A ship formation path planning method, device and system

Abstract

The present application relates to the technical field of ships, and provides a ship formation path planning method, device and system. The method comprises: establishing a formation path planning model; wherein the formation path planning model comprises an objective function and a constraint function; solving the formation path planning model to obtain a planning path of each formation; establishing a ship dynamics model; and calculating a required propulsive force of the ship according to the planning path of each formation and in combination with the ship dynamics model, so that each ship in the formation travels along the planning path under the action of the propulsive force. The present application can plan paths for multiple formations by modeling and solving the multiple formations, and can make it possible for the multiple formations to jointly perform a task by introducing a time constraint during modeling.

Description

Technical Field

[0001] This invention relates to the field of ship technology, and in particular to a method, apparatus and system for ship formation path planning. Background Technology

[0002] Given the small size and ease of use of unmanned surface vessels (USVs), which can play a significant role in various fields, path planning for USVs has become a research hotspot in recent years. As an essential function for autonomous navigation of USVs, path planning can plan obstacle-avoiding, collision-free paths from the starting point to the destination, and can also provide relevant local path planning when USVs encounter obstacles, enabling dynamic collision avoidance. Therefore, path planning technology is a fundamental guarantee for the navigation of USVs.

[0003] In existing technologies, a single unmanned vessel is often used to perform corresponding tasks. However, in practical applications, the path planning of a single vessel is prone to limitations such as restricted working range and inability to continue the task in the event of a malfunction. Therefore, overcoming the shortcomings of the existing technology is an urgent problem to be solved in this technical field. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method for planning the route of a ship formation.

[0005] The present invention adopts the following technical solution:

[0006] In a first aspect, the present invention provides a method for planning the path of a ship formation, comprising n formations, the method comprising:

[0007] Establish a formation path planning model; wherein, the formation path planning model includes an objective function and a constraint function, and the objective function is minP = P safe (w L L+w E E+w J J+w T T), the constraint function is Among them, P safe The cost of obstacle avoidance for ships, d is the distance between the ship and the obstacle. safe The preset safety distance; L is the ship's sailing length, L max E is the preset maximum length; J is the energy consumption of the flight path; T is the turning cost; T is the total time difference between formations. max To preset the maximum time difference; w L w represents the path constraint weights. E w represents the energy constraint weights. J w represents the corner constraint weight. T The time constraint weights represent the ship's speed; v is the ship's speed.min To preset the minimum speed, v max The preset maximum speed;

[0008] Solve the formation path planning model to obtain the planned paths for each formation;

[0009] Establish a ship dynamics model: Mv + C(v)v + D(v)v = τ w +τ thr ; where τ w =τ wind +τ wave +τ curren t, τ wind τ represents the force exerted by wind on a ship. wave τ represents the force exerted by waves on a ship. current τ represents the force exerted by the current on the ship. thr Let v represent the ship's own propulsion force, v be the velocity vector of the ship as a result of the combined effects of the environment and the propulsion system, M be the inertia matrix that measures the ease with which the ship's state changes during navigation, C(v) be the centripetal force matrix, and D(v) be the damping matrix, denoted as D(v)∈R. 2×2 ;

[0010] Based on the planned routes of each formation and in conjunction with the ship dynamics model, the required propulsion force for each ship is calculated, enabling each ship in the formation to travel along the planned route under the action of the propulsion force.

[0011] Preferably, the wind force τ on the ship wind The first force τ windx With the second force τ windy The combined force;

[0012] Among them, the first force τ windx The force exerted on the ship from the first direction (upwind). Second force τ windy The force exerted on the ship from the upper direction (second direction). ρ a A represents air density. f It is the projected area of ​​the part of the ship above the waterline, A s It is the side projection area of ​​the part of the ship above the waterline, C wx (α R C represents the wind pressure coefficient in the first direction. wy (α R V represents the wind pressure coefficient in the second direction. w Represents wind speed.

[0013] Preferably, the wave force τ on the ship wave The third force τ wavexWith the fourth force τ wavey The resultant force; among which, the third force τ wavex The force exerted by the waves on the ship in the first direction. Fourth force τ wavey This refers to the force exerted by the waves on the ship in the second direction. ρ is the density of seawater; ζ D χ represents the average wave amplitude during navigation, L is the ship's hull length, χ is the encounter angle during navigation, and C is the average wave amplitude during navigation. XD (λ) is the wave drift coefficient in the first direction, C YD (λ) is the wave drift force coefficient in the second direction.

[0014] Preferably, the force τ exerted by the flow on the ship current The fifth force τ currentx With the sixth force τ currenty The resultant force; among which, the fifth force τ currentx The force exerted on the ship by the upstream current in the first direction. The sixth force τ cuurenty The force exerted on the ship by the upstream current in the second direction. A f V is the orthographic projection area of ​​the part of the ship above the waterline, As is the lateral projection area of ​​the part of the ship above the waterline, and V is the lateral projection area of ​​the part of the ship above the waterline. c β and C are the velocity of the flow relative to the ship and the encounter angle, respectively. X (β) is the force coefficient of the wave in the first direction, C Y (β) is the force coefficient of the wave in the second direction.

[0015] Preferred total time difference between formations Where, ΔT i =T k,max -T i，min T k,max This represents the time taken for the longest formation to reach the target point.

[0016] Preferably, solving the formation path planning model to obtain the planned paths for each formation specifically includes:

[0017] The process involves multiple iterations until a preset number of iterations is reached. Based on the positions obtained from the last iteration, the path of the ship is then determined.

[0018] In each iteration, it is determined whether a preset condition is met; wherein the preset condition is p. <rand(0,1);

[0019] If the preset conditions are not met, the next position is determined using a first predation model; wherein, the first predation model is...

