A path planning method and system for multi-ship cooperative offshore operation

Abstract

The application relates to the technical field of autonomous navigation and control of ships, and particularly discloses a path planning method and system for multi-ship cooperative offshore operation, which comprises the following steps: acquiring parameters of an offshore operation area and operation ships; grouping operation points of the offshore operation area based on a clustering algorithm, and performing double verification through spatial convergence and quantity balance to obtain operation point clustering results; taking the maximum fitness as a target, optimizing operation paths of each operation ship by using an improved genetic algorithm, and generating optimal multi-ship cooperative path planning results through operations such as initial population generation by a greedy strategy, fitness calculation, and parent tournament selection optimization screening. The application realizes cooperative and efficient operation of multi-ship and multi-task points, and is suitable for various ships to carry out offshore cooperative operation, fixed-point feeding in a smart fishing ground and other offshore scenes.

Description

Technical Field

[0001] This invention relates to the field of ship autonomous navigation and control technology, and in particular to a path planning method and system for multi-ship collaborative maritime operations. Background Technology

[0002] Currently, in scenarios such as fixed-point feeding in smart fish farms, there are common problems such as poor coordination among multiple vessels and unreasonable path planning. Existing technologies often adopt the method of a single vessel completing all tasks at each work point sequentially, or simply distributing work points equally among the vessels, without fully considering the coordination and cooperation of multiple vessels. This results in high total energy consumption, excessive total time, and redundant total path length, which seriously affects the efficiency of feeding operations.

[0003] Meanwhile, existing technologies lack a scientific and comprehensive feasibility evaluation index system, making it impossible to accurately assess the merits of route planning schemes and guide their optimization. For example, some schemes only focus on route length, ignoring the impact of energy consumption and time on operational efficiency; while others consider energy consumption but fail to take into account the dynamic impact of the marine environment on ships, resulting in significant discrepancies between evaluation results and actual operational conditions, thus failing to guarantee the feasibility and effectiveness of the planning schemes.

[0004] Therefore, how to construct a comprehensive and feasible evaluation index system and design a path planning method that can achieve multi-ship collaboration, reduce total energy consumption, shorten total time, and reduce total path length has become an urgent technical problem to be solved. Summary of the Invention

[0005] The present invention aims to solve the above-mentioned problems. To this end, the present invention provides a path planning method and system for multi-vessel collaborative maritime operations. By rationally allocating the operational tasks of individual vessels, constructing a reasonable feasibility evaluation index system, and combining an improved path planning algorithm, it can achieve efficient collaborative operations of multiple vessels and multiple task points. It is applicable to various types of vessels carrying out collaborative maritime operations, smart fishing ground fixed-point feeding, and other maritime scenarios.

[0006] This invention provides a path planning method for multi-vehicle cooperative maritime operations, and the technical solution adopted is as follows: including: S1: Obtain the coordinates and environmental parameters of the offshore operation area, as well as the number and performance parameters of the operating vessels; S2: Based on the number of operating vessels and the coordinates of the offshore operating area, the operating points in the offshore operating area are grouped according to the clustering algorithm, and the clustering results are obtained by double verification through spatial convergence and quantity balance. Among them, the double verification of spatial convergence and quantity balance is used to ensure that the operating points are concentrated in the geographic space and that the number of operating points in each cluster is balanced. S3: With the goal of maximizing fitness, an improved genetic algorithm is used to optimize the operation path of each vessel based on the clustering results of the operation points. The optimal multi-vessel collaborative path planning result is generated through a greedy strategy for initial population generation, fitness calculation, parent tournament selection, partial matching crossover to ensure path legality, segmented adaptive mutation, and elite preservation population update mechanism. Among them, fitness is calculated using environmental parameters of the offshore operation area, performance parameters of the operation vessels, and operation path length.

[0007] Furthermore, a dual verification is performed through spatial convergence and quantitative balance, the process of which is as follows: Perform spatial convergence verification on the clusters obtained by grouping. If spatial convergence is not satisfied, redistribute the job points. If spatial convergence is satisfied, perform number balance verification on the clusters. If number balance is not satisfied, perform number balance redistribution. If number balance is satisfied, output the clustering results of the job points.

