Method and system for planning full-coverage path of kelp harvesting ship and medium

Abstract

The invention provides a full-coverage path planning method and system for a kelp harvesting ship and a medium, and belongs to the technical field of path planning, and the method comprises the steps: constructing a two-dimensional planning drawing, and determining a plurality of parallel operation lines in each sub-region and all angular points in boundary endpoints; taking all angular points of each sub-region as backtracking points; performing traversal path planning on one of the sub-regions through a sub-region internal traversal method to obtain a full-coverage path traversal result of the sub-region and an end point of the full-coverage path traversal result; according to the end point of the sub-region and the backtracking points of the remaining sub-regions, an improved A * algorithm is adopted to carry out connection route planning, and the backtracking point where the starting point of the next to-be-planned sub-region is located is determined; and through multiple iterations until all the backtracking points are traversed, determining a joint route plan and all sub-region full-coverage path traversal result. According to the method, through a path planning mode of regional traversal, inter-region connection and full-region coverage, reasonable connection between regions is achieved, and efficient full-coverage operation of the kelp harvesting ship is achieved.

Description

Technical Field

[0001] This invention relates to the field of path planning technology, specifically to a method, system, and medium for full-coverage path planning for kelp harvesting vessels. Background Technology

[0002] In kelp farming environments, the cultivation methods are very similar to those used in field planting, both employing a row-shaped, covered cultivation method.

[0003] Currently, the two key challenges in route planning for kelp harvesting vessels are sub-region traversal and sub-region connection. It is particularly important to note that, unlike ordinary route planning, when the vessel changes course, the turning radius must be carefully considered. Only by ensuring that the vessel's turning trajectory is greater than the minimum turning radius can the smoothness of continuous operation be guaranteed.

[0004] The current traditional reciprocating process route using sub-region traversal is used, and the sub-region connection route is judged based on experience. The problems that have been exposed are as follows: there is a specific correlation between row spacing and the turning radius of the hull. When the relationship is in a certain state, the relationship between the two is unbalanced, which causes the hull to travel in a pear shape at the turning point. This not only makes the harvesting path longer and the turning more difficult, but also makes the ship spend too much time at the turning point. Summary of the Invention

[0005] To address the aforementioned issues, this invention provides a full-coverage path planning method for kelp harvesting vessels. This method achieves efficient full-coverage operation for kelp harvesting vessels through a technical route of regional traversal, inter-regional connection, and full-region coverage.

[0006] To achieve the above objectives, the present invention provides the following technical solution.

[0007] A method for full-coverage route planning for kelp harvesting vessels includes the following steps: Based on the layout of the kelp farming area, a two-dimensional planning map is constructed by dividing it into multiple rectangular sub-regions. Multiple parallel operation rows and all corner points at the boundary endpoints of each sub-region are determined. All corner points of each sub-region are used as backtracking points. One sub-region is traversed using a sub-region traversal method to plan a traversal path, obtaining the full-coverage path traversal result for that sub-region and its endpoint; wherein, the sub-region traversal method is based on the working row width W in the sub-region and the minimum turning radius R of the kelp harvesting vessel. min Iterate through the size of W, and when W≥2R, min When using the traditional reciprocating traversal method, the line break and reversal method has a radius of R. min The arc shape; when R min <W<2R min When using a line-by-line traversal, the method of changing direction after a line break is used, with a radius of R. minThe arc shape, in which a non-cross-line traversal line break and turn is pear-shaped; Based on the endpoint of the sub-region and the backtracking point of the remaining sub-regions, the improved A* algorithm is used to plan the connecting route and determine the backtracking point of the starting point of the next sub-region to be planned. The improved A* algorithm adds a smoothing optimization mechanism to the original A* algorithm, including bidirectional deletion of redundant nodes, obstacle distance judgment, and bidirectional smoothness optimization. By iterating multiple times until all backtracking points are traversed, the connecting route planning and the result of traversing the full coverage path of all sub-regions are determined.