[0020] If the preset conditions are met, then determine Is it greater than or equal to 1? If the value is greater than or equal to 1, then a random search model is used to determine the next position; wherein, the random search model is...

[0021] like If the value is less than 1, the next position is determined using the second predation model; wherein, the predation model is...

[0022] In the first predator model and the second predator model, t is the current iteration number, b is a constant, and the shape of the logarithmic dropout is defined. It is the position vector of the global optimal solution. It is the position vector of the current solution. and For the coefficient vector,

[0023] Preferably, solving the formation path planning model to obtain the planned path for each formation further includes: updating the individual formations in each iteration, specifically including:

[0024] Generate test vectors Update the individual using the test vector. Where i, j, and t represent the individual index, dimension index, and iteration number, respectively; where CR∈(0,1), x i,j,t+1 x is the crossover rate. i,j, t represents the position of the individual in the previous iteration.

[0025] In a second aspect, the present invention also provides a ship formation path planning apparatus for implementing the ship formation path planning method described in the first aspect, the apparatus comprising:

[0026] At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor for performing the ship formation path planning method described in the first aspect.

[0027] Thirdly, the present invention also provides a non-volatile computer storage medium storing computer-executable instructions that are executed by one or more processors to perform the ship formation path planning method described in the first aspect.

[0028] Fourthly, the present invention also provides a ship formation path planning system, including a modeling module and a calculation module;

[0029] The modeling module is used to establish a formation path planning model; wherein, the formation path planning model includes an objective function and a constraint function, and the objective function is minP = P safe (w L L+w E E+w J J+w T T), the constraint function is Among them, P safe The cost of obstacle avoidance for ships, d is the distance between the ship and the obstacle. safe The preset safety distance; L is the ship's sailing length, L max The maximum preset length is T; T is the total time difference between formations. max To preset the maximum time difference; w L w represents the path constraint weights. E w represents the energy constraint weights. J w represents the corner constraint weight. T The time constraint weights represent the ship's speed; v is the ship's speed. min To preset the minimum speed, v max The preset maximum speed;

[0030] The calculation module is used to solve the formation path planning model to obtain the planned path for each formation;

[0031] The modeling module is also used to establish a ship dynamics model Mv+C(v)v+D(v)v=τ w +τ thr ; where τ w =τ wind +τ wave +τ current , τ wind τ represents the force exerted by wind on a ship. wave τ represents the force exerted by waves on a ship. current τ represents the force exerted by the current on the ship. thr Let v represent the ship's own propulsion force, v be the velocity vector of the ship as a result of the combined effects of the environment and the propulsion system, M be the inertia matrix that measures the ease with which the ship's state changes during navigation, C(v) be the centripetal force matrix, and D(v) be the damping matrix, denoted as D(v)∈R. 2×2 ;

[0032] The calculation module is also used to calculate the required propulsion force for each ship based on the planned path of each formation and the ship dynamics model, so that each ship in the formation can travel along the planned path under the action of the propulsion force.

[0033] This invention enables path planning for multiple formations by modeling and solving them, and introduces time constraints during modeling to make it possible for multiple formations to perform tasks together. Attached Figure Description

[0034] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments of the present invention will be briefly described below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0035] Figure 1 This is a flowchart illustrating a ship formation path planning method provided in an embodiment of the present invention;

[0036] Figure 2 This is a schematic diagram of the first ship formation path planning method provided in the embodiments of the present invention;

[0037] Figure 3 This is a flowchart illustrating another ship formation path planning method provided in an embodiment of the present invention;

[0038] Figure 4 This is a schematic diagram of a scenario for a ship formation path planning method provided in an embodiment of the present invention;

[0039] Figure 5 This is a flowchart illustrating the hierarchical analysis method in a ship formation path planning method provided by an embodiment of the present invention;

[0040] Figure 6 This is a schematic diagram of the second ship formation path planning method provided in the embodiments of the present invention;

[0041] Figure 7 This is a schematic diagram of the third ship formation path planning method provided in the embodiments of the present invention;

[0042] Figure 8 This is a schematic diagram of the fourth ship formation path planning method provided in the embodiments of the present invention;

[0043] Figure 9 This is a schematic diagram of the fifth ship formation path planning method provided in the embodiments of the present invention;

[0044] Figure 10 This is a schematic diagram of the sixth ship formation path planning method provided in the embodiments of the present invention;

[0045] Figure 11 This is a schematic diagram of the seventh ship formation path planning method provided in the embodiments of the present invention;

[0046] Figure 12 This is a schematic diagram of the architecture of a ship formation path planning system provided in an embodiment of the present invention;

[0047] Figure 13 This is a schematic diagram of the architecture of another ship formation path planning device provided in an embodiment of the present invention. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0049] In this invention, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0050] Furthermore, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0051] Example 1:

[0052] In existing technologies, single unmanned vessels are often used to perform corresponding tasks. However, in practical applications, the path planning of a single unmanned vessel is prone to limitations such as limited working range and inability to continue the task in case of failure. To solve this problem, Embodiment 1 of this invention provides a method for performing tasks using vessel formations, including n formations, each formation comprising one or more vessels, with the n formations jointly performing the corresponding task, thereby expanding the working range and improving the task success rate. However, existing technologies do not have path planning methods for multiple formations. To solve this problem, Embodiment 1 of this invention provides a vessel formation path planning method, such as... Figure 1 As shown, the method includes:

