Method for planning coverage path of large-scale unmanned ship group in any polygonal area

By establishing a raster map in the polygon task area and using simulated annealing algorithm optimization, combining spanning tree and area division algorithm, the flexibility and coverage problems of unmanned boat groups covering path planning in the polygon area are solved, and efficient Path planning.

CN118936471APending Publication Date: 2024-11-12DALIAN MARITIME UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410974578.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-19
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

The existing unmanned boat coverage path planning method cannot be directly applied to polygonal task areas, which has insufficient flexibility and applicability, and it is difficult to ensure the accuracy of coverage and scalability of proportional division.

Method used

A cover path planning method for large-scale unmanned boat clusters in any polygonal area is adopted. By establishing a raster map of polygonal task area, using optimization model and simulated annealing algorithm to generate a raster map with higher coverage, and combining the spanning tree algorithm and the area division algorithm, divide the task area proportionally and generate coverage paths.

Benefits of technology

It significantly improves the flexibility and applicability of the unmanned boat fleet in the polygonal mission area, achieves a coverage path with higher coverage rate and fewer turn counts, and is suitable for the practical application of large-scale unmanned boat fleets.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118936471A_ABST
    Figure CN118936471A_ABST
Patent Text Reader

Abstract

The invention provides a coverage path planning method for a large-scale unmanned ship group in an arbitrary polygonal area, and relates to the technical field of ship and ocean engineering, and the method comprises the following steps: S1, selecting a polygonal task area, building a grid map of the polygonal task area, S2, building an optimization model, and optimizing the grid map by using the optimization model, generating a grid map with a higher coverage rate; s3, dividing regions on the grid map with higher coverage rate according to a proportion by adopting a region division algorithm to allocate a task region for each unmanned ship; and generating a coverage path in the task area of each unmanned ship by adopting a spanning tree algorithm. The method not only can efficiently cover the rectangular task area, but also can flexibly adapt to the polygonal task area, and the flexibility and applicability in practical application are remarkably improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of ship and marine engineering technology, and in particular to a coverage path planning method for a large-scale unmanned boat group in an arbitrary polygonal area. Background Art

[0002] The ocean is one of the most precious resources on earth and an important cornerstone for human exploration, survival and development. Unmanned boats are widely used in exploring the ocean, developing the ocean and deploying underwater equipment. The tasks that a single unmanned boat can perform are limited and inefficient. At present, there is a general tendency to use the collaboration between multiple unmanned boats to achieve various tasks and goals. Unmanned boat path planning is the core link in its autonomous navigation process, which directly determines whether the unmanned boat can complete various tasks efficiently and safely. Unmanned boat path planning can be divided into unmanned boat coverage path planning and unmanned boat target arrival path planning. Among them, unmanned boat coverage path planning is a key component of the unmanned boat path planning task. Its goal is to ensure that multiple unmanned boats sail from their respective starting points to cover the entire mission area and avoid obstacles, while reducing path duplication.

[0003] In recent years, the field of coverage path planning has attracted in-depth research and exploration by scholars at home and abroad, and proposed coverage path planning methods including biologically inspired neural network algorithms and spanning tree-based coverage algorithms. However, the existing unmanned boat coverage path planning methods have the following limitations: First, the existing unmanned boat coverage path planning methods have limitations when dealing with mission areas. These methods are usually only applicable to coverage path planning in rectangular areas, and cannot be directly applied to polygonal mission areas, which limits their flexibility and applicability in practical applications. Second, the raster map created by the existing unmanned boat coverage path planning method has certain errors in the representation of the actual planning area, and it is difficult to ensure that the overall coverage rate meets the accuracy requirements. Third, the area division algorithm in the existing unmanned boat coverage path planning method can only provide equal-area area division, and cannot achieve proportional division of the mission area, and has poor scalability. Summary of the invention

[0004] In view of this, the purpose of the present invention is to propose a coverage path planning method for a large-scale unmanned boat group in an arbitrary polygonal area, so as to solve the technical problem that the existing coverage path planning method cannot be directly applied to polygonal mission areas.

[0005] The technical means adopted by the present invention are as follows:

[0006] A coverage path planning method for a large-scale unmanned boat group in an arbitrary polygonal area comprises the following steps:

[0007] S1. Select the polygonal task area and create a grid map of the polygonal task area.