[0008] Preferably, when R min <W<2R min When using a line-by-line traversal, the method of changing direction after a line break is used, with a radius of R. min The arc shape, wherein a single non-cross-line traversal line break is pear-shaped, wherein the arc is symmetrical with two radii of 90° each, R. min The pear shape is composed of arcs and straight line segments; the pear shape consists of three segments with a radius of R. min It is composed of smooth, circular arc connections.

[0009] Preferably, the improved A* algorithm adds a smoothing optimization mechanism to the original A* algorithm, which includes bidirectional removal of redundant nodes, obstacle distance judgment, and bidirectional smoothness optimization. Specifically, this includes: While maintaining a safe distance from obstacles, delete midpoints on the same straight line in the path nodes, and retain the starting point, inflection points and the ending point; Between every two retained nodes, take one node at a preset step size, determine whether there are obstacles on the path between the two nodes retained in the first optimization and the two newly generated nodes in the second optimization. If there are no obstacles, calculate the distance between the obstacles next to the path and the path, and use the size relationship to determine whether to select the current node as the path node. Reverse the point selection process and, using the secondary optimization method, perform judgments and calculations from the target point back to the starting point.

[0010] Preferably, the improved A* algorithm optimizes the search point selection mechanism during the generation of adjacent nodes in the original A* algorithm, specifically including: Based on the angle between the current position and the target position α Identify the five closest directions and discard the search in three directions.

[0011] Preferably, the improved A* algorithm further includes optimization of the evaluation function, specifically including: Construct the evaluation function: ; In the formula, h ( n This adds to the estimated cost value in the early stages of the search;p For obstacle density, r This represents the distance from the current node to the target point. R Let be the distance from the starting point to the target point; where, when the current node's position is close to the endpoint, let the actual value be... g ( n (Approximately estimated cost) h ( n Size; when far from the target point, increase the estimated cost. h ( n When the number of obstacles increases, the estimated cost decreases. h ( n ).

[0012] Preferably, the improved A* algorithm includes a safety threshold set during the generation of adjacent nodes in the original A* algorithm, specifically including: The safety threshold is set as the safe distance between the moving path and the vertices of static obstacles; Assuming the current node is n1, connect n1 and n2; if the distance from line segment n1n2 to the vertex of the static obstacle is less than the safety threshold, delete the node and select the next node, until the distance from the line segment connecting the previous node and the current node to the vertex of the static obstacle is greater than or equal to the safety threshold.

[0013] This invention also proposes a full-coverage path planning system for kelp harvesting vessels, the system comprising: The map building module is used to construct a two-dimensional planning map by dividing the kelp farming area into multiple rectangular sub-regions based on the layout of the area. It determines multiple parallel work rows and all corner points at the boundary endpoints of each sub-region; and uses all corner points of each sub-region as backtracking points. The path planning module is used to plan a traversal path for a sub-region using a sub-region traversal method, obtaining a full-coverage path traversal result for that sub-region and its endpoint; wherein, the sub-region traversal method is based on the working row width W in the sub-region and the minimum turning radius R of the kelp harvesting vessel. min Iterate through the size of W, and when W≥2R, min When using the traditional reciprocating traversal method, the line break and reversal method has a radius of R. min The arc shape; when R min <W<2R min When using a line-by-line traversal, the method of changing direction after a line break is used, with a radius of R. minThe algorithm is an arc shape, where the line break and reversal method of a non-cross-line traversal is a pear shape; it is also used to plan the connecting route using an improved A* algorithm based on the endpoint of the sub-region and the backtracking points of the remaining sub-regions, and to determine the backtracking point where the starting point of the next sub-region to be planned is located; the improved A* algorithm adds a smoothing optimization mechanism to the original A* algorithm, including bidirectional deletion of redundant nodes, obstacle distance judgment, and bidirectional smoothness optimization; through multiple iterations until all backtracking points are traversed, the connecting route planning and the result of full coverage path traversal of all sub-regions are determined.

