A path planning method and system for cooperative launching of multiple unmanned underwater vehicles in a safety zone of a vehicle

Abstract

This invention discloses a path planning method and system for the collaborative deployment of multiple unmanned underwater vehicles (UUVs) within a safe zone, relating to the field of UUVs. It includes: a preprocessing module for receiving information on the safe zone extent, initial ship parameters, UUV target parameters, and navigation speed, generating a set of candidate deployment points within the safe zone, and estimating the navigation time matrix of the ship and UUVs; and an initialization module for receiving the navigation time matrix and determining the initial search interval for the total mission time based on the matrix. This invention generates candidate deployment points by adaptively matching the safe zone boundary with UUV deployment requirements, accurately calculates the navigation time matrix, rationally delineates the initial search interval, efficiently explores feasible deployment paths using an optimized search algorithm, combines multi-dimensional fitness logic to select the optimal solution, dynamically adjusts the search interval and accuracy threshold, and balances mission timeliness, full deployment, and deployment equilibrium to quickly lock in the optimal total mission time.

Description

Technical Field

[0001] This invention relates to the field of unmanned underwater vehicle technology, specifically to a path planning method and system for the coordinated deployment of multiple unmanned underwater vehicles within a safe zone. Background Technology

[0002] Unmanned underwater vehicle (UUV) path planning is a core technology for autonomously completing tasks such as ocean exploration and resource exploration. By combining marine environmental perception data and comprehensively considering constraints such as obstacle avoidance, energy consumption, and mission timeliness, it generates the optimal navigation route through intelligent algorithms, which can effectively improve the operational efficiency and navigation safety of UUVs and provide reliable autonomous navigation support for stable operations in complex deep-sea scenarios.

[0003] The invention patent application with application number 201710538828.6 discloses an unmanned underwater vehicle path planning device and method based on the detection threat domain. The application aims to solve the problem that "most current unmanned underwater vehicle path planning methods only consider the problem of the existence of obstacles, and do not consider how to handle waypoints to meet constraints such as turning radius, as well as safety constraints such as underwater acoustics-based threat detection".

[0004] However, the technical objective of minimizing the total mission time for a single vessel to reach its designated target point within a safe operating area by planning its own navigation path and the timing of each UUV deployment remains difficult to achieve with existing technologies.

[0005] To address this, we propose a path planning method and system for the coordinated deployment of multiple unmanned underwater vehicles within a safe zone. Summary of the Invention

[0006] In view of the above-mentioned shortcomings of the existing technology, the present invention provides a path planning method and system for the coordinated deployment of multiple unmanned underwater vehicles within the safe zone of the aircraft, which can effectively solve the problems of the existing technology.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions; This invention discloses a path planning system for the coordinated deployment of multiple unmanned underwater vehicles within a safe zone, comprising: The preprocessing module receives information on the safety zone range, initial ship parameters, UUV target parameters, and navigation speed, generates a set of candidate deployment points within the safety zone, and estimates the navigation time matrix of the ship and UUV. The initialization module receives the navigation time matrix and determines the initial search interval for the total mission time based on the matrix. The verification module analyzes the estimated total mission time based on the current search interval, searches for feasible ship deployment paths based on the estimated value, and simultaneously verifies the deployment feasibility of all UUVs under this estimated value. The decision-making module evaluates the merits of feasible paths according to a preset fitness logic, and selects the optimal deployment point sequence and deployment timing that can cover all UUV deployment needs. The optimization module dynamically adjusts the search interval based on the feasibility verification results, controls the verification module to run repeatedly until the preset accuracy requirements are met and the optimal total mission time is locked. The visualization module outputs the optimal total mission time, ship navigation paths, and UUV deployment information, generating a visual chart of the collaborative deployment path. The preprocessing module is interconnected with the initialization module and the verification module via a wireless network. The verification module is interconnected with the decision module via a wireless network. The decision module is interconnected with the optimization module via a wireless network. The optimization module is interconnected with the visualization module via a wireless network.