[0008] S2. Establish an optimization model, use the optimization model to optimize the grid map, and generate a grid map with higher coverage;

[0009] S3. Use a regional division algorithm to divide the area proportionally on a grid map with higher coverage to assign a mission area to each unmanned boat; use a spanning tree algorithm to generate a coverage path within the mission area of ​​each unmanned boat.

[0010] Furthermore, S1 specifically includes the following steps:

[0011] S11, processing the electronic nautical chart information, packaging the sea and land information generated by the processing and storing it as a local file;

[0012] S12. Create a grid map based on the extracted electronic chart information.

[0013] Furthermore, S11 specifically includes the following steps:

[0014] The sea and land information and longitude and latitude information of the polygonal mission area are obtained from the electronic nautical chart, wherein the sea and land information includes data of the passable area or obstacle area at each longitude and latitude coordinate position; the sea and land information is packaged and stored as a local file.

[0015] Furthermore, S12 specifically includes the following steps:

[0016] S121. Select any polygonal task area in the electronic nautical chart as the task area, and select an obstacle area in the polygonal task area. The vertex coordinates of the polygonal task area and the vertex coordinates of the obstacle area are expressed in the format of longitude and latitude coordinates, as shown below:

[0017] PMA={(X1,Y1),...(X r ,Y r )},

[0018] W={{(X α1|1 ,Y α1|1 ),...(X α1|β1 ,Y α1|β1 )},...{(X αp|1 ,Y αp|1 ),...(X αp|pu ,Y αp|βq )}},

[0019] Where PMA represents the polygonal task area, X and Y are the vertex coordinates of the polygonal task area, r is the number of vertices in the polygonal task area; W represents the obstacle area, X αi1|βi2 and Y αi1|βi2 is the vertex coordinate of the obstacle area, α i1represents the i1th obstacle area, i1=1,...p, β i2 Represents the number of vertices in each obstacle area, i2=1,...q, p is the maximum number of obstacle areas, q is the maximum number of vertices in the obstacle area;

[0020] S122, establishing a grid map according to the vertex coordinate information of the polygonal task area, wherein the vertex coordinates of the grid map can be expressed as:

[0021] Grid = {(X min ,Y min ),(X max ,Y min ),(X min ,Y max ),(X max ,Y max )}

[0022] Among them, Grid represents a grid map, X max , X min Indicates the maximum and minimum horizontal coordinates of the polygonal task area, Y max , Y min Indicates the maximum and minimum ordinates of the polygonal task area;

[0023] S123, generating m×n grids with a side length of σ in the grid map according to the set planning accuracy;

[0024]

[0025] Among them, σ is the planning accuracy;

[0026] The grid includes three states: navigable area, obstacle area and initial position of the unmanned boat.

[0027] Furthermore, S2 specifically includes the following steps:

[0028] S21, establishing an initial optimization model and defining the initial optimization model;

[0029] S22. Use a simulated annealing algorithm to solve the defined initial optimization model to obtain an optimization model.

[0030] Furthermore, S21 specifically includes the following steps:

[0031] Define the translation amount of the grid map in the X-axis direction as L x , the range is [0, σ]; the Y-axis translation of the polygon is defined as L y , the range is [0, σ]; define the rotation angle of the grid map as L θ , the range is [0, 90°];

[0032] Define the optimization model Q, expressed as:

[0033] Q=λ1·Q1+λ2·Q2-λ3·Q3

[0034] Among them, Q is composed of three normalized Q i The term is composed and λ1, λ2 and λ3 are used as constants to adjust each term, and the constants take values ​​in the following range, namely:

[0035]

[0036] λ1+λ2=1

[0037] The Q1 term is defined to represent the basic goal of the entire optimization process, which is to control the number of grids in the navigable area within the polygonal mission area, as follows:

[0038]

[0039] Where δ represents the number of grids in the navigable area within the polygonal mission area; S PWA represents the area of ​​the polygon, then the theoretical maximum number of grids in the navigable area within the polygon mission area is After region selection, S PWA and σ are both constants. When the Q1 term is maximized, the number of navigable area grids placed in the polygon is also maximized;

