Multi-USV hunting method based on multi-factor scoring and adaptive task allocation

Abstract

The invention relates to the technical field of unmanned ship hunting, and discloses a multi-USV hunting method based on multi-factor scoring and adaptive task assignment, comprising the following steps: (1) establishing USV mathematical models including a kinematics model and a dynamics model; (2) establishing a multi-USV hunting task model; (3) establishing a multi-factor scoring rule: obtaining a distance factor score, an angle factor score and an energy consumption factor score, and obtaining a comprehensive score between the surrounding USV and the confronting USV according to each factor score; and (4) determining an adaptive task allocation strategy according to the comprehensive score between the hunting USV and the confrontation USV, and realizing multi-USV hunting. According to the method, the multiple confronting USVs are trapped to the designated target area to be trapped, the trapped tasks are distributed to each trapped USV in a self-adaptive mode, it is ensured that each confronting USV can be trapped by the trapped USV at the same time, the trapped USV can be trapped at the same time, trapped efficiency is improved, energy consumption needed for trapped USV is lower, and trapped voyage is shorter.

Description

Technical Field

[0001] The present invention relates to a multi-USV (Unmanned Surface Vehicle) encirclement method based on multi-factor scoring and adaptive task allocation, belonging to the technical field of USV encirclement. Background Art

[0002] Unmanned Surface Vehicles (USVs) have strong mobility, low operating costs, and the ability to perform tasks in complex environments. In various applications, the encirclement tasks of multi-USVs have received extensive attention due to their unique status in military and maritime security. Efficiently completing the encirclement tasks of USVs is an important guarantee for maritime security.

[0003] A large number of studies on the encirclement tasks of USVs have been carried out. Common encirclement methods include heuristic methods, deep reinforcement learning algorithms, distributed encirclement methods, etc. However, there are still some problems to be solved in these encirclement methods: 1. Most of the current encirclement methods mainly focus on distributed encirclement methods, and the encirclement methods are relatively single; 2. Mainly consider the encirclement performance of the encirclement method, but relatively few studies on the energy consumption generated during the encirclement process, and do not fully consider the energy-saving requirements of USVs in marine interactions; 3. The main verification environment is carried out under static obstacle conditions, and there is less verification of complex environments with dynamic obstacles; 4. For the formation of both sides of the encirclement task, mainly the majority of the encircling parties confront a single escaping party, and there is relatively little encirclement of multiple escaping parties. Summary of the Invention

[0004] Aiming at the deficiencies of the existing USV encirclement technology, the present invention proposes a multi-USV encirclement method based on multi-factor scoring and adaptive task allocation to achieve the goals of low energy consumption, high efficiency, and short voyage. This method comprehensively considers factors such as distance, energy consumption, and steering angle, and besieges multiple escaping USVs to a specified target area for encirclement.

[0005] The multi-USV encirclement method based on multi-factor scoring and adaptive task allocation of the present invention includes the following steps: (1) Establish a mathematical model of USV (Unmanned Surface Vehicle), including a kinematic model and a dynamic model; (2) Establish a multi-USV encirclement task model; (3) Establish a multi-factor scoring rule: In the encirclement task of the USV, the influencing factors in four aspects, namely the distance between the encircling USV and the adversarial USV, the angle to be turned, the distance between the adversarial USV and the target area, and the energy consumption required for the encircling USV to execute the task, are combined to form a four-factor scoring strategy, obtaining a distance factor score, an angle factor score, and an energy consumption factor score. Based on the scores of each factor, a comprehensive score between the encircling USV and the adversarial USV is obtained; (4) Determine an adaptive task allocation strategy according to the comprehensive score between the encircling USV and the adversarial USV to achieve multi-USV encirclement.

[0006] The establishment process of the kinematic model in step (1) is as follows: When the unmanned ship moves on the sea surface, it has six degrees of freedom of motion. Two coordinate systems are established: the earth coordinate system fixed on the earth and the appendage coordinate system fixed on the hull ; The six-degree-of-freedom USV model is simplified to three degrees of freedom. The kinematic model of the three degrees of freedom is as follows: ; In the above formula, represents the position information of the USV, represents the first derivative of the position information, represents the heading angle of the USV, represents the angular velocity, represents the forward speed of the USV, represents the lateral movement speed of the USV, represents the angular velocity of the USV's turning.