[0014] The present invention also proposes a computer-readable storage medium storing a data processing program, which, when executed by a processor, implements the kelp harvesting vessel full-coverage path planning method.

[0015] The present invention also proposes a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the kelp harvesting vessel full-coverage path planning method.

[0016] The beneficial effects of this invention are: This invention proposes a full-coverage path planning method for kelp harvesting vessels. This method utilizes an improved A* algorithm combined with a cross-row reciprocating approach. Regarding the sub-region traversal method, by analyzing the harvesting operation path, the influence of parameters such as the vessel's turning radius and the size of the operating area on sub-region path planning was discovered. This led to the determination of a sub-region reciprocating path planning method, effectively reducing path distance, lowering turning difficulty, and significantly reducing the time spent by the vessel at turns due to fewer turning operations.

[0017] In terms of sub-region connection algorithms, this invention proposes an improved A* algorithm. By adjusting and updating the algorithm mechanisms such as adding bidirectional redundant node deletion, obstacle distance judgment and bidirectional smoothness optimization, search point selection mechanism and evaluation function optimization, it effectively solves path planning problems such as tortuous paths, low path coverage and high repetitive operation rate. Attached Figure Description

[0018] Figure 1 This is a flowchart of a method according to an embodiment of the present invention; Figure 2 This is a map coordinate analysis diagram according to an embodiment of the present invention; Figure 3 These are schematic diagrams illustrating three reciprocating traversal methods according to embodiments of the present invention, wherein... Figure 3 (a) in the text represents the traditional reciprocating type. Figure 3 (b) in the equation represents a repetitive expression spanning one row. Figure 3 (c) in the equation represents a reciprocal expression spanning two rows; Figure 4 These are schematic diagrams illustrating three U-turns according to embodiments of the present invention, wherein, Figure 4 (a) in the text is pear-shaped. Figure 4 (b) is a semicircle. Figure 4 (c) in the diagram represents a bow-shaped turn; Figure 5 These are three methods for obtaining a roadmap according to embodiments of the present invention, wherein, Figure 5 (a) in the text represents the traditional reciprocating type. Figure 5 (b) in the equation represents a repetitive expression spanning one row. Figure 5 (c) in the equation represents a reciprocal expression spanning two rows; Figure 6 This is a schematic diagram illustrating the trend of path length traversal using three methods according to embodiments of the present invention; Figure 7 This is a schematic diagram of the search node in an embodiment of the present invention; Figure 8 This describes the workflow of the backtracking mechanism in this embodiment of the invention. Figure 9 This is a diagram showing the first set of test results according to an embodiment of the present invention. Detailed Implementation

[0019] 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.

[0020] Example 1 This embodiment proposes a method for full-coverage path planning for kelp harvesting vessels. The specific steps are as follows: Figure 1 As shown, it includes: S1: Based on the layout of the kelp farming area, a two-dimensional planning map is constructed by dividing it into multiple rectangular sub-regions. Multiple parallel work rows and all corner points at the boundary endpoints of each sub-region are determined; all corner points of each sub-region are used as backtracking points.

[0021] S2: For one sub-region, perform path planning using the sub-region traversal method to obtain the full-coverage path traversal result and its endpoint for that sub-region; where the sub-region traversal method is based on the working row width W and the minimum turning radius R of the kelp harvesting vessel within the sub-region. min Iterate through the size of W, and when W≥2R, min When using the traditional reciprocating traversal method, the line break and reversal method has a radius of R. min The arc shape; when R min <W<2R min When using a line-by-line traversal, the method of changing direction after a line break is used, with a radius of R. minThe arc shape, in which a non-cross-line traversal's line break and reversal method is pear-shaped.