[0053] In step 201, a formation path planning model is established; wherein, the formation path planning model includes an objective function and a constraint function, and the objective function is minP = P safe (w L L+w E E+w J J+w T T), the constraint function is Among them, P safe The cost of obstacle avoidance for ships, d is the distance between the ship and the obstacle.safe This represents the preset safe distance; L is the ship's sailing length. L i,i+1 Let m represent the length of each travel segment, and let L represent the number of track segments (i.e., travel segments). max E represents the preset maximum length; E represents the energy consumption of the trajectory. E i,i+1 J represents the work done by the ship to overcome environmental resistance during navigation; J is the turning cost. α represents the ship's real-time horizontal turning angle, while α max The maximum horizontal turning angle limit for the vessel; T is the total time difference between formations. T max To preset the maximum time difference, ΔT i =T k,max -T i，min T k,max The time taken for the longest formation to reach the target point; w L w represents the path constraint weights. E w represents the energy constraint weights. J w represents the corner constraint weight. T The time constraint weights are represented by v; v is the ship's speed, which is usually the average speed between road segment node i and road segment node i+1. min To preset the minimum speed, v max The preset maximum speed is defined as follows: the preset safety distance, preset maximum length, preset maximum time difference, preset minimum speed, and preset maximum speed are all obtained by those skilled in the art based on experience. The path constraint weights, energy constraint weights, turning angle constraint weights, and time constraint weights are obtained by those skilled in the art using the analytic hierarchy process (AHP) to ensure that w... L +w E +w J The constraint condition is 1.

[0054] Since multiple formations are performing the same mission, they should arrive at the destination within the same time frame. However, different formations may have different sailing speeds and departure times. Therefore, to ensure that all formations arrive within the same time frame, the aforementioned T≤T is proposed. max Given the time constraints, let the speed of each formation be [v]. min ,v max If we calculate the path length L1 for formation 1, then the time interval for reaching the target point is... And reach the shortest time and longest time Then we get Figure 2The time coordination strategy shown sorts the formations according to their travel time. The formation with the longest travel time departs first, followed by the next longest, and so on, until the formation with the shortest travel time departs. All formations arrive at the target point simultaneously. Through this time coordination, multiple formations can reach a predetermined target point simultaneously in the shortest possible time. In this process, let k be the formation with the longest travel time, and T be the maximum time required for it to reach the target point. k,max Then, the time difference between formation i and formation k reaching the target point is calculated as ΔT. i =T k,max -T i，min This indicates that formation i must be in T. k,max Depart beforehand, so that in T k,max When formation k and formation k arrive at the target point simultaneously, the total time difference T between formations is calculated. This total time difference is equal to the sum of the absolute values ​​of the time differences between all formations. The value of T indicates the departure interval between formations; a large departure interval indicates poor time coordination between formations. Therefore, it is desirable to minimize the value of T to strengthen the time coupling between formations. Under ideal conditions, T... max =0, meaning all formations arrive at the target location simultaneously.

[0055] In step 202, the formation path planning model is solved to obtain the planned path for each formation; wherein, the solution method includes various methods such as ant colony algorithm, particle swarm algorithm and whale algorithm.

[0056] In step 203, the ship dynamics model Mv + C(v)v + D(v)v = τ is established. w +τ thr ; where τ w =τ wind +τ wave +τ current , τ wind τ represents the force exerted by wind on a ship. wave τ represents the force exerted by waves on a ship. current τ represents the force exerted by the current on the ship. t hr represents the ship's own propulsion force, v is the velocity vector of the ship as a result of the combined effects of the environment and the propulsion system, M is the inertia matrix that measures the ease with which the ship's state changes during navigation, C(v) is the centripetal force matrix, and D(v) is the damping matrix, denoted as D(v)∈R. 2×2 .

[0057] In step 204, based on the planned paths of each formation and the ship dynamics model, the required thrust for each ship is calculated, enabling the ships in the formation to travel along the planned paths under the influence of this thrust. Specifically, the required velocity vector *v* is first calculated based on the planned paths, and then this velocity vector *v* is substituted into the ship dynamics model to calculate the ship's own thrust. Here, "ship" is a general term for waterborne mobile tools, including one or more types such as boats, boats, and ships. The phrase "enabling each ship in the formation to travel along the planned paths under the influence of this thrust" actually means ensuring that all ships in the formation, under the influence of this thrust, allow the formation to travel along the planned paths.

[0058] In practical applications, when a formation contains only one vessel, the position of that vessel is used as the formation's position for path planning. When a formation contains multiple vessels, the position of one vessel can be selected as the formation's position for path planning. Considering that in real-world scenarios, vessels in a formation may experience anomalies, causing some vessels to detach from the formation, this embodiment also provides a preferred implementation method to avoid formation chaos: selecting the mathematical center of the formation as the formation's position for path planning.

[0059] This embodiment models and solves for multiple formations, enabling path planning for them. Furthermore, this embodiment introduces time constraints during modeling, making it possible for multiple formations to perform tasks together.

[0060] Among them, the force τ exerted by the wind on the ship wind The first force τ windx With the second force τ windy The combined force.

[0061] Among them, the first force τ windx The force exerted on the ship from the first direction (upward). Second force τ windy The force exerted on the ship from the upper direction (second direction). ρ a A represents air density. f It is the projected area of ​​the part of the ship above the waterline, A s It is the side projection area of ​​the part of the ship above the waterline, C wx (α R C represents the wind pressure coefficient in the first direction. wy (α R V represents the wind pressure coefficient in the second direction. w Represents wind speed.

[0062] In practical use, the first direction is horizontal, and the second direction is vertical.