[0007] The dynamic model in step (1) is as follows: ; In the above formula, , , represent the mass coefficients composed of the mass of the USV, , , represent the hydrodynamic damping coefficients, represents the driving force in the forward direction of the USV, represents the driving force for the USV to turn, represents the forward speed of the USV, represents the lateral movement speed of the USV, represents the angular velocity of the USV's turning, represents the first derivative of the forward speed of the USV, represents the first derivative of the lateral movement speed of the USV, Represents the first derivative of the USV steering angular velocity.

[0008] The process of establishing the multi-USV pursuit mission model in step (2) is as follows: The pursuing USV represents the pursuer in the pursuit mission, and the opposing USV represents the escaping party in the pursuit mission. Represents the number of pursuing USVs. Represents the number of opposing USVs. ; When the pursuing USV detects the opposing USV, the pursuing USV starts from the starting point position and points in the direction of the opposing USV. When it reaches a position within the range of the opposing USV, it moves along a circular trajectory centered on the opposing USV with a radius of to the driving position . The driving point position is on the side of the line connecting the target area and the opposing USV that is far from the target area, that is, the opposing USV is in the middle of the target area and the pursuing USV, so as to surround the opposing USV into the target area and achieve the pursuit mission; When the opposing USV is surrounded and enters the target area with a radius of , it means the pursuit is successful.

[0009] To prevent collisions between USVs during movement, it should be ensured that , represents the length of the USV.

[0010] The process of obtaining the distance factor score in step (3) is: Define the distance factor score as: ; ; represents the distance score between the th pursuing USV and the j th opposing USV. represents the distance between the th pursuing USV and the th opposing USV. represents the distance score between the th opposing USV and the target area. represents the distance between the th opposing USV and the target area. The closer the distance between the pursuing USV and the opposing USV, and the closer the distance between the opposing USV and the target area, the higher the score.

[0011] The process of obtaining the angle factor score in step (3) is; Define the angular factor score as: ; represents the angular score between the forward direction of the th encircling USV and the th counter USV; represents the angle between the forward direction of the th encircling USV and the th counter USV; The smaller the angle between the forward direction of the encircling USV and the counter USV, that is, the higher the approach degree, the higher the score.

[0012] The process of obtaining the energy consumption factor score in step (3) is as follows: When the USV moves in the marine environment, the energy consumed includes: the forward energy consumption term (energy consumption in the forward direction of the USV), the turning energy consumption term (energy consumption for turning), and the ocean current influence term (energy consumption required for the USV to interact with the ocean current). Therefore, the calculation method of energy consumption is the sum of these three energy consumption forms: ; The forward energy consumption term The calculation method is as follows: ; represents the forward thrust of the USV (in the surge direction), represents the forward speed of the USV, represents the movement time; The turning energy consumption term The calculation method is as follows: ; represents the control torque in the yaw direction, represents the angular velocity, represents the movement time; The ocean current influence term The calculation method is as follows: ; represents the ocean current force; represents the velocity component of the forward speed direction of the USV in the direction perpendicular to the ocean current (i.e., the USV has assistance when flowing with the ocean current and increases consumption when flowing against or laterally to the ocean current); represents the velocity component of the forward speed direction of the USV in the direction of the ocean current; represents the movement time; Finally, the total energy consumption weight formula is obtained: ; represents the th encircling USV's energy consumption required to encircle the th counter USV; represents the moving time; Therefore, the energy consumption factor score is defined as: ; represents the th encircling USV's energy consumption score for encircling the th counter USV, represents the th encircling USV's energy consumption required to encircle the th counter USV; the lower the required energy consumption, the higher the score.