[0022] S3: Based on the endpoint of the sub-region and the backtracking point of the remaining sub-regions, the improved A* algorithm is used to plan the connecting route and determine the backtracking point of the starting point of the next sub-region to be planned; the improved A* algorithm adds a smoothing optimization mechanism to the original A* algorithm, including bidirectional deletion of redundant nodes, obstacle distance judgment, and bidirectional smoothness optimization.

[0023] S4: Through multiple iterations until all backtracking points are traversed, determine the connecting route planning and the result of the full coverage path traversal of all sub-regions.

[0024] The aquaculture area is the target of the kelp harvesting vessel. Considering the complexity of the actual site environment, it is necessary to conduct on-site measurements to obtain the vertex location information of the aquaculture area in order to obtain its planar coordinates. The aquaculture area is abstracted as a 2D geometric polygon and analyzed in the first quadrant of a Cartesian coordinate system, transforming the path planning problem into a geometric mathematical problem, such as... Figure 2 As shown.

[0025] This example presents three reciprocating path planning methods. The core logic of reciprocating path planning is as follows: within the work area, the agricultural machinery starts working from the boundary endpoint, travels in a straight line along the boundary, turns around upon reaching the boundary, and travels in the opposite direction in a straight line, repeating this cycle until the entire work area has been traversed. This traversal method is particularly advantageous in rectangular plots, ensuring that the work trajectory is complete and without overlap, forming a continuous closed path. This work route generally refers to the work path running along the work rows, with turning angles typically 180 degrees. The work rows are parallel to each other, and the turning stroke occurs at the end of the field, generally representing an empty stroke. The difference between cross-row and traditional reciprocating methods is that a work method where the working strokes are sequentially adjacent is called a traditional reciprocating method, while a work method where the working strokes are not adjacent is called a cross-row reciprocating method. This invention considers three reciprocating traversal path planning methods, such as... Figure 3 As shown: traditional reciprocating (a), reciprocating across one line (b), and reciprocating across two lines (c). The aim is to further reduce path length and turning difficulty by adjusting the intervals between work lines, thereby improving work efficiency while meeting working conditions.

[0026] Furthermore, in S2, this embodiment provides pear-shaped, semi-circular, and bow-shaped turning methods, as illustrated in the diagram. Figure 4 As shown, Figure 4 Images (a), (b), and (c) show schematic diagrams of pear-shaped, semi-circular, and bow-shaped U-turns, respectively. The bow shape is a symmetrical 90° U-turn with two radii of R. min It is composed of arcs and straight line segments; the pear shape consists of three segments with radius R. min It is composed of smooth, circular arc connections.

[0027] This embodiment provides an example where R... min <W<2R min Three simulation results: Minimum turning radius R of kelp harvesting vessel min The map is 4.5 meters long and 50 meters wide, with 10 rows. The row width is R. min Up to 2R min With a scale of 0.1 meters, taking a width W of 4.6 meters as an example, path simulation analysis was performed using three methods: traditional reciprocating (a), reciprocating across one row (b), and reciprocating across two rows (c). The path lengths were then compared, and the results are as follows: Figure 6 As shown.

[0028] Table 1 compares the path lengths of the three traversal methods; Figure 5 Images (a), (b), and (c) show the traversal patterns of three methods: traditional reciprocating, cross-row reciprocating, and cross-two-row reciprocating. Table 1 presents the path length data for these three traversal routes. Figure 5 As shown in Table 1, the reciprocating motion across a single row has the shortest distance and the turning method is simple, mostly in an arc shape. Therefore, this invention proposes that when R... min <W<2R min When using a line-by-line traversal, the method of changing direction after a line break is used, with a radius of R. min The arc shape, in which a non-cross-line traversal's line break and reversal method is pear-shaped.

[0029] This invention selects the A* algorithm to study the connection path planning of kelp harvesting vessels. However, the traditional A* algorithm also has some drawbacks: (1) it searches for too many nodes, which affects the computational efficiency. (2) it crosses obstacle vertices, which does not meet the working conditions. (3) the path obtained by the A* algorithm has redundant collinear nodes and redundant turning points, which makes the walking path discontinuous.