[0063] The force τ of waves on a ship wave The third force τ wavex With the fourth force τ wavey The resultant force; among which, the third force τ wavex The force exerted by the waves on the ship in the first direction. Fourth force τ wavey This refers to the force exerted by the waves on the ship in the second direction. ρ is the density of seawater; ζ D χ represents the average wave amplitude during navigation, L is the ship's hull length, χ is the encounter angle during navigation, and C is the average wave amplitude during navigation. XD (λ) is the wave drift coefficient in the first direction, C YD (λ) is the wave drift force coefficient in the second direction.

[0064] The force τ exerted by the current on the ship current The fifth force τ currentx With the sixth force τ currenty The resultant force; among which, the fifth force τ currentx The force exerted on the ship by the upstream current in the first direction. The sixth force τ cuurenty The force exerted on the ship by the upstream current in the second direction. A f V is the orthographic projection area of ​​the part of the ship above the waterline, As is the lateral projection area of ​​the part of the ship above the waterline, and V is the lateral projection area of ​​the part of the ship above the waterline. c β and C are the velocity of the flow relative to the ship and the encounter angle, respectively. X (β) is the force coefficient of the wave in the first direction, C Y (β) is the force coefficient of the wave in the second direction.

[0065] When solving the formation path planning model using existing technologies, problems such as slow convergence speed, poor global optimization ability, and low convergence accuracy were found. To solve this problem, this embodiment also provides a preferred implementation method, namely, solving the formation path planning model to obtain the planned paths for each formation, such as... Figure 3 As shown, it specifically includes:

[0066] In step 301, multiple iterations are performed until the number of iterations reaches a preset number. Then, the path of the ship is determined based on the positions obtained from the last iteration. The preset number of iterations is obtained by those skilled in the art based on experience.

[0067] In step 302, in each iteration, it is determined whether a preset condition is met; wherein the preset condition is p. <rand(0,1)。

[0068] In step 303, if the preset conditions are not met, the next position is determined using a first predation model; wherein, the first predation model is...

[0069] In step 304, if the preset condition is met, then a determination is made. Is it greater than or equal to 1? If the value is greater than or equal to 1, then a random search model is used to determine the next position; wherein, the random search model is...

[0070] In step 305, if If the value is less than 1, the next position is determined using the second predation model; wherein, the predation model is...

[0071] In the first predator model and the second predator model, t is the current iteration number, b is a constant, and the shape of the logarithmic dropout is defined. It is the position vector of the global optimal solution. It is the position vector of the current solution. and For the coefficient vector,

[0072] The above preferred implementation can be understood as an improvement on the existing whale algorithm. Specifically, the existing whale algorithm has three localization methods: the first is bounding, the second is random search, and the third is spiral update. The existing models corresponding to bounding and spiral update are as follows: When p > rand(0,1), a spiral update is performed, i.e., during the spiral update... This embodiment introduces a sine wave into the spiral update formula. As interference and auxiliary terms, we obtain the formula. This improves convergence accuracy.

[0073] In practical applications, it has been found that the later iterations of solving the formation path planning model are prone to getting trapped in local optima, making it impossible to find the global optimum. To address this issue, this embodiment also provides a preferred implementation method, namely, solving the formation path planning model to obtain the planned paths for each formation further includes: updating the individual elements in each iteration, specifically including:

[0074] Generate test vectors Update the individual using the test vector. Where i, j, and t represent the individual index, dimension index, and iteration number, respectively; where CR∈(0,1), x i,j,t+1x is the crossover rate. i,j,t This represents the position of the individual in the previous iteration.

[0075] The above implementation can be understood as introducing the crossover and selection operations of the Differential Evolution Algorithm (DE) into the whale algorithm. This leverages the global optimization capabilities of the traditional whale algorithm, proactively adapts to coupling constraints, and protects dominant individuals. Solution space analysis of the traditional algorithm reveals that the crossover operation in DE reduces the likelihood of pushing individuals into infeasible regions, resulting in a higher search frequency in each iteration. This leads to greater diversity in search results, increases the opportunity to search other regions, and avoids getting trapped in local optima.

[0076] When solving the formation path planning model, the minimum value P of the comprehensive function and the corresponding length cost, turning cost, energy consumption cost and coordination cost can be obtained. At the same time, the position coordinates of each step can be obtained. By smoothing and connecting these coordinates, the planned path of each formation can be obtained.

[0077] Example 2:

[0078] Based on the method described in Embodiment 1, this invention combines specific application scenarios and uses technical descriptions in relevant scenarios to illustrate the implementation process of the features of this invention in those scenarios.

[0079] For example, if there are multiple formations, each containing multiple unmanned surface vessels (i.e., the vessels in Example 1), it is necessary to consider their operation in situations such as... Figure 4 Formation path planning is performed within a two-dimensional square water area of ​​40×40km. The following conditions must be met during formation path planning:

[0080] 1) Obstacles and threat zones can be viewed as various irregular obstacles composed of square areas.

[0081] 2) When performing global planning for unmanned surface vessels (USVs), it is essential to first ensure that there are no collisions between the USV and obstacles before considering whether its planned path can achieve global optimization.

[0082] Figure 4 Each black block in the diagram represents an obstacle. The ship formation path planning method provided in this embodiment includes two parts: model building and model solving.

[0083] The model building process specifically includes:

[0084] For path planning modeling of a single unmanned surface vessel (USV), obstacle avoidance factors are considered first, and an obstacle avoidance cost function for the USV is established. d is the distance between the unmanned surface vessel and the obstacle, and d safeThe pre-set safe distance between the unmanned surface vessel and obstacles.

[0085] Set sailing length Can be viewed L i,i+1 Let m be the length of each travel segment, and m be the number of track segments.

[0086] Set corner cost In the formula, α represents the real-time horizontal turning angle of the unmanned surface vessel, and α max This is the maximum horizontal turning angle limit for unmanned surface vessels.