[0013] The comprehensive score formula between the encircling USV and the counter USV in step (3) is as follows: ; represents the th encircling USV's comprehensive score for encircling the th counter USV. The higher the score, the higher the priority allocation degree, that is, the closer the distance between the encircling USV and the counter USV, the closer the distance between the counter USV and the target area, the higher the orientation proximity, and the lower the energy consumption required for the encircling USV, the higher the comprehensive score; Among them, represents the th encircling USV and the th counter USV's distance score, represents the th counter USV and the target area's distance score, represents the th encircling USV's forward direction and the th counter USV's angle score, represents the th encircling USV's energy consumption score for encircling the th counter USV; represents the scoring factor coefficient, satisfying , and the encircling relationship between the encircling USV and the counter USV is determined through the four-factor comprehensive scoring rule.

[0014] The adaptive task allocation strategy in step (4) is; For the multi-factor scoring between each encircling USV and each opposing USV, the scoring situation is represented by the following formula: ; In the formula, represents the number of encircling USVs, represents the number of opposing USVs, , the multi-factor score of the th encircling USV is , , and this encircling USV selects the th opposing USV with the highest score to perform the encircling task.

[0015] The method of the present invention is an adaptive task allocation strategy based on multi-factor scoring rules (MFSR-ATAS). It comprehensively considers the influence of four aspects: the distance factor, the orientation factor, the energy consumption factor between each encircling USV and each opposing USV, and the distance factor between the opposing USV and the target area. It traps multiple escaping USVs to a specified target area for encirclement, adaptively assigns encircling tasks to each encircling USV, ensures that each opposing USV can be simultaneously encircled by the encircling USVs, greatly improves the encircling efficiency; it performs well in the encircling tasks of multiple USVs and is an effective method with low energy consumption, high efficiency, and short voyage.

[0016] Through experimental verification, the method of the present invention can efficiently achieve the encircling tasks of multiple USVs. It shortens the encircling time by more than 40% compared with the Hungarian algorithm and the greedy algorithm. The required scoring energy consumption is 5 kJ to 10 kJ, and the energy consumption is reduced by about 70%. The encircling USVs require lower energy consumption and shorter encircling voyages, greatly improving the encircling efficiency. It greatly reduces the energy loss during the task process and fully demonstrates the significant advantage in terms of energy conservation. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a schematic diagram of the kinematic model.

[0018] Figure 2 is a schematic diagram of the dynamic model.

[0019] Figure 3 is a schematic diagram of the multi-USV encircling task model.

[0020] Figure 4 is a schematic diagram of the multi-factor scoring rules.

[0021] Figure 5 is a schematic diagram of the energy consumption factor.

[0022] Figure 6 is a schematic diagram of the task allocation.

[0023] Figure 7It is the trajectory diagram of two pursuit USVs chasing two adversarial USVs.

[0024] Figure 8 It is the graph of the pursuit time results of the comparison experiments of different algorithms.

[0025] Figure 9 It is the graph of the energy consumption results of the comparison experiments of different algorithms. Specific implementation mode

[0026] Aiming at the pursuit problem of multiple USVs, the present invention proposes a multi-USV pursuit method based on multi-factor scoring and adaptive task allocation. This method is an Adaptive Task Allocation Strategy Based on Multi-Factor Scoring Rules (MFSR-ATAS), which comprehensively considers factors such as distance, energy consumption, and steering angle, and traps multiple escaping USVs into a specified target area for pursuit. The specific process is described as follows.

[0027] I. Mathematical model of USV.

[0028] 1. Kinematic model of USV; When a USV moves on the sea surface, it usually has six degrees of freedom of motion. Therefore, generally two coordinate systems are established: the earth coordinate system fixed on the earth and the appendage coordinate system fixed on the hull , and the kinematic model of the USV is as Figure 1 shown ， The present invention simplifies the six-degree-of-freedom USV model to three degrees of freedom, and the kinematic model of the three degrees of freedom is shown in Equation (1).

[0029] ;(1) In Equation (1), represents the position information of the USV, represents the first derivative of the position information, represents the heading angle of the USV, represents the angular velocity, represents the forward speed of the USV, represents the lateral movement speed of the USV, represents the angular velocity of the USV's steering.

[0030] 2. Dynamic model of USV; The dynamic schematic diagram of the USV is as Figure 2 shown, and the dynamic model of the USV can be obtained as Equation (2).