[0030] Therefore, the present invention provides the following optimization method: (1) Search point selection Traditional A* algorithms expand the 8-neighborhood grid of the current node when performing path planning, such as... Figure 7 As shown, the green grid represents the current node, and n1 to n8 represent the eight directions in which the current grid can move.

[0031] Table 2 Search Direction Selection Table; In gridded path search, the traditional eight-neighbor search mechanism wastes computation time and storage space. Let the angle between the line connecting the current position and the target position be... α Then the included angleα The correspondence between the three directions and the discarded directions is shown in Table 2.

[0032] (2) Improvement of evaluation function Consider adding estimated cost value in the early stages of the search. h ( n This improves search efficiency; when the current node's position is close to the endpoint, the actual cost is increased. g ( n ) and estimated cost h ( n When the values ​​are close, the execution efficiency is optimal; when the target point is far away, the estimated cost... h ( n The value should be relatively large to increase search speed; when the number of obstacles is relatively large, the estimated cost should be... h ( n The value should be small to reduce search speed, ensure the accuracy of the found path, and improve security. Based on this, the evaluation function is set as follows:

[0033] ; In the formula, p For obstacle density, r This represents the distance from the current node to the target point. R This represents the distance from the starting point to the target point.

[0034] (3) Path smoothing optimization The core optimization idea is: for any two non-adjacent nodes in the path, if their straight-line distance does not exceed the original planned path length and they do not collide with obstacles, the node between the two nodes can be deleted. This effectively reduces path turns and length, and improves smoothness. The main measures include bidirectional deletion of redundant nodes, obstacle distance judgment, and bidirectional smoothness optimization. The specific optimization content for each of the three stages is as follows:

[0035] (a) Ensure that a safe distance is maintained from obstacles, delete the midpoints on the same straight line in the path nodes, and keep only the starting point, inflection point and ending point.

[0036] (b) Take one node every preset step length between every two retained nodes, and determine whether there are obstacles in the path between the two nodes (retained in the first optimization and newly generated in the second optimization). If there are obstacles, ignore them. If not, calculate the distance between the obstacle next to the path and the path, and use the size relationship to determine whether to select the current node as the path node.

[0037] (c) Take points in reverse and perform judgment and calculation again from the target point to the starting point according to the secondary optimization method.

[0038] (4) Set a safety threshold The safety threshold is the safe distance k between the moving path and the vertex of the static obstacle. Assuming the current node is n1, and n1 and n2 are connected, the distance from line segment n1n2 to the static obstacle must be greater than or equal to the safety threshold k. If the distance from line segment n1n2 to the vertex of the static obstacle is less than the safety threshold k, then the node is deleted and the next node is selected, until the distance from the line segment connecting the previous node and the current node to the vertex of the static obstacle is greater than or equal to the safety threshold k.

[0039] Furthermore, to achieve more accurate full-coverage path planning, this invention establishes a backtracking mechanism. Through a corresponding backtracking point path planning algorithm (improved A*) and a sub-region traversal method, full-coverage operation of the kelp harvesting vessel is achieved. Backtracking points refer to those uncovered points. The selection principle for backtracking points is: grouping backtracking points within the same sub-region, and then using the grouping backtracking principle combined with the Euclidean distance formula to find backtracking points. The function of backtracking points is to connect covered sub-regions with uncovered sub-regions.

[0040] The path switching between covered sub-regions serves as the starting point for subsequent sub-regions. It's important to note that the sub-region traversal method used in this invention is a reciprocating traversal; therefore, the backtracking point needs to be located at the corner of each sub-region to ensure efficient coverage by the kelp harvesting vessel within the sub-regions. The workflow of the backtracking mechanism is as follows: Figure 8 The harvesting vessel traverses each sub-region point by point. Only when the harvesting vessel has completely traversed all backtracking points in the backtracking list for that group can it enter other sub-regions for traversal. This mechanism not only satisfies the requirement of no omissions in the operation, but also optimizes the connection path through the grouped backtracking strategy.