[0040] The Q2 term is defined to represent the area ratio of the polygonal task area to its enhanced area, that is, to achieve the optimal placement of the grid within the polygonal task area, thereby increasing the coverage effect of the edge area of ​​the polygonal task area and improving the coverage rate; the definition of Q2 is as follows:

[0041]

[0042] The vertex coordinates of the enhanced region of the polygonal task area are defined as follows:

[0043]

[0044] EBA={(X * min ,Y * min ),(X * max ,Y * min ),(X * min ,Y * max ),(X * max ,Y *max )}

[0045] Among them, S u represents the area of ​​the enhanced bounding box, {X * max ,X * min ,Y * max ,Y * min} are the vertices of the enhanced bounding box, EBA represents the enhanced bounding box; after region selection, S PWA is a constant, and S is minimized by maximizing the Q2 term u ; Q2 controls the rotation angle L θ Achieve optimal grid placement;

[0046] The Q3 item is defined to control the absolute difference between the edge of the polygonal mission area and the enhanced bounding box in each direction, that is, to achieve the centering of the navigable grid with respect to the polygonal mission area. Q3 is defined as follows:

[0047]

[0048] Q3 is controlled by L x and L y , so that the grid is centered.

[0049] Furthermore, S22 specifically includes the following steps:

[0050] Set the initialization temperature T and the minimum temperature T min , temperature drop rate α, number of iterations K, planning accuracy σ, proportional coefficients λ1, λ2, λ3, constant parameter;

[0051] Create an initial solution cur (L x ,L y ,L θ ), for sol cur Initialize L x =σ,L y =σ,L θ =0;

[0052] for sol cur Create a new state and solve the current optimization coefficient Q cur =λ1Q1+λ2Q2+λ3Q3;

[0053] When T>T min , when i<(K-1), obtain the optimization index Q of the current solution cur =λ1Q1+λ2Q2+λ3Q3;

[0054] If Q cur >Q max , update the maximum optimization index Q max Q cur ;

[0055] When i=K-1, a new state is randomly generated for the new solution, namely:

[0056] L xnew =L xcur +random(0,1)*2*σ

[0057] L ynew =L ycur +random(0,1)*2*σ

[0058] L θnew =L θcur +random(0,1)*90

[0059] Otherwise, a neighboring state is created for the new solution, namely:

[0060] when hour,

[0061]

[0062] otherwise,

[0063]

[0064] when hour,

[0065]

[0066] otherwise,

[0067]

[0068] when hour,

[0069] L θnew =L θcur +random(0,1)*45

[0070] otherwise,

[0071] L θnew =L θcur -random(0,1)*45

[0072] Calculate the probability of acceptance

[0073] When P>random(0,1), Q max Updated to Qnew , update the temperature T = T * α; repeat the above process until the temperature T drops to T min and below.

[0074] Furthermore, S3 specifically includes the following steps:

[0075] S31. Allocate a mission area to each unmanned boat in proportion;

[0076] S32: Generate a coverage path in the mission area of ​​each unmanned boat.

[0077] Furthermore, S31 specifically includes the following steps:

[0078] Proportional region division is performed based on the region division algorithm, where the region division ratio is defined by the user. The region division algorithm formula is as follows:

[0079]

[0080] Among them, H i represents the area proportion of the i-th unmanned boat, and n is the number of unmanned boats;

[0081] The optimized area division algorithm allocates each sub-area to the corresponding unmanned boat after dividing the area, and different sub-areas ensure the generation of non-intersecting paths, so that multiple unmanned boats can operate at the same time without conflict.

[0082] Further, S32 specifically includes the following steps:

[0083] After the task area is divided, a coverage path is generated in each sub-area of ​​the task area; the spanning tree coverage algorithm generates a minimum spanning tree in each sub-area, and then generates a coverage path around each minimum spanning tree; the generated path is as follows:

[0084] path all ={{(X m1|1 ,Y m1|1 ),...(X m1|t1 ,Y m1|t1 )},...{(X mn|1 ,Y mn|1 ),...(X mn|tz ,Y mz|tz )}},

[0085] Among them, path all represents the path coverage of the unmanned boat, t represents the number of waypoints of each unmanned boat, and m represents the path.