[0031] ;(2) In Equation (2), , , represent the mass coefficients composed of the USV mass, , , represent the hydrodynamic damping coefficients, represents the driving force in the forward direction of the USV, represents the driving force for the USV to turn, represents the forward speed of the USV, represents the lateral movement speed of the USV, represents the angular velocity of the USV's turning, represents the first derivative of the forward speed of the USV, represents the first derivative of the lateral movement speed of the USV, represents the first derivative of the angular velocity of the USV's turning.

[0032] II. Multi-USV Encirclement Mission Model.

[0033] In the present invention, the key research content is the encirclement mission of multiple USVs. In order to encircle multiple adversarial USVs to the target area, an encirclement model as shown in Figure 3 is designed.

[0034] In Figure 3 , the encircling USV represents the encircling party of the encirclement mission, and the adversarial USV represents the escaping party of the encirclement mission. represents the number of encircling USVs, represents the number of adversarial USVs. In order to efficiently complete the encirclement mission, generally is taken; when the encircling USV detects the adversarial USV, the encircling USV starts from the starting point position and points in the direction of the adversarial USV. When it reaches the position within the range of from the adversarial USV , it moves along a circular trajectory with the adversarial USV as the center and as the radius to the driving position . The driving point position is on the side away from the target area on the line connecting the target area and the adversarial USV, that is, the adversarial USV is in the middle of the target area and the encircling USV, thereby encircling the adversarial USV to the target area and achieving the encirclement mission. Among them, represents the radius of the target area. When the adversarial USV is encircled and enters this area, it represents the success of the encirclement. For the selection of , in order to prevent the USVs from colliding during the movement, it should be ensured that , represents the length of the USV.

[0035] III. Adaptive task allocation strategy based on multi-factor scoring rules.

[0036] In the encirclement task of the USV, the present invention comprehensively considers the distance between the encircling USV and the opposing USV , the angle that needs to be turned , the distance between the opposing USV and the target area and the energy consumption required for the encircling USV to execute the task The influencing factors in these four aspects form a four-factor scoring strategy, as Figure 4 shown.

[0037] 1. Distance factor scoring; Define the distance factor score as: ; (3) ; (4) represents the distance score between the th encircling USV and the th opposing USV; represents the distance between the th encircling USV and the th opposing USV; represents the distance score between the th opposing USV and the target area; represents the distance between the th opposing USV and the target area.

[0038] According to the definition of the distance factor, it can be seen that: the closer the distance between the encircling USV and the opposing USV, and the closer the distance between the opposing USV and the target area, the higher the score.

[0039] 2. Angle factor scoring; Define the angle factor score as: ; (5) represents the angle score between the forward direction of the th encircling USV and the th opposing USV; represents the angle between the forward direction of the th encircling USV and the th opposing USV.

[0040] According to the definition of the angle factor, it can be seen that: the smaller the angle between the forward direction of the encircling USV and the opposing USV, that is, the higher the orientation proximity, the higher the score.

[0041] 3. Energy consumption factor scoring; The USV moves in the marine environment, and the energy consumed includes: the forward energy consumption term (the energy consumption in the forward direction of the USV), the turning energy consumption term (the energy consumption for turning), and the ocean current influence term (the energy consumption required for the interaction between the USV and the ocean current). The energy consumption factors are as Figure 5 shown.

[0042] Therefore, the calculation method of the energy consumption is the sum of these three energy consumption forms.

[0043] ;(6) The forward energy consumption term The calculation method is as follows: ;(7) represents the forward propulsion force of the USV (in the surge direction), represents the forward speed of the USV, represents the movement time.

[0044] The turning energy consumption term The calculation method is as follows: ;(8) represents the control torque in the yaw direction, represents the angular velocity, represents the movement time.

[0045] The ocean current influence term The calculation method is as follows: ;(9) represents the ocean current acting force; represents the velocity component of the forward speed direction of the USV in the direction perpendicular to the ocean current, that is, the USV has assistance when flowing with the ocean current and increases consumption when flowing against or laterally to the ocean current; represents the velocity component of the forward speed direction of the USV in the ocean current direction; represents the movement time; Finally, the total energy consumption weight formula is obtained as shown in Equation (10).