[0041] To verify the effectiveness of the environmental modeling method and path planning algorithm proposed in this invention, a full-coverage path planning simulation software for kelp harvesting vessels was developed based on the MATLAB platform.

[0042] Nine sets of typical kelp farming environment parameters were selected, and the simulation lengths of the traditional reciprocating and cross-row reciprocating methods were compared. The results are shown in Table 3. Table 3 shows that the advantages of the cross-row reciprocating traversal path length are fully verified, as well as the effectiveness of the improved A* algorithm in improving the shortness, smoothness, and safety of cross-regional connection paths, confirming the practical value and application potential of the algorithm improvement.

[0043] Table 3. Simulation results of full coverage traversal; The results of the first group of experiments are shown in the following figure. Figure 9 As shown in the figure, the black line represents the connection route (improved A*), and the blue, purple, red and green lines are the traversal routes of each sub-region.

[0044] The above is an embodiment of the kelp harvesting vessel full-coverage path planning method provided in this example. Based on the same idea, this embodiment also provides a corresponding kelp harvesting vessel full-coverage path planning system. Specific limitations of the kelp harvesting vessel full-coverage path planning system can be found in the limitations of the kelp harvesting vessel full-coverage path planning method described above, and will not be repeated here. Each module in the above-mentioned kelp harvesting vessel full-coverage path planning system can be implemented entirely or partially through software, hardware, or a combination thereof. Each module can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0045] This embodiment also provides a computer-readable storage medium storing a computer program that can be used to execute the above-described... Figure 1 A method for providing full-coverage route planning for kelp harvesting vessels.

[0046] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0047] The above are merely preferred embodiments of the present invention and are 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 for full-coverage path planning for kelp harvesting vessels, characterized in that, Includes the following steps: Based on the layout of the kelp farming area, a two-dimensional planning map is constructed by dividing it into multiple rectangular sub-regions. Multiple parallel operation rows and all corner points at the boundary endpoints of each sub-region are determined. All corner points of each sub-region are used as backtracking points. One sub-region is traversed using a sub-region traversal method to plan a traversal path, obtaining the full-coverage path traversal result for that sub-region and its endpoint; wherein, the sub-region traversal method is based on the working row width W in the sub-region and the minimum turning radius R of the kelp harvesting vessel. min Iterate through the size of W, and when W≥2R, min When using the traditional reciprocating traversal method, the line break and reversal method has a radius of R. min The arc shape; when R min <W<2R min When using a line-by-line traversal, the method of changing direction after a line break is used, with a radius of R. min The arc shape, in which a non-cross-line traversal line break and turn is pear-shaped; Based on the endpoint of the sub-region and the backtracking point of the remaining sub-regions, the improved A* algorithm is used to plan the connecting route and determine the backtracking point of the starting point of the next sub-region to be planned. The improved A* algorithm adds a smoothing optimization mechanism to the original A* algorithm, including bidirectional deletion of redundant nodes, obstacle distance judgment, and bidirectional smoothness optimization. By iterating multiple times until all backtracking points are traversed, the connecting route planning and the result of traversing the full coverage path of all sub-regions are determined.

2. The method for full-coverage path planning of kelp harvesting vessels according to claim 1, characterized in that, When R min <W<2R min When using a line-by-line traversal, the method of changing direction after a line break is used, with a radius of R. min The arc shape, wherein a single non-cross-line traversal line break is pear-shaped, wherein the arc is symmetrical with two radii of 90° each, R. min The pear shape is composed of arcs and straight line segments; the pear shape consists of three segments with a radius of R. min It is composed of smooth, circular arc connections.