[0008] Furthermore, step S2 includes: S2.1: Extract the coordinates of the work points in the offshore work area from the coordinates of the offshore work area, perform normalization processing, and obtain the plane rectangular coordinate matrix of the work points; S2.2: From Randomly selected without repetition from each work site 10 work points were selected as the initial cluster centers; S2.3: Traverse all For each work point, calculate its distance to a Cartesian coordinate matrix. The Euclidean distance between the cluster centers is used to assign cluster groups based on the minimum distance. One cluster; S2.4: Calculate the mean coordinates of all operation points in the cluster and denote them as the new cluster centers; S2.5: Perform spatial convergence verification on the clustering. If spatial convergence is not satisfied, proceed to step S2.3. Perform a cluster balance check. If the cluster balance is not met, proceed to step S2.6. S2.6: For clusters where the number of work points exceeds the quantity balance verification threshold, calculate the adjustment priority index for each work point in the cluster and select the work points to be adjusted. From clusters with fewer job points than the average number of job points across all clusters, select the cluster that is closest to the job point to be adjusted as the target cluster; Adjust the work points to be adjusted to the target cluster, update the work point set of each cluster; calculate the coordinates of the cluster center, and return to step S2.3.

[0009] Furthermore, the formula for calculating the priority index of work points is as follows: in, For work points Adjustment priority indicators For work points Euclidean distance to the current cluster center For the first The number of actual operation points in each cluster. The average number of job points across all clusters. This is the distance weighting coefficient.

[0010] Furthermore, step S3 includes: S3.1: Pair the clustering results of the work sites with the work vessels. Based on a greedy strategy, generate multiple gene sequences for each work vessel according to the clustering results of the work sites to obtain the initial population. S3.2: Perform fitness calculation, parent tournament selection, partial matching crossover to ensure path legality, segmented adaptive mutation, elite preservation population update mechanism, and termination judgment operation on the initial population in sequence to calculate the path planning result.

[0011] Furthermore, step S3.2 includes: S3.2.1: Using the environmental parameters of the offshore operation area and the performance parameters of the operating vessel, and based on a multivariate feasibility evaluation index system, calculate the fitness of each gene sequence in the population; and take the maximum value of all fitness values ​​as the optimal fitness of the population. S3.2.2: Randomly select from the entire gene sequence From these, select the one with the highest fitness as the preferred parent; repeat this process until a parent is selected. A preferred parent generation; S3.2.3: In Randomly select one pair from the preferred parent generations, perform partial matching and crossover to obtain offspring; repeat this operation until the desired offspring are obtained. One offspring; S3.2.4: Randomly select two gene positions in the offspring's gene sequence and exchange them to obtain crossover variant offspring; S3.2.5: Select some individuals with high fitness from the parent generation and form a new population with the crossover offspring; S3.2.6: Repeat steps S3.2.1 to S3.2.5. When the change in the optimal fitness of the population over multiple generations is less than the termination threshold, output the gene sequence corresponding to the optimal fitness of the population at this time as the optimal path planning result for the working vessel.

[0012] Furthermore, the fitness is a weighted sum of energy consumption, duration, and path length indicators; The formula for calculating the energy consumption index is as follows: in, For the first Energy consumption of a single operating vessel For the first The length of the operating path of each operating vessel. For the first Energy consumption per unit distance for a single operating vessel in still water; For the first The increase in energy consumption caused by environmental factors of each operating vessel. , This is the coefficient for the increase in ship water flow energy consumption. This is the wind energy consumption increment coefficient. For water flow velocity, Wind speed; The duration indicator is the time from the departure of the working vessel to its return to the starting point; Path length metric uses .

[0013] Furthermore, the formula for calculating fitness is: in, For the first The adaptability of the operating vessels Energy consumption weighting coefficient This is the duration weighting coefficient. This is the path length weighting coefficient; The preset maximum energy consumption, The preset maximum duration, This is the preset maximum path length.

[0014] Furthermore, in step S3.2.4, the probability of mutation occurs. The calculation formula is: in, Based on the probability of mutation, The maximum mutation probability, The fitness of the current gene sequence, This represents the average fitness value of the current generation of the population.