[0046] ;(10) represents the th encircling USV's energy consumption required to encircle the th opposing USV; represents the movement time; Therefore, the energy consumption factor score is defined as: ; (11) represents the th encircling USV's energy consumption score for encircling the th counter USV. represents the th encircling USV's required energy consumption for encircling the th counter USV. According to the definition of the energy consumption factor, the lower the energy consumption required for the encircling USV, the higher the score.

[0047] Therefore, the comprehensive scoring formula between the encircling USV and the counter USV is as follows: ; (12) represents the th encircling USV's comprehensive score for encircling the th counter USV. The higher the score, the higher the priority allocation degree, that is, the closer the distance between the encircling USV and the counter USV, the closer the distance between the counter USV and the target area, the higher the orientation proximity, and the lower the energy consumption required for the encircling USV, the higher the comprehensive score. Among them, represents the scoring factor coefficient, satisfying , and the encircling relationship between the encircling USV and the counter USV is determined through the four-factor comprehensive scoring rule.

[0048] IV. Adaptive task allocation strategy based on MFSR.

[0049] After the scoring rule is determined, the comprehensive score between each encircling USV and the counter USV can be determined for adaptive task allocation, as shown in Figure 6 .

[0050] In Figure 6 , represents the number of encircling USVs, represents the number of counter USVs. To achieve the encircling task, usually . For the multi-factor scoring between each encircling USV and each counter USV, the scoring situation can be represented by Equation (13): ; (13) In Equation (13), represents the number of encircling USVs, represents the number of counter USVs. Usually , so the multi-factor score of the rd encircling USV is , , and this encircling USV selects the with the highest score.A ship against the USV performs an encirclement mission.

[0051] V. Simulation experiments.

[0052] 1. Multi-USV encirclement experiment; In the multi-USV encirclement experiment, the present invention sets up 2 encircling USVs and 2 opposing USVs. The initial position of encircling USV1 is (8, 8), the initial position of encircling USV2 is (93, 8), the initial position of opposing USV1 is (-4, 180), and the initial position of opposing USV2 is (101, 183). And 2 square dynamic obstacles with side lengths of 10 m are added to the simulation environment. The initial positions of the dynamic obstacles are (29, 76) and (160, 198) respectively. Affected by ocean disturbances, they will move along with the ocean current direction.

[0053] After the experiment starts, the four-factor scores of the 2 encircling USVs and 2 opposing USVs are shown in Table 1.

[0054] Table 1 Four-factor comprehensive score table

[0055] As can be seen from Table 1, the final scores of encircling USV2 and opposing USV2 are 0.416, and the final scores of encircling USV2 and opposing USV1 are 0.393. Therefore, encircling USV2 will be assigned to oppose USV2 to perform the encirclement mission; in order to avoid repeated task assignment, when opposing USV2 has been assigned to encircling USV2, encircling USV1 can only encircle opposing USV1. At this time, the final score of encircling USV1 and opposing USV1 is 0.412. After the task assignment, the encirclement mission will be executed, and the encirclement trajectory is as Figure 7 shown.

[0056] Table 2 Encirclement results of two encircling USVs

[0057] As can be seen from Table 2, encircling USV1 completed the encirclement mission of opposing USV1 within 29.99 s, and encircling USV2 completed the encirclement mission of opposing USV2 within 37.84 s. The moving distances are 464.96 m and 508.26 m respectively; however, the total energy consumption of encircling USV1 is 5036.89 J, and the total energy consumption of encircling USV2 is 4545.74 J. As Figure 7 can be seen, during the encirclement process of encircling USV1, it mainly moves perpendicular to the ocean current direction and hardly moves in the same direction as the ocean current, which leads to an increase in operating energy consumption; while during the encirclement process of encircling USV2, the moving direction is consistent with the ocean current direction for a long time, thus reducing the navigation energy consumption of the USV.