[0086] Compared with the prior art, the present invention has the following advantages:

[0087] First, compared with the existing unmanned boat coverage path planning method, the present invention creates a grid map through the vertex coordinate information of any polygon, so that it can efficiently cover the rectangular task area and can more flexibly adapt to the polygonal task area, which significantly improves the flexibility and applicability in practical applications.

[0088] Second, compared with the existing unmanned boat coverage path planning method, the present invention establishes an optimization model based on a grid map, realizes the optimal placement of nodes by rotating and translating the grid map, and uses a simulated annealing algorithm to solve the optimal rotation angle and translation amount, and finally obtains a grid map that is more conducive to coverage, thereby generating a coverage path with higher coverage and fewer turns.

[0089] Third, compared with the existing unmanned boat coverage path planning algorithm, the present invention divides the unmanned boat mission area in proportion and then generates a coverage path of corresponding proportion, which ultimately makes it easier for users to simultaneously operate heterogeneous unmanned boats with different energy levels in the same mission. BRIEF DESCRIPTION OF THE DRAWINGS

[0090] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0091] Figure 1 It is a flow chart of the present invention.

[0092] Figure 2 A schematic diagram of a grid map of any polygon is established for the present invention.

[0093] Figure 3 Schematic diagram of the grid map enhanced bounding box of the present invention.

[0094] Figure 4 This is a schematic diagram of the present invention before node optimization placement.

[0095] Figure 5 This is a schematic diagram of the node after optimized placement of the present invention.

[0096] Figure 6 This is a simulation diagram of the coverage path of the unmanned boat group area ratio division of the present invention.

[0097] Figure 7 This is a simulation diagram of the coverage path of a single unmanned boat of the present invention.

[0098] Figure 8 This is a simulation diagram of the coverage path of a large-scale unmanned boat group of the present invention. DETAILED DESCRIPTION

[0099] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0100] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0101] like Figure 1 As shown, the present invention provides a coverage path planning method for a large-scale unmanned boat group in an arbitrary polygonal area, comprising the following steps:

[0102] A. Electronic chart information processing

[0103] Electronic nautical charts are digital nautical charts that describe maritime geographic information and navigation information in digital form. The sea and land information and longitude and latitude information of the mission area are obtained from the electronic nautical chart. The sea and land information contains the data of the passable area or obstacle area at each longitude and latitude coordinate position. After the sea and land information is packaged, it is stored as a local file.

[0104] B. Grid map creation

[0105] First, select any polygonal mission area in the electronic nautical chart, and select the obstacle area in the polygonal mission area. The vertex coordinates of the polygonal mission area and the vertex coordinates of the obstacle area are expressed in the format of the WGS 84 coordinate system, as shown below:

[0106] PMA={(X1,Y1),...(X r ,Y r )},

[0107]

[0108] Where PMA represents the polygonal mission area, (X i ,Y i )(i=1,...r) are the vertex coordinates of the polygonal task area, r is the number of vertices in the polygonal task area; W represents the obstacle area, is the vertex coordinate of the obstacle area, α i (i=1,...p) represents the i-th obstacle area, β i (i=1,...q) represents the number of vertices in each obstacle area.

[0109] Secondly, a grid map is established based on the vertex information of the polygonal task area. The vertex coordinates of the grid map can be expressed as:

[0110] Grid = {(X min ,Y min ),(X max ,Y min ),(X min ,Y max ),(X max ,Y max )}

[0111] Grid represents a grid map, X max , X min Indicates the maximum and minimum horizontal coordinates of the polygonal task area, Y max , Y min Indicates the maximum and minimum vertical coordinates of the polygonal task area.

[0112] Finally, according to the set planning accuracy, m×n grids with a side length of σ are generated in the grid map.

[0113]

[0114] Among them, σ is the planning accuracy.

[0115] The grid has three states: navigable area, obstacle area and initial position of the unmanned boat.

[0116] C. Grid map optimization based on simulated annealing algorithm

[0117] C1. Optimization model establishment

[0118] The optimization strategy of the present invention is to rotate and move the grid map so that the polygonal mission area will contain more navigable areas, thereby effectively improving the coverage rate.

[0119] Define the translation amount of the grid map in the X-axis direction as L x , the range is [0, σ]; the Y-axis translation of the polygon is defined as L y, the range is [0, σ]; define the rotation angle of the grid map as L θ , the range is [0, 90°].