3. The method for full-coverage path planning of kelp harvesting vessels according to claim 1, characterized in that, The improved A* algorithm adds a smoothing optimization mechanism to the original A* algorithm, including bidirectional redundant node removal, obstacle distance judgment, and bidirectional smoothness optimization. Specifically, it includes: While maintaining a safe distance from obstacles, delete midpoints on the same straight line in the path nodes, and retain the starting point, inflection points and the ending point; Between every two retained nodes, take one node at a preset step size, determine whether there are obstacles on the path between the two nodes retained in the first optimization and the two newly generated nodes in the second optimization. If there are no obstacles, calculate the distance between the obstacles next to the path and the path, and use the size relationship to determine whether to select the current node as the path node. Reverse the point selection process and, using the secondary optimization method, perform judgments and calculations from the target point back to the starting point.

4. The method for full-coverage path planning of kelp harvesting vessels according to claim 1, characterized in that, The improved A* algorithm optimizes the search point selection mechanism during the generation of adjacent nodes in the original A* algorithm, specifically including: Based on the angle between the current position and the target position α Identify the five closest directions and discard the search in three directions.

5. The method for full-coverage path planning of kelp harvesting vessels according to claim 3, characterized in that, The improved A* algorithm also includes optimization of the evaluation function, specifically including: Construct the evaluation function: ； In the formula, h ( n This adds to the estimated cost value in the early stages of the search; p For obstacle density, r This represents the distance from the current node to the target point. R Let be the distance from the starting point to the target point; where, when the current node's position is close to the endpoint, let the actual value be... g ( n (Approximately estimated cost) h ( n Size; when far from the target point, increase the estimated cost. h ( n When the number of obstacles increases, the estimated cost decreases. h ( n ).

6. The method for full-coverage path planning of kelp harvesting vessels according to claim 1, characterized in that, The improved A* algorithm incorporates a safety threshold during the generation of adjacent nodes in the original A* algorithm, specifically including: The safety threshold is set as the safe distance between the moving path and the vertices of static obstacles; Assuming the current node is n1, connect n1 and n2; if the distance from line segment n1n2 to the vertex of the static obstacle is less than the safety threshold, delete the node and select the next node, until the distance from the line segment connecting the previous node and the current node to the vertex of the static obstacle is greater than or equal to the safety threshold.

7. A full-coverage path planning system for kelp harvesting vessels, characterized in that, The system includes: The map building module is used to construct a two-dimensional planning map by dividing the kelp farming area into multiple rectangular sub-regions based on the layout of the area. It determines multiple parallel work rows and all corner points at the boundary endpoints of each sub-region; and uses all corner points of each sub-region as backtracking points. The path planning module is used to plan a traversal path for a sub-region using a sub-region traversal method, obtaining a full-coverage path traversal result for that sub-region and its endpoint; wherein, the sub-region traversal method is based on the working row width W in the sub-region and the minimum turning radius R of the kelp harvesting vessel. min Iterate through the size of W, and when W≥2R, min When using the traditional reciprocating traversal method, the line break and reversal method has a radius of R. min The arc shape; when R min <W<2R min When using a line-by-line traversal, the method of changing direction after a line break is used, with a radius of R. min The algorithm is an arc shape, where the line break and reversal method of a non-cross-line traversal is a pear shape; it is also used to plan the connecting route using an improved A* algorithm based on the endpoint of the sub-region and the backtracking points of the remaining sub-regions, and to determine the backtracking point where the starting point of the next sub-region to be planned is located; the improved A* algorithm adds a smoothing optimization mechanism to the original A* algorithm, including bidirectional deletion of redundant nodes, obstacle distance judgment, and bidirectional smoothness optimization; through multiple iterations until all backtracking points are traversed, the connecting route planning and the result of full coverage path traversal of all sub-regions are determined.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a data processing program, which, when executed by a processor, implements the kelp harvesting vessel full-coverage path planning method as described in any one of claims 1 to 6.

9. A computer device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the full-coverage path planning method for kelp harvesting vessels as described in any one of claims 1 to 6.

Images