[0058] 2. Comparative experiments of different algorithms; To compare the differences between the method of the present invention and different task allocation methods, the present invention sets up comparative experiments of different algorithms for 2 hunting USVs and 2 adversarial USVs. The method of the present invention is compared with the Hungarian algorithm and the greedy algorithm respectively, and 10 comparative experiments are carried out respectively. The hunting time of different algorithms and the average energy consumption of the two hunting USVs are counted. The experimental results are as Figure 8 and Figure 9 shown.

[0059] In Figure 8 , the hunting times of the three algorithms to achieve task allocation and hunt 2 adversarial USVs to the target area are shown. The hunting times of the Hungarian algorithm and the greedy algorithm mainly concentrate between 40 s and 65 s, and even extend to 80 s in individual experiments. The overall hunting time fluctuates greatly, the error range is wide, and the stability is poor; while the hunting time of the MFSR-ATAS method proposed by the present invention mainly concentrates between 25 s and 35 s, and the hunting time is shortened by more than 40%. The hunting process is more stable, with smaller fluctuations and a narrower error range, showing stronger time stability and reliability.

[0060] Figure 9 shows the average energy consumption required for each USV when the three algorithms execute the hunting task. The average energy consumption of the Hungarian algorithm and the greedy algorithm mostly distributes between 15 kJ and 35 kJ, and there are large fluctuations; while the scoring energy consumption required by the method of the present invention is 5 kJ to 10 kJ, saving about 70%, greatly reducing the energy loss during the task process, and fully reflecting the significant advantage of the method of the present invention in terms of energy conservation.

Claims

1. A multi-USV roundup method based on multi-factor scoring and adaptive task allocation, characterized in that: The following steps are involved: (1) Establish the mathematical model of USV, including kinematic model and dynamic model; (2) Establish a multi-USV capture mission model; (3) Establish a multi-factor scoring rule: In the USV capture mission, the four influencing factors of the distance between the capture USV and the confrontation USV, the angle required for turning, the distance between the confrontation USV and the target area, and the energy consumption required for the capture USV to perform the mission are combined into a four-factor scoring strategy to obtain the distance factor score, angle factor score, and energy consumption factor score. The comprehensive score between the capture USV and the confrontation USV is obtained based on the scores of each factor. (4) Determine the adaptive task allocation strategy based on the comprehensive score between the encircling USV and the adversarial USV to achieve multi-USV encirclement.

2. The multi-USV roundup method based on multi-factor scoring and adaptive task allocation according to claim 1 is characterized in that: The process of establishing the kinematic model in step (1) is: When the unmanned ship moves on the sea, it moves with six degrees of freedom and establishes two coordinate systems: the geodetic coordinate system fixed on the ground and the attached coordinate system fixed to the hull ; The six-degree-of-freedom USV model is simplified to three-degree-of-freedom, and the kinematic model of three-degree-of-freedom is as follows: ； In the above formula, Indicates the location information of USV, The first-order derivative representing the position information, represents the heading angle of USV, represents the angular velocity, Indicates the forward speed of USV, Indicates the lateral movement speed of USV, Indicates the angular velocity of the USV turning.

3. The multi-USV roundup method based on multi-factor scoring and adaptive task allocation according to claim 1 is characterized in that: The kinetic model in step (1) is as follows: ； In the above formula, , , represents the mass coefficient composed of the USV mass, , , represents the hydrodynamic damping coefficient, Indicates the driving force in the forward direction of the USV, Indicates the driving force of USV steering, Indicates the forward speed of USV, Indicates the lateral movement speed of USV, represents the USV steering angular velocity, represents the first derivative of the USV forward speed, represents the first-order derivative of the lateral movement speed of the USV, Represents the first-order derivative of the USV steering angular velocity.

4. The multi-USV roundup method based on multi-factor scoring and adaptive task allocation according to claim 1 is characterized in that: The process of establishing the multi-USV encirclement task model in step (2) is as follows: The capture USV represents the capture party in the capture mission, and the counter USV represents the escape party in the capture mission. Indicates the number of captured USVs, Indicates the number of adversarial USVs, ; When the capture USV detects the opposing USV, the capture USV moves from the starting position Start, point to the direction of the opposing USV, and reach the distance of the opposing USV. scope When the USV is in position, Move to the driving position on a circular trajectory with a radius of , drive position is the side of the line connecting the target area and the counter USV that is far away from the target area, that is, the counter USV is between the target area and the encirclement USV, so that the counter USV is encircled to the target area to achieve the encirclement task; when the counter USV is encircled and enters the radius When the target area is reached, the capture is successful.