[0015] This invention also provides a path planning system for multi-vehicle collaborative maritime operations, the technical solution of which includes: The data acquisition unit is used to acquire the coordinates and environmental parameters of the offshore operation area, as well as the number and performance parameters of the operating vessels; The work point grouping unit is used to group the work points in the offshore work area based on the number of working vessels and the coordinates of the offshore work area using a clustering algorithm. The clustering results are obtained through dual verification of spatial convergence and quantity balance. The dual verification of spatial convergence and quantity balance is used to ensure that the work points are concentrated in the geographic space and that the number of work points in each cluster is balanced. The multi-vessel collaborative path planning unit is used to optimize the operation path of each vessel based on the clustering results of the operation points, with the goal of maximizing fitness. The optimal multi-vessel collaborative path planning result is generated through a greedy strategy for initial population generation, fitness calculation, parent tournament selection, partial matching crossover to ensure path legality, segmented adaptive mutation, and elite preservation population update mechanism. The fitness is calculated using environmental parameters of the offshore operation area, performance parameters of the operation vessels, and operation path length.

[0016] The above-described one or more technical solutions in the embodiments of the present invention have at least one of the following technical effects: 1. This invention proposes an improved dual-balance clustering algorithm, which divides the work points into spatially concentrated and quantitatively balanced work areas through a dual-condition closed-loop verification and balanced redistribution mechanism of spatial convergence and quantitative balance. This ensures the continuity of the work points of a single ship in geographical space, avoids the energy consumption and time loss caused by long-distance cross-regional navigation, and realizes the balanced distribution of the work load of each ship, laying a reasonable task allocation foundation for multi-ship collaborative operations.

[0017] 2. This invention constructs a multi-dimensional feasibility evaluation index system that includes energy consumption, duration, and path length. It not only considers the basic index of path length, but also incorporates the energy consumption (including energy consumption increments caused by environmental factors) and operation duration (including travel time and operation stay time) in the actual operation of the ship. By calculating the comprehensive evaluation score through weighted normalization, it realizes a comprehensive, accurate, and quantitative evaluation of the path planning scheme.

[0018] 3. This invention makes multi-dimensional targeted improvements to the traditional genetic algorithm, designing a comprehensive improvement strategy encompassing greedy initial population generation, fitness calculation, tournament-based parent selection, and fitness-based segmented adaptive mutation. The initial population generated by the greedy strategy inherently possesses high-quality path characteristics, accelerating algorithm convergence; tournament selection prevents the algorithm from prematurely converging to local optima; partial matching crossover combined with matching table methods ensures path validity and eliminates invalid solutions; adaptive mutation dynamically adjusts the mutation probability based on individual fitness, balancing population diversity and local search accuracy; and the elite retention mechanism prevents the loss of high-quality genes, ensuring the algorithm iterates towards the optimal solution.

[0019] 4. This invention achieves end-to-end optimization of path planning for multi-vessel collaborative maritime operations through a comprehensive design of "dual-balanced clustering allocation - multi-dimensional evaluation - improved genetic algorithm optimization." The resulting path scheme effectively reduces total energy consumption, shortens total operation time, and reduces total path length, significantly improving the overall efficiency of multi-vessel collaborative operations. This invention is applicable to various multi-vessel collaborative maritime operation scenarios, such as fixed-point feeding in smart fishing grounds. It exhibits strong adaptability to operational area boundaries, vessel performance parameters, and marine environmental parameters. The planning scheme demonstrates high feasibility and practicality, providing direct technical guidance for actual maritime collaborative operations and possessing significant engineering practical value.

[0020] 5. This invention fully considers the actual constraints of offshore operations in its entire process design. It incorporates the dynamic influence of marine environmental parameters such as water flow and wind speed during the energy consumption calculation stage, and combines vessel performance parameters during the path optimization stage, ensuring that the planning scheme always conforms to the actual execution conditions of offshore operations. Furthermore, parameters such as weight coefficients, thresholds, and probabilities in the algorithm can be flexibly adjusted according to actual operational scenarios and needs, adapting to the path planning requirements of different sea areas, different vessels, and different operation types, demonstrating excellent scenario adaptability and engineering flexibility.

[0021] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0023] Figure 1 This is a flowchart of the method provided by the present invention.

[0024] Figure 2 This is a diagram illustrating the task allocation effect of multi-ship collaborative operations at sea.

[0025] Figure 3 This is a schematic diagram illustrating the path planning for multi-vessel collaborative operations at sea. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention. The following embodiments are used to illustrate this invention but should not be used to limit the scope of this invention.