[0087] Set track energy consumption v is the average speed of the unmanned surface vessel between nodes i and i+1, E i,i+1 This refers to the work done by the unmanned surface vessel (USV) to overcome environmental interference and resistance during navigation.

[0088] The weights w are determined using the analytic hierarchy process. L w E w J Thus, the comprehensive cost of path planning for a single vessel is established.

[0089] minP=P safe (w L L+w E E+w J J)

[0090] function

[0091] Among them, the analytic hierarchy process (AHP) is as follows: Figure 5 As shown, it includes:

[0092] In step 401, a reasonable and systematic hierarchical structure is established.

[0093] In step 402, a judgment matrix is ​​constructed and assigned values.

[0094] In step 403, a consistency check is performed.

[0095] Among them, w L w E w J The weights representing path length, energy consumption, and cornering costs are obtained using the analytic hierarchy process (AHP) and satisfy w. L +w E +w J The constraint condition is 1, and the solution to the objective function must be carried out under the premise of obstacle avoidance to ensure the safe navigation of the unmanned surface vessel. At the same time, the length L of the unmanned surface vessel's travel path should not exceed the maximum length L within the navigation area. max The speed v of the unmanned surface vessel while it is in motion must not be less than its own minimum speed v. min(i.e., the preset minimum speed) cannot exceed its own maximum speed v. max (i.e., the preset maximum speed).

[0096] Establish a single-boat dynamic model: Mv + C(v)v + D(v)v = τ w +τ thr (i.e., ship dynamics model) and τ w =τ wind +τ wave +τ current M is the inertia matrix that measures the ease with which the unmanned surface vessel changes state during navigation; C(v) is the centripetal force matrix; D(v) is the damping matrix, denoted as D(v)∈R. 2×2 ;τ wind , τ wave and τ current τ represents the forces exerted on unmanned surface vessels by marine environmental factors such as wind, waves, and currents; thr Let v represent the propulsion force of the unmanned system's own propulsion system, and v be the velocity vector of the unmanned surface vessel caused by the combined effects of the environment and the propulsion system.

[0097] And modeling of ocean factors such as wind, waves, and currents for a single vessel, including:

[0098] Wind Model Where ρ a A represents air density. f A is the projected area of ​​the unmanned surface vessel above the waterline. s It is the side projection area of ​​the unmanned surface vessel above the waterline, while C wx (α R ) and C wy (α R V represents the longitudinal and lateral wind pressure coefficients. w Represents wind speed.

[0099] Wave Model Where τ wavex It is the lateral wave interference force, τ wavey Represents the longitudinal disturbance force; ρ is the density of seawater; ζ D The average wave amplitude during navigation is given by χ, where L is the length of the unmanned surface vessel (USV); χ is the encounter angle during navigation; and C is the average wave amplitude during navigation. XD (λ) and C YD (λ) is the wave drift force coefficient of the wave in the transverse direction (i.e., the first direction in Example 1) and longitudinal direction (i.e., the second direction in Example 1).

[0100] and the model of flow τ currentx and τ currenty These represent the lateral and longitudinal disturbance forces experienced by the unmanned surface vessel, respectively, while A fV is the orthographic projection area of ​​the unmanned surface vessel above the waterline, As is the lateral projection area of ​​the unmanned surface vessel above the waterline, and V is the lateral projection area of ​​the unmanned surface vessel above the waterline. c β and C are the velocity of the flow relative to the ship and the encounter angle, respectively. X (β) and C Y (β) is the force coefficient of the wave in the horizontal and vertical directions.

[0101] After establishing the path planning comprehensive cost function and dynamics model of a single unmanned surface vessel (USV), the control mode among the individual vessels in the formation is modeled. This involves using the positions of each USV as virtual points to establish a virtual navigation point model. This model can guide the USV formation's navigation based on actual conditions. By simply monitoring the virtual navigator's movements, the formation's trajectory can be tracked. The selection of the virtual navigation center point for the USV formation can be achieved in two ways:

[0102] The first method involves selecting a virtual point within the unmanned surface vessel (USV) formation as a virtual navigation center. This center point is used to calculate the individual positions and related information of the surrounding USVs, thereby ensuring the navigation and related activities of the USVs.

[0103] The second approach: Select the mathematical center of the unmanned surface vessel formation as the virtual navigation point model, such as... Figure 6 As shown.

[0104] While the first method can determine formation and navigation information by acquiring only a small amount of data from the virtual navigation point model, damage or anomalies in the individual component of this virtual navigation point model can lead to chaos in the unmanned surface vessel (USV) formation. Therefore, the second method, which selects a mathematical center point, is not only logically simple and quick but also easy to implement in practical applications. Subsequent implementations will specifically describe the second method, analyzing the USV formation as a virtual navigation point model with its mathematical center.

[0105] After determining the internal structure model of the unmanned surface vessel (USV) formation, a time coupling model between USV formations is established, assuming the velocity of each formation is [v]. min ,v max If we calculate the path of length L1 for unmanned surface vessel formation 1, then the time interval for reaching the target point is... And reach the shortest time and longest time Then we get Figure 4 The time coordination strategy shown sorts the unmanned surface vessel (USV) formations according to their travel time. The formation with the longest travel time departs first, followed by the next longest, and so on, until the formation with the shortest travel time departs. All formations arrive at the target point simultaneously. Through this time coordination, multiple USV formations can reach a predetermined target point simultaneously in the shortest possible time. In this process, let k be the formation with the longest travel time, and T be the maximum time required for it to reach the target point.k,max Then, the time difference between formation i and formation k reaching the target point is calculated as ΔT. i =T k,max -T i，min This indicates that formation i must be in T. k,max Depart beforehand, so that in T k,max If the unmanned surface vessel (USV) formation can arrive at the target point simultaneously with formation k, then the total time difference T between formations is calculated. This total time difference is equal to the sum of the absolute values ​​of the time differences between all formations. The value of T indicates the departure interval between formations; if the departure interval is too large, it indicates poor time coordination between USV formations. Therefore, it is desirable to minimize the value of T to strengthen the time coupling between formations.