[0120] Define the optimization model Q, expressed as

[0121] Q=λ1·Q1+λ2·Q2-λ3·Q3

[0122] where Q is composed of three normalized Q i The term is composed and λ1, λ2 and λ3 are used as constants to adjust each term, and the constants take values ​​in the following range, namely:

[0123]

[0124] λ1+λ2=1

[0125] The Q1 term is defined to represent the basic goal of the entire optimization process, which is to control the number of grids in the navigable area within the polygonal mission area, as follows:

[0126]

[0127] Where δ represents the number of grids in the navigable area within the polygonal mission area; S PWA represents the area of ​​the polygon. Then the theoretical maximum number of grids in the navigable area within the polygon mission area is After region selection, S PWA Both σ and Q1 are constants, so when the Q1 term is maximized, the number of navigable area cells placed in the polygon is also maximized.

[0128] The Q2 term is defined to represent the area ratio of the polygonal task area to its enhanced area, that is, to achieve the optimal placement of the grid within the polygonal task area, thereby increasing the coverage effect of the edge area of ​​the polygonal task area and improving the coverage rate. The definition of Q2 is as follows:

[0129]

[0130] The vertex coordinates of the enhanced region of the polygonal task area are defined as follows:

[0131]

[0132] EBA={(X * min ,Y * min ),(X * max ,Y * min ),(X * min ,Y *max ),(X * max ,Y * max )}

[0133] Where S u represents the area of ​​the enhanced bounding box, {X * max ,X * min ,Y * max ,Y * min} are the vertices of the enhanced bounding box, and EBA represents the enhanced bounding box. After region selection, S PWA is a constant, and S can be minimized by maximizing the Q2 term u Q2 controls the rotation angle L θ Achieve optimal grid placement.

[0134] The Q3 item is defined to control the absolute difference between the edge of the polygonal mission area and the enhanced bounding box in each direction, that is, to achieve the centering of the navigable grid with respect to the polygonal mission area. Q3 is defined as follows:

[0135]

[0136] Q3 is controlled by L x and L y , so that the grid is centered.

[0137] C2. Simulated annealing algorithm to solve the optimization model

[0138] In order to optimize the node representation and placement given the polygonal task area and planning accuracy, a simulated annealing algorithm is used to find the optimal solution (L x ,L y ,L θ ) array, the specific method is as follows:

[0139] First, set the initialization temperature T and the minimum temperature T min , temperature drop rate α, number of iterations K, planning accuracy σ, proportional coefficients λ1, λ2, λ3, constant And other parameters.

[0140] First create the initial solution sol cur (L x ,L y ,L θ ), for sol cur Initialize L x =σ,L y =σ,L θ =0.

[0141] for sol cur Create a new state and solve the current optimization coefficient Q cur =λ1Q1+λ2Q2+λ3Q3

[0142] When T>T min , when i<(K-1), obtain the optimization index Q of the current solution cur =λ1Q1+λ2Q2+λ3Q3.

[0143] If Q cur >Q max , update the maximum optimization index Q max Q cur .

[0144] When i=K-1, a new state is randomly generated for the new solution. That is:

[0145] L xnew =L xcur +random(0,1)*2*σ

[0146] L ynew =L ycur +random(0,1)*2*σ

[0147] L θnew =L θcur +random(0,1)*90

[0148] Otherwise, a neighboring state is created for the new solution, i.e.

[0149] when hour,

[0150]

[0151] otherwise,

[0152]

[0153] when hour,

[0154]

[0155] otherwise,

[0156]

[0157] when hour,

[0158] L θnew =L θcur +random(0,1)*45

[0159] otherwise,

[0160] L θnew =L θcur -random(0,1)*45

[0161] Calculate the probability of acceptance

[0162] When P>random(0,1), Q max Updated to Q new , update the temperature T = T*α. Repeat the above process (from T>T min The cycle starts at T until the temperature T drops to T min and below.

[0163] D. Area division and path generation

[0164] In this section, the region partitioning algorithm is first improved to achieve proportional region partitioning. After that, the area represented by the previous grid is divided into specific sub-regions according to the proportional area and combined with the spanning tree cover algorithm to provide an independent path for each unmanned boat.