[0027] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0028] The following is combined Figures 1 to 3 The present invention will be further described in detail below, providing a path planning method and system for multi-vehicle cooperative maritime operations: In this embodiment, as Figure 1 As shown, a path planning method for multi-vehicle collaborative maritime operations is provided, including the following steps: S1: Obtain the coordinates and environmental parameters of the offshore operation area, as well as the number and performance parameters of the operating vessels.

[0029] The coordinates of an offshore operation area include the coordinates of the vessel's launch and recovery points, the coordinates of multiple operation points, and the boundary coordinates of the offshore operation area. A common type of offshore operation area is the smart fishing ground.

[0030] Environmental parameters for offshore operations include water flow velocity and wind speed.

[0031] Determine the number of vessels in operation It obtains performance parameters such as energy consumption per unit distance and maximum sailing speed of each operating vessel.

[0032] S2: Based on the number of operating vessels and the coordinates of the offshore operating area, the operating points in the offshore operating area are grouped according to the clustering algorithm, and the clustering results of the operating points are obtained through dual verification of spatial convergence and quantity balance. Among them, the dual verification of spatial convergence and quantity balance is used to ensure that the operating points are concentrated in the geographic space and that the number of operating points in each cluster is balanced.

[0033] This step uses a closed-loop logic of clustering iteration and balance verification to ensure that each cluster simultaneously meets the requirements of spatial convergence and quantity balance, thereby achieving a concentrated distribution of work points in geographical space and a balanced number of work points in each cluster.

[0034] S2.1: Data Standardization. Extract the coordinates (latitude and longitude coordinates) of all operation points in the offshore operation area from the coordinates of the offshore operation area, perform normalization processing, and obtain a Cartesian coordinate matrix of all operation points.

[0035] S2.2: Initialize cluster centers. From all Randomly selected without repetition from each work site 10 work points were selected as the initial cluster centers.

[0036] S2.3: Assign job points. Iterate through all... For each work point, calculate its distance to a Cartesian coordinate matrix. The Euclidean distance between the cluster centers is used to assign cluster groups based on the minimum distance. One cluster.

[0037] S2.4: Update cluster centers. Filter all job points belonging to a cluster, calculate the mean coordinates of these job points, and denote them as the new cluster centers.

[0038] S2.5: Convergence-Balance Dual Verification. The clustering process undergoes spatial convergence and quantity balance verification for task allocation. If both spatial convergence and quantity balance conditions are met, the iteration terminates; otherwise, task points are reassigned.

[0039] Specifically, spatial convergence is checked for clustering. If spatial convergence is not met, proceed to step S2.3 and redistribute job points. If spatial convergence is met, number balance is checked for clustering. If number balance is not met, proceed to step S2.6 and redistribute number points. If number balance is met, terminate the iteration and output the job point clustering results.

[0040] Spatial convergence verification: Calculate the displacement deviation of cluster centers between two iterations. When the displacement deviation is less than or equal to a preset spatial convergence threshold, the cluster centers are considered to have achieved spatial convergence. The criterion formula is as follows: in, The global cluster center displacement deviation between the two iterations is calculated using the squared Euclidean distance. For the first The first cluster of the th cluster The cluster center coordinates obtained in the next iteration For the first The first cluster of the th cluster The cluster center coordinates obtained in the next iteration; The spatial convergence threshold is typically set to

[10] . -6 10 -4 ].

[0041] Number Balance Verification: Number balance is determined by the number deviation rate of each cluster. When the number deviation rate of all clusters is less than or equal to the preset number balance verification threshold, the number balance is considered to be achieved. The criterion formula is as follows: in, For the first Number of clusters bias rate For the first The number of actual operation points in each cluster. The average number of job points across all clusters. , The threshold for quantity balance verification is generally set to [0.1, 0.3].

[0042] S2.6: Balanced redistribution of quantity.

[0043] Specifically, the following steps are included: S2.6.1: Calculation of adjustment priority. Clustering for job points where the number exceeds the quantity balance verification threshold ( The formula for calculating the adjustment priority index for each work point in the over-standard cluster is as follows: in, For work points Adjustment priority indicators For work points Euclidean distance to the current cluster center Characterizing the degree to which the number of out-of-standard clusters exceeds the standard. This is the distance weighting coefficient, with a value range of [0.6, 0.9], used to balance the priority effects of spatial distance and quantity deviation.

[0044] Select the operation point with the highest adjustment priority index from each out-of-standard cluster. One work point was designated as the work point to be adjusted. , To adjust the ratio, it is usually taken as [0.1, 0.2].

