimed in claim 1 is characterized in that: In step 4, based on the escort information and the intra-cluster formation information, the escort error and sliding mode variable are constructed, which specifically includes the following steps: Step 41: For the i-th group definition And define Assume that there are no unmanned boats in this formation The position and heading angle vector are The corresponding speed is Right now Unmanned boat The escort error is defined as in Describes the time-varying formation configuration of the unmanned boats in the corresponding cluster. is corresponding to The time derivative of , η0 is the position and heading angle vector of the navigator, ζ0 is the velocity vector corresponding to the navigator, that is Step 42: construct auxiliary escort and formation error variables according to the escort information and formation information. The specific construction method is as follows: in It's an unmanned boat In the collection The communication neighbor set in , l = 1,2, if Then l = 1, otherwise, l = 2; Step 43: Construct escort sliding mode variables 6. The method for swarm escort of multiple unmanned boats based on rotation scheduling as claimed in claim 1, characterized in that: In step 5, according to the dynamic and kinematic models of the unmanned boat, the escort controller is designed for each unmanned boat, which specifically includes the following steps: Step 51: Unmanned Boat The dynamic model of is expressed in the geodetic coordinate system as: in Unmanned Boat location, represents the heading angle, Unmanned Boat The velocity vector in the geodetic coordinate system, is the rotation matrix, is the time derivative of the rotation matrix, M ij is the inertia matrix of the unmanned boat, is the system torque, is the control input; Step 52: Unmanned Boat The escort controller is designed as follows: where α is a sufficiently large control gain, i = 1,...,N / 2, j = 1,..,n i .

7. A system for implementing the multi-cluster unmanned boat swarm escort method based on rotation scheduling as claimed in claim 1, characterized in that: include: Preprocessing module, interactive modeling module and escort processing module; The preprocessing module preprocesses the given multiple clusters to ensure that the number of clusters is even, and groups the preprocessed clusters into pairs; The interactive modeling module models the information interaction between the two clusters in each formation as a structurally balanced undirected symbolic graph, and establishes a directed communication graph from the leader to the unmanned boats in the cluster; the escort processing module constructs an escort scheme based on round-robin scheduling according to the preset escort distance, and determines the escort vector of each group. Based on the escort information and the formation information within the cluster, the escort error and sliding mode variables are constructed, and an escort controller is designed for each unmanned boat according to the dynamic and kinematic models of the unmanned boat.

Images