[0106] Based on the aforementioned time coordination and the objective function of path planning for a single vessel considering the marine environment, the formation path planning model is determined as follows: And w L +w E +w J +w T =1.

[0107] Based on the above model, the model is solved. In this embodiment, an improved whale algorithm is used for the solution. The solution approach of the improved whale algorithm is as follows: Figure 7 and Figure 8 As shown, the improved whale algorithm makes the following improvements compared to the existing whale algorithm:

[0108] By introducing crossover and selection operations from the Differential Evolutionary Algorithm (DE), the global optimization capability of the traditional whale algorithm is leveraged, allowing for proactive adaptation to coupling constraints and protection of dominant individuals. Solution space analysis of the traditional algorithm reveals that the crossover operation in DE reduces the likelihood of pushing individuals into infeasible regions. Therefore, based on the crossover operation in the DE algorithm, test vectors are generated. Where i, j, and t represent the individual index, dimension index, and iteration number, respectively; CR∈(0,1), x i,j,t+1 The crossover rate is calculated using various models of the improved whale algorithm, including the first predator model, the second predator model, and the random search model. It can be understood as the iterative result (i.e., position) obtained in each iteration of the improved whale algorithm. After that, based on the position Calculate x i,j,t+1 ;x i,j,t This represents the position of the individual from the previous iteration. To prevent individuals from moving towards positions where fitness crosses, the hybrid algorithm can introduce a selection operation from the Dependency Evolution (DE). This allows for updating individuals after each iteration, resulting in a higher search frequency in each iteration. This leads to greater diversity in search results, increases the chance of searching other regions, and avoids getting trapped in local optima.

[0109] Furthermore, this embodiment also improves the existing whale algorithm model used for spiral update and prey search, namely... That is, by adding a sine wave As interference and auxiliary terms, they improve convergence accuracy. According to the above equation, when... Retain if no better solution is found If a solution is not found, a better solution is sought, and the solution exhibits diversity. In summary, the improved whale algorithm flowchart is as follows: Figure 8 As shown.

[0110] For ease of understanding, the improved whale algorithm proposed in this embodiment will be compared with the existing whale algorithm:

[0111] Existing whale algorithms construct and solve mathematical models based on the hunting behavior of humpback whales. They use... The formula summarizes the behavior of humpback whales locating their prey during a fall, where t represents the current iteration number. and For the coefficient vector, It is the position vector of the global optimal solution, and This is the position vector of the current solution. If a better solution appears during the iteration process, it should be updated immediately. and The expression is During the iteration process, a decreases linearly from 2 to 0. and It is a random vector between [0, 1].

[0112] In addition, whale falls also employ a bubble-web attack pattern, which consists of a shrinking encirclement mechanism and a spiraling repositioning. This mechanism is essentially derived from the encirclement and predation activities of humpback whale pods. The spiral repositioning is another non-monetary behavior of humpback whales, and its process can be summarized by a mathematical model: the whale's position (X,Y) and the prey's position (X... * ,Y * The distance between the whale and its prey, and the positional modeling between them, can be expressed as... Where b is a constant, defining the shape of the logarithmic fallout. It can be observed that the humpback whales swim around their prey within a continuously shrinking circle, while simultaneously moving along a spiral path. For humpback whales' predation activities, both encirclement methods occur simultaneously, therefore, mathematical modeling of both predation behaviors needs to be considered. The value of the whale's position is calculated to determine the optimal position at each step in the iteration process. Since whales may simultaneously engage in two different behaviors during predation, the formula is used... Describe the probability of whales choosing the contraction encirclement mechanism and the spiral model mechanism during predation, and update the whale's position in predation behavior accordingly.

[0113] Besides the bubble-web attack pattern, whale pods also use random search methods to find prey, and individual whales within the pod will also search randomly based on the location of their companions. Greater than 1 or When the value equals 1, it randomly selects a value to distance itself from the prey and find a more suitable one. This search mechanism differs from the bubble web attack phase. In the whale fall algorithm, the position of the search phase is updated by randomly selecting a search agent, thereby improving the overall position of the whale fall and enhancing its search capability. Its mathematical model is as follows: Here, is a randomly selected position vector within the current population, representing a random whale in the population during predation. Therefore, the above formula can be derived. The WOA algorithm starts with a set of random solutions, and in each iteration, its position is updated based on a randomly selected search agent or the current optimal solution.

[0114] In the existing whale algorithm, the solution process can be understood as follows:

[0115] The process involves multiple iterations until a preset number of iterations is reached. Based on the positions obtained from the last iteration, the path of the ship is then determined.

[0116] In each iteration, it is determined whether a preset condition is met; wherein the preset condition is p. <rand(0,1)。

[0117] If the preset conditions are not met, then use Determine the next position (i.e., perform a spiral update); where,

[0118] If the preset conditions are met, then determine Is it greater than or equal to 1? If greater than or equal to 1, then use (i.e., the random search model in Example 1) determines the next location (i.e., to search for prey).

[0119] like If less than 1, then use (i.e., the predation model in Example 1) determines the next location (i.e., performs a contraction and encirclement); wherein, t is the current iteration number, b is a constant, and the shape of the logarithmic dropout is defined. It is the position vector of the global optimal solution. It is the position vector of the current solution. and For the coefficient vector,

[0120] The improved whale algorithm proposed in this embodiment introduces a first predation model. Spiral updates are performed using the first predator model to improve convergence accuracy.