[0045] S2.6.2: Target Clustering Matching. Starting from a cluster with fewer than [number of job points]... Clustering ( In the missing clustering, the cluster that is closest to the work point to be adjusted is selected as the target cluster.

[0046] S2.6.3: Job point reassignment. Adjust the job points to be reassigned from the oversaturated clusters to the matching target clusters, and update the job point set for each cluster; then calculate the coordinates of the cluster centers as initial values ​​for the iteration. Return to step S2.3.

[0047] To prevent infinite iterations, this embodiment also sets an upper limit for quantity balancing re-iteration. When the number of times quantity balancing redistribution is executed reaches the upper limit for quantity balancing re-iteration, the quantity balancing verification threshold is increased, for example, by 0.1 each time.

[0048] S3: With the goal of maximizing fitness, based on the clustering results of the work points, an improved genetic algorithm is used to optimize the work path of each work vessel. The optimal multi-vessel collaborative path planning result is generated through a greedy strategy for initial population generation, fitness calculation, parent tournament selection, partial matching crossover to ensure path legality, segmented adaptive mutation, and elite retention population update mechanism. The fitness score is calculated using environmental parameters of the offshore operating area, performance parameters of the operating vessel, and length of the operating path.

[0049] Specifically, the following steps are included: S3.1: Greedy strategy for initial population generation. The clustering results of the work points are paired with the work vessels (the performance differences between work vessels are small and can be ignored). Integer encoding is used, and the work path of each work vessel is represented in the form of a gene sequence. Where 0 is both the starting point and the ending point. The work site is numbered 1. The work site is numbered 2. Assign operation point number 3; the number of gene locations is equal to the number of operation points in the cluster corresponding to this operation vessel. Based on a greedy strategy, generate a clustering result for each operation vessel according to the operation point clustering results. The gene sequences are combined to obtain the initial population.

[0050] S3.1.1: Initial site selection is random. One site is randomly selected from the assigned work sites on this vessel as the initial site for the gene sequence. And update the current gene location to the coordinates of the task point.

[0051] S3.1.2: Greedy sequence update. Select the job point closest to the current gene position from the set of currently unvisited job points, and add that job point to the gene sequence (as part of the gene sequence). The process involves deleting the unvisited job set and updating the current gene position to the coordinates of that job. After traversing all job sets, zeros are added to the end of the gene sequence to form a new gene sequence.

[0052] S3.1.3: Repeat this operation to form a population. Repeat steps S3.1.1 to S3.1.2 to obtain the population formed by... An initial population consisting of a greedy strategy gene sequence This refers to the population size.

[0053] S3.2: Improved genetic operations. The initial population is subjected to fitness calculation, parent tournament selection, partial matching crossover to ensure path legitimacy, segmented adaptive mutation, elite preservation population update mechanism, and termination judgment operation in sequence to calculate the optimal path planning result.

[0054] S3.2.1: Fitness Calculation for Multidimensional Adaptability: Using environmental parameters of the offshore operating area and performance parameters of the operating vessels, and based on a multivariate feasibility evaluation index system, the fitness of each gene sequence in the population (initial population or new population) is calculated. The maximum value of all fitness values ​​is taken as the optimal fitness of the population.

[0055] The construction of a multi-dimensional feasibility evaluation index system. The evaluation index system includes three core indicators: energy consumption, duration, and path length. A weighted summation method is used to calculate the fitness, thereby evaluating the merits of the path planning scheme.

[0056] (1) The formula for calculating the energy consumption index is: in, For the first Energy consumption of a single operating vessel For the first The length of the operation path for each vessel is calculated during path planning based on the coordinates of the operation points corresponding to the gene sequence. For the first Energy consumption per unit distance for a single operating vessel in still water; For the first The increase in energy consumption caused by environmental factors of each operating vessel. ; This is the coefficient for the increase in ship water flow energy consumption. This is the wind energy consumption increment coefficient. and These are input parameters preset based on the sea conditions for the day's operations; For water flow velocity, This refers to wind speed.

[0057] (2) The duration index is the time from the departure of the working vessel to the return starting point, and the calculation formula is: in, For the first The mission duration of each operating vessel For the first The travel time of the operating vessels , For the first The speed of each operating vessel must not exceed its maximum speed. For the first The operating time of each working vessel , For the number of work points, The preset dwell time at the work site.