[0165] D1. Regional division

[0166] In the present invention, proportional area division can be performed based on the area division algorithm, where the area division ratio is defined by the user, so as to facilitate the simultaneous operation of heterogeneous unmanned boats with different energy levels in the same mission:

[0167]

[0168] Among them, H i It represents the area proportion of the i-th unmanned boat, and n is the number of unmanned boats.

[0169] The optimized area division algorithm allocates each sub-area to the corresponding unmanned boat after dividing the area, and different sub-areas ensure the generation of non-intersecting paths, so that multiple unmanned boats can operate at the same time without conflict, which helps to improve the safety and efficiency of the entire system.

[0170] D2. Path Generation

[0171] After completing the task area division, a coverage path is generated in each sub-area. The present invention adopts a spanning tree coverage algorithm as a basic method for generating a coverage path. The principle of the spanning tree coverage algorithm is to generate a minimum spanning tree in each sub-area, and then generate a coverage path around each minimum spanning tree. The generated path is as follows:

[0172] path all ={{(X m1|1 ,Y m1|1 ),...(Xm1|t1 ,Y m1|t1 )},...{(X mn|1 ,Y mn|1 ),...(X mn|tz ,Y mz|tz )}},

[0173] where path all represents the path coverage of the unmanned boat, t i (i=1,..z) represents the number of waypoints for each unmanned boat, m i (i=1,...,n) represents the i-th path.

[0174] Example

[0175] The first step is to perform initial settings according to the mission requirements, including determining the number and initial positions of unmanned boats, formulating planning accuracy, selecting the mission area by selecting the vertices of the polygonal mission area, and selecting the obstacle area within the selected mission area.

[0176] The second step is to generate a grid map based on the sea and land information extracted from the electronic chart and the vertex coordinate information of the polygonal mission area and obstacle area. The triangle represents the navigable area, the circle represents the obstacle area, and the star represents the initial position of the unmanned boat (such as Figure 2 ). Next, create an environment model (such as Figure 3 ), and calculate the optimal angle and distance of rotating and translating the grid map through the simulated annealing algorithm to generate a grid map that is more conducive to coverage. (such as Figure 4 , Figure 5 )

[0177] In the third step, the area is divided proportionally through the optimized area division algorithm, a sub-area is assigned to each unmanned boat, and a coverage path is generated for each unmanned boat in combination with the spanning tree coverage method.

[0178] In the present invention, the simulation parameters are selected as follows: planning accuracy is 20 meters, initialization temperature is 1000, minimum temperature is 5, temperature drop rate is 0.9, number of iterations for each temperature is 500 times, λ1, λ2, λ3 are 0.68, 0.32 and 0.25 respectively.

[0179] The simulation results are as follows Figure 6-8 As shown, Figure 6 The coverage paths of the unmanned boat group area division ratio are shown, and the area division ratios of the unmanned boat mission area are set to 10%, 10%, 20%, 20%, and 40% respectively. From the simulation results, it can be seen that the optimized area division algorithm realizes the task area division of different ratios and generates coverage paths of corresponding ratios in the corresponding areas. Figure 7For the coverage path of a single unmanned boat, the number of turns of the unmanned boat trajectory before optimization is 20, and the track area coverage rate is 76.35%. After the introduction of Q1, the number of turns of the trajectory is 16, and the track coverage rate is 81.36%. After the introduction of Q1 and Q2, the number of turns of the trajectory is 16, and the track coverage rate is 83.67%. After the introduction of Q1, Q2 and Q3, the number of turns of the trajectory is 16, and the track coverage rate is 91.42%. From the simulation results, it can be seen that the coverage path planning optimization algorithm reduces the number of turns of the unmanned boat trajectory and improves the path coverage rate of the unmanned boat. Figure 8 The simulation path of a large-scale unmanned boat group based on the optimized coverage path planning algorithm is shown in Figure 1. The number of unmanned boats is 100, the total number of turns of the unmanned boat trajectory before optimization is 1526, and the track area coverage rate is 97.05%. The total number of turns of the optimized trajectory is 1348, and the track coverage rate is 98.32%. It can be seen from the simulation results that the unmanned boat coverage path planning method in the present invention is also applicable to large-scale unmanned boat groups.