[0121] After solving the formation path planning model using the improved whale algorithm, the minimum value P of the comprehensive function and the corresponding length cost, corner cost, energy cost, and coordination cost can be obtained. Simultaneously, the position coordinates at each step are calculated. By smoothing and connecting these coordinates, the planned paths for each formation can be obtained. The specific ideas for path planning and solution are as follows: Figure 9 As shown.

[0122] Assuming the starting coordinates of the unmanned surface vessel (USV) are (0, 0) and the destination target is (29, 29), the solution result for single-vessel modeling is as follows: Figure 10 As shown, the solution results for multi-unmanned surface vessel formations are as follows: Figure 11 As shown in the figure. The actual coordinates obtained are those of the unmanned surface vessel (USV), and connecting these coordinates results in a straight line. To accurately represent the nature of its route, a fitting technique is used to smooth it into a curve, resulting in the following... Figure 10 He Ru Figure 11 The planned path is shown. Based on the single-vessel dynamics model and the planned path, the required thrust for each unmanned surface vessel can then be obtained.

[0123] This embodiment considers global optimization as well as obstacle avoidance, energy, and turning constraints, and uses global path planning as the final generated path, which can realistically reflect the trajectory characteristics of the unmanned surface vessel (USV) formation. The path planning based on the improved whale algorithm can achieve better time coordination of USV formations and demonstrates robustness and superior experimental results in complex sea areas and formation experiments.

[0124] Example 3:

[0125] Based on Embodiments 1 and 2, this embodiment also provides a ship formation path planning system, such as Figure 12 As shown, it includes a modeling module and a calculation module.

[0126] The modeling module is used to establish a formation path planning model; wherein, the formation path planning model includes an objective function and a constraint function, and the objective function is minP = P safe (w L L+w E E+w J J+w T T), the constraint function is Among them, P safe The cost of obstacle avoidance for ships, d is the distance between the ship and the obstacle. safe The preset safety distance; L is the ship's sailing length, L max The maximum preset length is T; T is the total time difference between formations. max To preset the maximum time difference; w L w represents the path constraint weights. E w represents the energy constraint weights. J w represents the corner constraint weight. T The time constraint weights represent the ship's speed; v is the ship's speed. min To preset the minimum speed, v max This is the preset maximum speed.

[0127] The calculation module is used to solve the formation path planning model to obtain the planned path for each formation.

[0128] The modeling module is also used to establish a ship dynamics model Mv+C(v)v+D(v)v=τ w +τ thr ; where τ w =τ wind +τ wave +τ current , τ wind τ represents the force exerted by wind on a ship. wave τ represents the force exerted by waves on a ship. current τ represents the force exerted by the current on the ship. thr Let v represent the ship's own propulsion force, v be the velocity vector of the ship as a result of the combined effects of the environment and the propulsion system, M be the inertia matrix that measures the ease with which the ship's state changes during navigation, C(v) be the centripetal force matrix, and D(v) be the damping matrix, denoted as D(v)∈R. 2×2 .

[0129] The calculation module is also used to calculate the required propulsion force for each ship based on the planned path of each formation and the ship dynamics model, so that each ship in the formation can travel along the planned path under the action of the propulsion force.

[0130] This embodiment also provides a ship formation path planning device, such as Figure 13 The diagram shown is a schematic representation of the architecture of a ship formation path planning device according to an embodiment of the present invention. The ship formation path planning device of this embodiment includes one or more processors 21 and a memory 22. Figure 13 Take a processor 21 as an example.

[0131] Processor 21 and memory 22 can be connected via a bus or other means. Figure 13 Taking the example of a connection between China and Israel via a bus.

[0132] The memory 22, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs and non-volatile computer-executable programs, such as the ship formation path planning method in Embodiment 1. The processor 21 executes the ship formation path planning method by running the non-volatile software programs and instructions stored in the memory 22.

[0133] Memory 22 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, memory 22 may optionally include memory remotely located relative to processor 21, which can be connected to processor 21 via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0134] The program instructions / modules are stored in the memory 22 and, when executed by one or more processors 21, execute the ship formation path planning method in Embodiment 1 above.

[0135] It is worth noting that the information interaction and execution process between the modules and units in the above-mentioned device and system are based on the same concept as the processing method embodiment of the present invention. For details, please refer to the description in the method embodiment of the present invention, and will not be repeated here.

[0136] Those skilled in the art will understand that all or part of the steps in the various methods of the embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include: read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, etc.

[0137] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method of ship formation path planning, characterized by, The method comprises n platoons, and the method comprises the following steps: A formation path planning model is established, wherein the formation path planning model comprises a target function and a constraint function, the target function is min P=P safe (w L L+w E E+w J J+w T T), and the constraint function is wherein P safe is an obstacle avoidance cost of the ship, d is a distance between the ship and the obstacle, d safe is a preset safety distance; L is a sailing length of the ship, L max is a preset maximum length; E is a track energy consumption; J is a turning angle cost; T is a total time difference between formations, T max is a preset maximum time difference; w L represents a path constraint weight value, w E represents an energy constraint weight value, w J represents a turning angle constraint weight value, w T represents a time constraint weight value; v is a running speed of the ship, v min is a preset minimum speed, v max is a preset maximum speed; Solving the platoon path planning model to obtain a planning path of each platoon; A ship dynamics model Mv + C(v)v + D(v)v = τ is established w +τ thr ; wherein, τ w = τ wind + τ wave + τ current , τ wind represents the force of wind on the ship, τ wave represents the force of waves on the ship, τ current represents the force of flow on the ship, τ thr represents the propulsive force of the ship itself, v is the speed vector of the ship under the joint action of the environment and the propulsion system, M is an inertia matrix measuring the difficulty of state change of the ship during navigation, C(v) is a centripetal force matrix; D(v) is a damping matrix, represented as D(v) ∈ R 2 ×2 ; According to the planning path of each platoon, the ship dynamics model is combined to calculate a required propulsive force of the ship, so that each ship in the platoon travels along the planning path under the action of the propulsive force.