[0058] (3) The path length index adopts .

[0059] (4) Fitness: The fitness of a gene sequence is calculated by weighted average, and the formula is as follows: in, For the first The adaptability of the operating vessels Energy consumption weighting coefficient This is the duration weighting coefficient. This is the path length weighting coefficient. , , It is an adjustable parameter; The preset maximum energy consumption, The preset maximum duration, The preset maximum path length, , , Used for normalization.

[0060] S3.2.2: Parental tournament selection. From the population... Randomly selected from gene sequences From these candidates, select the one with the highest fitness as the preferred parent in the parent generation tournament. Repeat this process until a parent is selected. A preferred parent; in this embodiment, Desirable This mechanism ensures the priority inheritance of high-quality genes while avoiding premature convergence of the algorithm to a local optimum through random sampling.

[0061] S3.2.3: Partial matching and intersection for path validity assurance. In Randomly select one pair of preferred parents, extract gene sequences from the preferred parent path, randomly select two intersection points for gene exchange, and use a matching table method to eliminate duplicate work points in the gene sequences, ensuring that each work point is unique, thus obtaining offspring. Repeat this operation until the desired offspring are obtained. One offspring; in this embodiment, Also advisable This mechanism ensures a balanced structure where offspring inherit from their parents during the crossover process, and guarantees the feasibility of a unique path sequence, thus avoiding invalid solutions.

[0062] S3.2.4: Fitness-based segmented adaptive mutation. The mutation processing method for segmented adaptive mutation is as follows: two gene positions are randomly selected in the offspring's gene sequence for exchange, resulting in crossover mutated offspring. The probability of mutation occurrence is... The formula for calculating the probability of gene position exchange is: in, The basic mutation probability ranges from 0.02 to 0.05. The maximum mutation probability ranges from 0.08 to 0.12. The fitness of the current gene sequence; This represents the average fitness value of the current generation of the population.

[0063] S3.2.5: Elite Preservation Population Renewal Mechanism. In The top 20% of individuals with the highest fitness among the parent generations were selected as elite parent individuals and directly retained, and then... Each crossover mutation offspring forms a new population, and the population size remains the same. .

[0064] S3.2.6: Termination Judgment. Repeat steps S3.2.1 to S3.2.5. When the change in the optimal fitness of the population over 10 consecutive generations is less than a preset termination threshold (1% in this embodiment), stop the iteration and output the gene sequence corresponding to the optimal fitness of the population at this time, as the optimal path planning result for the working vessel. The change in the optimal fitness of the population is calculated based on the optimal fitness of the population selected in step S3.2.1.

[0065] The optimal path planning results of multiple operating vessels are summarized to obtain the multi-vessel collaborative path planning results.

[0066] This embodiment also verifies the path planning effect of the proposed method.

[0067] In this embodiment, the work area is set to 1A0.0°E-1A0.1°E and A0.0°N-A0.1°N, the launch and retrieval position (starting point) of the work vessel is [1A0.05E, A0.05N], the number of vessels is 4, and 100 work points are randomly generated within the work area.

[0068] Based on the number of vessels operating, the work sites were clustered into 4 groups. Through iterative clustering and load balancing, the final classification results are as follows: the first vessel was assigned 23 work sites, the second vessel 28, the third vessel 28, and the fourth vessel 21. (See attached diagram) Figure 2 As shown.

[0069] In this embodiment, the relevant parameters of the multi-dimensional feasibility evaluation index system are set as follows: Energy consumption weighting coefficient Set to 0.2, duration weighting coefficient Set to 0.2, path length weight coefficient Take 0.6 as the maximum energy consumption. Take 20kWh, maximum duration Take 10h as the maximum path length. Take 50km as the incremental coefficient of ship water flow energy consumption. Take 0.2 as the wind energy consumption increment coefficient. Take 0.1.

[0070] After 64 generations of iterative computation using the genetic algorithm, the optimal result is shown in the appendix. Figure 3 As shown. Detailed specifications for each individual vessel are as follows: The first operating vessel had an energy consumption of 3.83 kWh, a duration of 3.02 hours, a route length of 22.55 km, and an adaptability of 63.07%. The second operating vessel had an energy consumption of 4.56 kWh, a duration of 3.61 hours, a route length of 26.80 km, and an adaptability of 56.06%. The third operating vessel had an energy consumption of 4.68 kWh, a duration of 3.68 hours, a route length of 27.50 km, and an adaptability of 54.96%. The fourth operating vessel had an energy consumption of 3.92 kWh, a duration of 3.01 hours, a route length of 23.08 km, and an adaptability of 62.36%.