[0180] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A coverage path planning method for a large-scale unmanned boat group in an arbitrary polygonal area, characterized in that: The steps include: S1. Select the polygonal task area and create a grid map of the polygonal task area. S2. Establish an optimization model, use the optimization model to optimize the grid map, and generate a grid map with higher coverage; S3, using a region division algorithm to divide the region proportionally on a grid map with a higher coverage rate and assigning a mission area to each unmanned boat; The spanning tree algorithm is used to generate coverage paths within the mission area of ​​each unmanned boat.

2. The coverage path planning method for a large-scale unmanned boat group in an arbitrary polygonal area according to claim 1 is characterized in that: S1 specifically includes the following steps: S11, processing the electronic nautical chart information, packaging the sea and land information generated by the processing and storing it as a local file; S12. Create a grid map based on the extracted electronic chart information.

3. The coverage path planning method for a large-scale unmanned boat group in an arbitrary polygonal area according to claim 2 is characterized in that: S11 specifically includes the following steps: Acquire sea and land information and longitude and latitude information of the polygonal mission area from the electronic nautical chart, wherein the sea and land information includes data of a passable area or an obstacle area at each longitude and latitude coordinate position; The sea and land information is packaged and stored as a local file.

4. The coverage path planning method for a large-scale unmanned boat group in an arbitrary polygonal area according to claim 2 is characterized in that: S12 specifically includes the following steps: S121. Select any polygonal task area in the electronic nautical chart as the task area, and select an obstacle area in the polygonal task area. The vertex coordinates of the polygonal task area and the vertex coordinates of the obstacle area are expressed in the format of longitude and latitude coordinates, as shown below: PMA={(X1,Y1),...(X r ,AND r )}, W={{(X α1|1 ,Y α1|1 ),...(X α1|β1 ,Y α1|β1 )},...{(X αp|1 ,Y αp|1 ),...(X αp|pu ,Y αp|βq )}}, Where PMA represents the polygonal task area, X and Y are the vertex coordinates of the polygonal task area, r is the number of vertices in the polygonal task area; W represents the obstacle area, X αi1|βi2 and Y αi1|βi2 is the vertex coordinate of the obstacle area, α i1 represents the i1th obstacle area, i1=1,...p, β i2 Represents the number of vertices in each obstacle area, i2=1,...q, p is the maximum number of obstacle areas, q is the maximum number of vertices in the obstacle area; S122, establishing a grid map according to the vertex coordinate information of the polygonal task area, wherein the vertex coordinates of the grid map can be expressed as: Grid={(X min ,AND min ),(X max ,AND min ),(X min ,AND max ),(X max ,AND max )} Among them, Grid represents a grid map, X max , X min Indicates the maximum and minimum horizontal coordinates of the polygonal task area, Y max , Y min Indicates the maximum and minimum ordinates of the polygonal task area; S123, generating m×n grids with a side length of σ in the grid map according to the set planning accuracy; Among them, σ is the planning accuracy; The grid includes three states: navigable area, obstacle area and initial position of the unmanned boat.

5. The coverage path planning method for a large-scale unmanned boat group in an arbitrary polygonal area according to claim 1 is characterized in that: S2 specifically includes the following steps: S21, establishing an initial optimization model and defining the initial optimization model; S22. Use a simulated annealing algorithm to solve the defined initial optimization model to obtain an optimization model.