2. The ship formation path planning method according to claim 1, characterized in that, the wind force τ on the ship wind is the first force τ windx is the second force τ windy is the resultant force; wherein τ windx is the first force of the wind on the ship in the first direction, τ windy is the second force of the wind on the ship in the second direction, ρ a represents the air density, A f is the frontal projection area of the ship above the waterline, A s is the lateral projection area of the ship above the waterline, C wx (α R ) represents the wind pressure coefficient in the first direction, C wy (α R ) represents the wind pressure coefficient in the second direction, V w represents the wind speed.

3. The ship formation path planning method of claim 1, wherein, The force τ of waves on a ship wave The third force τ wavex With the fourth force τ wavey The resultant force; among which, the third force τ wavex The force exerted by the waves on the ship in the first direction. Fourth force τ wavey This refers to the force exerted by the waves on the ship in the second direction. ρ is the density of seawater; ζ D χ represents the average wave amplitude during navigation, L is the ship's hull length, χ is the encounter angle during navigation, and C is the average wave amplitude during navigation. XD (λ) is the wave drift coefficient in the first direction, C YD (λ) is the wave drift force coefficient in the second direction.

4. The ship formation path planning method of claim 1, wherein, τ current is the fifth force τ currentx is the resultant of the sixth force τ currenty ; wherein the fifth force τ currentx is the force of the flow on the ship in the first direction, the sixth force τ cuurenty is the force of the flow on the ship in the second direction, A f is the right projection area of the ship above the waterline, As is the side projection area of the ship above the waterline, V c and β are the speed and encounter angle of the flow relative to the ship, C X (β) is the force coefficient of the wave in the first direction, C Y (β) is the force coefficient of the wave in the second direction.

5. The ship formation path planning method of claim 1, wherein, Total time difference between platoons where ΔT i = T k,max - T i，min , T k,max is the time of arrival at the target point for the platoon with the longest time.

6. The ship formation path planning method of claim 1, wherein, The solving of the platoon path planning model to obtain a planning path of each platoon specifically comprises the following steps: Performing multiple iterations until the number of iterations reaches a preset number, and determining the path of the ship based on the positions obtained in the last iteration; In each iteration, it is determined whether a preset condition is met; the preset condition is p<rand(0,1). If the preset condition is not met, a first predation model is used to determine the next position; wherein the first predation model is If the preset condition is met, it is determined whether is greater than or equal to 1, if is greater than or equal to 1, a random search model is used to determine the next position; wherein the random search model is If is less than 1, a second predation model is used to determine the next position; wherein the predation model is In the first and second predation models, t is the current iteration number, b is a constant, defining the shape of the log drop-off, is the position vector of the global optimal solution, is the position vector of the current solution, and is the coefficient vector, 7. The ship formation path planning method of claim 1, wherein, The solving of the platoon path planning model to obtain a planning path of each platoon further comprises the following steps in each iteration: Generating test vectors Updating individuals using the test vectors where i, j, t represent individual index, dimension index and iteration number respectively; where CR∈(0, 1), x i,j,t+1 is the crossover rate, x i,j,t is the position of the individual in the previous iteration.

8. A ship formation path planning device characterized by comprising: Comprise: At least one processor; And a memory connected in communication with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the processor to execute the ship platoon path planning method of any one of claims 1-7.

9. A non-transitory computer storage medium, comprising: The computer storage medium stores computer executable instructions, and the computer executable instructions are executed by one or more processors to complete the ship platoon path planning method of any one of claims 1-7.

10. A ship formation path planning system, characterized by, Comprise a modeling module and a calculation module; The modeling module is configured to establish a platoon path planning model; wherein the platoon path planning model comprises a target function and a constraint function, the target function is min P=P safe (w L L+w E E+w J J+w T T), and the constraint function is wherein P sa fe is an obstacle avoidance cost of the ship, d is a distance between the ship and the obstacle, d safe is a preset safety distance; L is a navigation length of the ship, L max is a preset maximum length; T is a total time difference between platoons, T max is a preset maximum time difference; w L represents a path constraint weight, w E represents an energy constraint weight, w J represents a turning angle constraint weight, w T represents a time constraint weight; v is a running speed of the ship, v min is a preset minimum speed, v max is a preset maximum speed; The calculation module is used for solving the platoon path planning model to obtain a planning path of each platoon; The modeling module is further configured to establish a ship dynamics model Mv+C(v)v+D(v)v=τ w +τ thr ; wherein τ w =τ wind +τ wave +τ current , τ wind represents the force of wind on the ship, τ wave represents the force of wave on the ship, τ current represents the force of flow on the ship, τ thr represents the propelling force of the ship itself, v is a speed vector of the ship under the joint action of the environment and the propulsion system, M is an inertia matrix measuring the difficulty of state change of the ship during navigation, C(v) is a centripetal force matrix, and D(v) is a damping matrix, represented as D(v)∈R 2×2 ; The calculation module is also used for combining the ship dynamics model according to the planning path of each platoon to calculate a required propulsive force of the ship, so that each ship in the platoon travels along the planning path under the action of the propulsive force.

Images