[0071] The average energy consumption (average of all ship energy consumption) of the multi-ship collaborative route planning results is 4.2475 kWh, the total duration of all ships (maximum duration of all ships) is 3.68 h, and the average route length of all ships (average route length of all ships) is 24.985 km. Therefore, the fitness (multivariate feasibility evaluation score) of the multi-ship collaborative route planning results is 58.41%. This score can be used to intuitively reflect the quality of the multi-ship collaborative route planning results to users.

[0072] This embodiment also provides a path planning system for multi-vehicle collaborative maritime operations, and the technical solution adopted is as follows: including: The data acquisition unit is used to acquire the coordinates and environmental parameters of the offshore operation area, as well as the number and performance parameters of the operating vessels; The work point grouping unit is used to group the work points in the offshore work area based on the number of working vessels and the coordinates of the offshore work area using a clustering algorithm. The clustering results are obtained through dual verification of spatial convergence and quantity balance. The dual verification of spatial convergence and quantity balance is used to ensure that the work points are concentrated in the geographic space and that the number of work points in each cluster is balanced. The multi-vessel collaborative path planning unit is used to optimize the operation path of each vessel based on the clustering results of the operation points, with the goal of maximizing fitness. The optimal multi-vessel collaborative path planning result is generated through a greedy strategy for initial population generation, fitness calculation, parent tournament selection, partial matching crossover to ensure path legality, segmented adaptive mutation, and elite preservation population update mechanism. The fitness is calculated using environmental parameters of the offshore operation area, performance parameters of the operation vessels, and operation path length.

[0073] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A path planning method for multi-vehicle cooperative maritime operations, characterized in that, include: S1: Obtain the coordinates and environmental parameters of the offshore operation area, as well as the number and performance parameters of the operating vessels; S2: Based on the number of operating vessels and the coordinates of the offshore operating area, the operating points in the offshore operating area are grouped according to the clustering algorithm, and the clustering results are obtained by double verification through spatial convergence and quantity balance. Among them, the double verification of spatial convergence and quantity balance is used to ensure that the operating points are concentrated in the geographic space and that the number of operating points in each cluster is balanced. S3: With the goal of maximizing fitness, an improved genetic algorithm is used to optimize the operation path of each vessel based on the clustering results of the operation points. The optimal multi-vessel collaborative path planning result is generated through a greedy strategy for initial population generation, fitness calculation, parent tournament selection, partial matching crossover to ensure path legality, segmented adaptive mutation, and elite-preserving population update mechanism. Among them, fitness is calculated using environmental parameters of the offshore operation area, performance parameters of the vessels, and operation path length. Step S3 includes: S3.1: Pair the clustering results of the work sites with the work vessels. Based on a greedy strategy, generate multiple gene sequences for each work vessel according to the clustering results of the work sites to obtain the initial population. S3.2: Perform fitness calculation, parent tournament selection, partial matching crossover to ensure path legality, segmented adaptive mutation, elite preservation population update mechanism, and termination judgment operation on the initial population in sequence to calculate the path planning result; Step S3.2 includes: S3.2.1: Using the environmental parameters of the offshore operation area and the performance parameters of the operating vessel, and based on a multivariate feasibility evaluation index system, calculate the fitness of each gene sequence in the population; and take the maximum value of all fitness values ​​as the optimal fitness of the population. S3.2.2: Randomly select from the entire gene sequence From these, select the one with the highest fitness as the preferred parent; repeat this process until a parent is selected. A preferred parent generation; S3.2.3: In Randomly select one pair from the preferred parent generations, perform partial matching and crossover to obtain offspring; repeat this operation until the desired offspring are obtained. One offspring; S3.2.4: Randomly select two gene positions in the offspring's gene sequence and exchange them to obtain crossover variant offspring; S3.2.5: Select some individuals with high fitness from the parent generation and form a new population with the crossover offspring; S3.2.6: Repeat steps S3.2.1 to S3.2.