6. The coverage path planning method for a large-scale unmanned boat group in an arbitrary polygonal area according to claim 5 is characterized in that: S21 specifically includes the following steps: Define the translation amount of the grid map in the X-axis direction as L x , the range is [0, σ]; the Y-axis translation of the polygon is defined as L y , the range is [0, σ]; define the rotation angle of the grid map as L θ , the range is [0, 90°]; Define the optimization model Q, expressed as: Q=λ1·Q1+λ2·Q2-λ3·Q3 Among them, Q is composed of three normalized Q i The term is composed and λ1, λ2 and λ3 are used as constants to adjust each term, and the constants take values ​​in the following range, namely: λ1+λ2=1 The Q1 term is defined to represent the basic goal of the entire optimization process, which is to control the number of grids in the navigable area within the polygonal mission area, as follows: Where δ represents the number of grids in the navigable area within the polygonal mission area; S PWA represents the area of ​​the polygon, then the theoretical maximum number of grids in the navigable area within the polygon mission area is After region selection, S PWA and σ are both constants. When the Q1 term is maximized, the number of navigable area grids placed in the polygon is also maximized; The Q2 term is defined to represent the area ratio of the polygonal task area to its enhanced area, that is, to achieve the optimal placement of the grid within the polygonal task area, thereby increasing the coverage effect of the edge area of ​​the polygonal task area and improving the coverage rate; the definition of Q2 is as follows: The vertex coordinates of the enhanced region of the polygonal task area are defined as follows: EBA={(X * min ,AND * min ),(X * max ,AND * min ),(X * min ,AND * max ),(X * max ,AND * max )} Among them, S u represents the area of ​​the enhanced bounding box, {X * max ,X * min ,Y * max ,Y * min } are the vertices of the enhanced bounding box, EBA represents the enhanced bounding box; after region selection, S PWA is a constant, and S is minimized by maximizing the Q2 term u ; Q2 controls the rotation angle L θ Achieve optimal grid placement; The Q3 item is defined to control the absolute difference between the edge of the polygonal mission area and the enhanced bounding box in each direction, that is, to achieve the centering of the navigable grid with respect to the polygonal mission area. Q3 is defined as follows: Q3 is controlled by L x and L y , so that the grid is centered.

7. The coverage path planning method for a large-scale unmanned boat group in an arbitrary polygonal area according to claim 5, characterized in that: S22 specifically includes the following steps: Set the initialization temperature T and the minimum temperature T min , temperature drop rate α, number of iterations K, planning accuracy σ, proportional coefficients λ1, λ2, λ3, constant parameter; Create an initial solution cur (L x ,L y ,L θ ), for sol cur Initialize L x =σ,L y =σ,L θ =0; for sol cur Create a new state and solve the current optimization coefficient Q cur =λ1Q1+λ2Q2+λ3Q3; When T>T min , when i<(K-1), obtain the optimization index Q of the current solution cur =λ1Q1+λ2Q2+λ3Q3; If Q cur >Q max , update the maximum optimization index Q max Q cur ; When i=K-1, a new state is randomly generated for the new solution, namely: L xnew =L xcur +random(0,1)*2*σ L ynew =L ycur +random(0,1)*2*σ L θnew =L θcur +random(0,1)*90 Otherwise, a neighboring state is created for the new solution, namely: when hour, otherwise, when hour, otherwise, when hour, L θnew =L θcur +random(0,1)*45 otherwise, L θnew =L θcur -random(0,1)*45 Calculate the probability of acceptance When P>random(0,1), Q max Updated to Q new , update temperature T = T*α; jump to the step of obtaining the optimization index of the current solution until the temperature T drops to T min and below.

8. The coverage path planning method for a large-scale unmanned boat group in an arbitrary polygonal area according to claim 1 is characterized in that: S3 specifically includes the following steps: S31. Allocate a mission area to each unmanned boat in proportion; S32: Generate a coverage path in the mission area of ​​each unmanned boat.

9. The coverage path planning method for a large-scale unmanned boat group in an arbitrary polygonal area according to claim 8, characterized in that: S31 specifically includes the following steps: Proportional region division is performed based on the region division algorithm, where the region division ratio is defined by the user. The region division algorithm formula is as follows: Among them, H i represents the area proportion of the i-th unmanned boat, and n is the number of unmanned boats; The optimized area division algorithm allocates each sub-area to the corresponding unmanned boat after dividing the area, and different sub-areas ensure the generation of non-intersecting paths, so that multiple unmanned boats can operate at the same time without conflict.

10. The coverage path planning method for a large-scale unmanned boat group in an arbitrary polygonal area according to claim 8, characterized in that: S32 specifically includes the following steps: After the task area is divided, a coverage path is generated in each sub-area of ​​the task area; the spanning tree coverage algorithm generates a minimum spanning tree in each sub-area, and then generates a coverage path around each minimum spanning tree; the generated path is as follows: path all ={{(X m1|1 ,Y m1|1 ),...(X m1|t1 ,Y m1|t1 )},...{(X mn|1 ,Y mn|1 ),...(X mn|tz ,Y mz|tz )}}, Among them, path all represents the path coverage of the unmanned boat, t represents the number of waypoints of each unmanned boat, and m represents the path.