5. The multi-USV roundup method based on multi-factor scoring and adaptive task allocation according to claim 4 is characterized in that: Said , Indicates the USV length.

6. The multi-USV roundup method based on multi-factor scoring and adaptive task allocation according to claim 1 is characterized in that: The process of obtaining the distance factor score in step (3) is: Define the distance factor score as: ； ； Indicates The USV and the The distance score between the opposing USVs, Indicates The USV and the The distance between the opposing USVs; Indicates The distance score between the adversarial USV and the target area; Indicates The distance between the adversarial USV and the target area; The closer the distance between the encircling USV and the opposing USV, and the closer the distance between the opposing USV and the target area, the higher the score.

7. The multi-USV roundup method based on multi-factor scoring and adaptive task allocation according to claim 1 is characterized in that: The process of obtaining the angle factor score in step (3) is: Define the angle factor score as: ； Indicates The direction of the first USV is similar to that of the second Angle score between opposing USVs; Indicates The direction of the first USV is similar to that of the second The angle between the opposing USVs; The smaller the angle between the heading direction of the encircling USV and the opposing USV, that is, the higher the orientation proximity, the higher the score.

8. The multi-USV roundup method based on multi-factor scoring and adaptive task allocation according to claim 1 is characterized in that: The process of obtaining the energy consumption factor score in step (3) is: When USV moves in the ocean environment, the energy consumed includes: forward energy consumption , Steering energy consumption and ocean current influence , so the energy consumption is calculated as the sum of these three forms of energy consumption: ； Energy consumption The calculation is as follows: ； represents the forward propulsion force of the USV, represents the forward speed of the USV, Indicates the moving time; Steering energy consumption The calculation is as follows: ； represents the control torque in the yaw direction, represents the angular velocity, Indicates the moving time; Ocean current influence The calculation is as follows: ； Indicates the force of ocean current; It represents the velocity component of the USV forward speed in the direction perpendicular to the ocean current; It represents the velocity component of the USV forward speed in the direction of the ocean current; Indicates the moving time; Finally, the total energy consumption weight formula is obtained: ； Indicates USV captures The energy consumption required to counter USV; Indicates the moving time; Therefore, the energy consumption factor score is defined as: ； Indicates USV captures Energy consumption score of the anti-USV, Indicates USV captures The energy consumption required to defeat a USV; the lower the energy consumption, the higher the score.

9. The multi-USV roundup method based on multi-factor scoring and adaptive task allocation according to claim 1 is characterized in that: The comprehensive scoring formula between the encircling USV and the confronting USV in step (3) is as follows: ； Indicates The USVs and the The comprehensive score of the USVs is: the higher the score, the higher the priority allocation, that is, the closer the distance between the USV and the USV, the closer the distance between the USV and the target area, the higher the proximity, the lower the energy consumption required for the USV, and the higher the comprehensive score; in, Indicates The USV and the The distance score between the adversarial USVs, Indicates The distance score between the adversarial USV and the target area, Indicates The direction of the first USV is similar to that of the second The angle score between the competing USVs, Indicates USV captures Energy consumption score of the anti-USV; Represents the scoring factor coefficient, satisfying , the encirclement relationship between the encircling USV and the antagonistic USV is determined by a four-factor comprehensive scoring rule.

10. The multi-USV roundup method based on multi-factor scoring and adaptive task allocation according to claim 1 is characterized in that: The adaptive task allocation strategy in step (4) is: For the multi-factor score between each siege USV and each confrontation USV, the score is expressed as follows: ； In the formula, Indicates the number of captured USVs, represents the number of adversarial USVs, , No. The multi-factor score of the captured USV is , The USV with the highest score is selected. A counter USV was carrying out a capture mission.

Images