5. When the change in the optimal fitness of the population over multiple generations is less than the termination threshold, output the gene sequence corresponding to the optimal fitness of the population at this time as the optimal path planning result for the working vessel. Fitness is a weighted sum of energy consumption, duration, and path length metrics; The formula for calculating the energy consumption index is as follows: in, For the first Energy consumption of a single operating vessel For the first The length of the operating path of each operating vessel. For the first Energy consumption per unit distance for a single operating vessel in still water; For the first The increase in energy consumption caused by environmental factors of each operating vessel. , This is the coefficient for the increase in ship water flow energy consumption. This is the wind energy consumption increment coefficient. For water flow velocity, Wind speed; The duration indicator is the time from the departure of the working vessel to its return to the starting point; Path length metric uses ; The formula for calculating fitness is: in, For the first The adaptability of the operating vessels Energy consumption weighting coefficient This is the duration weighting coefficient. This is the path length weighting coefficient; The preset maximum energy consumption, The preset maximum duration, This is the preset maximum path length.

2. The path planning method for multi-vessel cooperative maritime operations as described in claim 1, characterized in that, The process involves dual verification through spatial convergence and quantitative balance: Perform spatial convergence verification on the clusters obtained by grouping. If spatial convergence is not satisfied, reallocate the job points. If spatial convergence is satisfied, perform a clustering balance check. If the quantity balance is not satisfied, perform a quantity balance redistribution; if the quantity balance is satisfied, output the clustering results of the job points.

3. The path planning method for multi-vehicle cooperative maritime operations as described in claim 2, characterized in that, Step S2 includes: S2.1: Extract the coordinates of the work points in the offshore work area from the coordinates of the offshore work area, perform normalization processing, and obtain the plane rectangular coordinate matrix of the work points; S2.2: From Randomly selected without repetition from each work site 10 work points were selected as the initial cluster centers; S2.3: Traverse all For each work point, calculate its distance to a Cartesian coordinate matrix. The Euclidean distance between the cluster centers is used to assign cluster groups based on the minimum distance. One cluster; S2.4: Calculate the mean coordinates of all operation points in the cluster and denote them as the new cluster centers; S2.5: Perform spatial convergence verification on the clustering. If spatial convergence is not satisfied, proceed to step S2.

3. Perform a cluster balance check. If the cluster balance is not met, proceed to step S2.

6. S2.6: For clusters where the number of work points exceeds the quantity balance verification threshold, calculate the adjustment priority index for each work point in the cluster and select the work points to be adjusted. From clusters with fewer job points than the average number of job points across all clusters, select the cluster that is closest to the job point to be adjusted as the target cluster; Adjust the work points to be adjusted to the target cluster, update the work point set of each cluster; calculate the coordinates of the cluster center, and return to step S2.

3.

4. The path planning method for multi-vehicle cooperative maritime operations as described in claim 3, characterized in that, The formula for calculating the priority index of work sites is: in, For work points Adjustment priority indicators For work points Euclidean distance to the current cluster center For the first The number of actual operation points in each cluster. The average number of job points across all clusters. This is the distance weighting coefficient.

5. The path planning method for multi-vessel cooperative maritime operations as described in claim 1, characterized in that, In step S3.2.4, the probability of mutation occurs. The calculation formula is: in, Based on the probability of mutation, The maximum mutation probability, The fitness of the current gene sequence, This represents the average fitness value of the current generation of the population.

6. A path planning system for multi-vehicle cooperative maritime operations, characterized in that, A path planning method for performing a multi-vessel cooperative maritime operation as described in any one of claims 1 to 5 includes: The data acquisition unit is used to acquire the coordinates and environmental parameters of the offshore operation area, as well as the number and performance parameters of the operating vessels; The work point grouping unit is used to group the work points in the offshore work area based on the number of working vessels and the coordinates of the offshore work area using a clustering algorithm. The clustering results are obtained through dual verification of spatial convergence and quantity balance. The dual verification of spatial convergence and quantity balance is used to ensure that the work points are concentrated in the geographic space and that the number of work points in each cluster is balanced. The multi-vessel collaborative path planning unit is used to optimize the operation path of each vessel based on the clustering results of the operation points, with the goal of maximizing fitness. The optimal multi-vessel collaborative path planning result is generated through a greedy strategy for initial population generation, fitness calculation, parent tournament selection, partial matching crossover to ensure path legality, segmented adaptive mutation, and elite preservation population update mechanism. The fitness is calculated using environmental parameters of the offshore operation area, performance parameters of the operation vessels, and operation path length.

Images