[0008] Furthermore, in the stage of generating the candidate deployment point set within the safe zone, the preprocessing module adaptively adjusts the grid division density based on the polygonal boundary shape of the safe zone, uses the ratio of the safe zone area to the total number of UUVs to be deployed as the basic parameter of the grid density, and then completes the candidate deployment point set based on the coordinates of the UUV target points. The grid division density is expressed based on the grid node spacing. , This refers to the spacing between grid nodes; This is the adaptive adjustment coefficient; The polygon area of ​​the safe zone; This represents the total number of UUVs to be deployed. When supplementing the candidate deployment point set based on the UUV target point coordinates, a spatial topology judgment is performed on the target location coordinates of each UUV. If the target location is located inside or on the boundary of the safe zone, it is directly included in the candidate deployment point set. If the UUV target location is located outside the safe zone, the shortest distance from the target location to the boundary of the safe zone is calculated, and the projection point on the boundary line segment of the safe zone corresponding to the shortest distance is included as a supplementary deployment point in the candidate deployment point set. Finally, after traversing all UUV target points to complete the supplement, duplicate coordinate points in the candidate deployment point set are removed to obtain a discretized candidate deployment point set. The navigation time matrix includes the ship navigation time matrix. and UUV flight time matrix ,in The matrix is ​​(m+1)×(m+1) dimensional, where row index and column index 0 correspond to the initial position of the ship. The row index and column indexes 1~m correspond to the candidate delivery points respectively. Matrix elements This represents the time it takes for a ship to travel from position i to position j, where Tu is m × A 3D matrix, where row indices 1 to m correspond to candidate delivery points. Column index 1~ Corresponding to number 1~ One UUV, matrix element This indicates that the q-th UUV is delivered from the candidate delivery point. sail to its target location The time is calculated by dividing the Euclidean distance between the two points by the corresponding sailing speed.

[0009] Furthermore, when determining the initial search interval for the total task time in the initialization module, the following applies: Based on the flight time matrix generated by the preprocessing module, the theoretical minimum and maximum values ​​of the total mission time are calculated to form the initial search interval: ; In the formula: C represents the initial search range; C represents the set of candidate deployment points; and U represents the set of UUVs to be deployed. The time it takes for the ship to travel from its initial position to the candidate drop point c; For the i-th UUV, the distance from the candidate delivery point c to its target location is... The sailing time.

[0010] Furthermore, when the verification module runs for the first time, it is based on the initial search interval determined by the initialization module. The midpoint value is taken as the estimated total task time. During subsequent iterations and verifications, the estimated total task time is updated based on the midpoint value of the search interval dynamically adjusted by the optimization module according to the previous verification results. The verification module searches for feasible ship deployment paths using an improved genetic algorithm. For the current estimated total mission time, calculate the latest allowed time for the vessel to arrive at each candidate deployment point for each UUV to be deployed and each candidate deployment point. ,in This represents the estimated total task time. This indicates that the i-th UUV is delivered from the candidate delivery point. The time it takes to reach its target location; If the ship travels from its initial position to the candidate drop point via the preceding nodes in the path sequence... The cumulative time is less than Then determine the candidate delivery point. It can be used to deploy the i-th UUV; Traverse all candidate deployment points in the path sequence. If a feasible deployment point can be assigned to each UUV to be deployed, and all UUVs can be deployed... Once they reach their respective target positions, the current status is determined. If a feasible delivery path exists, the deployment is feasible; otherwise, the deployment is deemed infeasible. If the deployment is determined to be infeasible, the verification module outputs an infeasibility result to the optimization module, which then adjusts the current deployment. Set the new lower limit of the search range and re-trigger the verification module to update based on the adjusted new search range. Then, a feasibility assessment was performed again.

[0011] Furthermore, the fitness logic preset in the decision-making module is as follows: ; In the formula: This represents the path fitness value. As weight; UUV deployment completion rate; This is the upper bound of the current search interval; This represents the actual total task time corresponding to the current path; This represents the total number of candidate delivery points. The number of UUVs allocated to the p-th candidate delivery point; The number of UUVs is evenly distributed among the delivery points; The decision-making module uses the following logic to select feasible paths: First, filter out The set of paths with a value of 1 is selected, and then fitness values ​​are chosen from this set. Paths exceeding a preset fitness threshold; if the number of paths meeting the condition is not less than a preset number, then select one of them. The longest path is selected as the optimal path. If the number of paths that meet the criteria is less than the preset number, the preset fitness threshold is lowered by a preset ratio and the search is repeated until the optimal delivery point sequence and delivery timing are determined.

[0012] Furthermore, during the process of dynamically adjusting the search interval by the optimization module, the preset accuracy requirement is dynamically updated with each iteration: ; And the iteration termination condition also satisfies: and ; In the formula: The absolute precision threshold for the k-th iteration; The absolute precision threshold for the k-th iteration; This is the accuracy attenuation coefficient; , These are the upper and lower bounds of the search interval in the k-th iteration, respectively; To preset the relative accuracy threshold; When the iteration meets the termination condition, the optimal total task time is locked. If the termination condition is not met even after the preset maximum number of iterations is reached, the median of the current search interval is taken as the optimal total task time, and the error message indicating that the accuracy is not up to standard is recorded and output to the visualization module simultaneously.

[0013] Furthermore, when adjusting the search interval, the optimization module employs a dynamic shrinkage factor based on feasibility verification results. ; The This is the fitness value of the optimal feasible path in the current validation round; This is the average fitness value of all feasible paths in the current validation round; if there are no feasible paths in the current validation round, then... = =0, contractility factor Then take the preset baseline shrinkage factor; When the verification result is feasible, the upper bound of the new search interval is updated to... When the verification result is infeasible, the lower bound of the new search interval is updated to... ; After each update, if the new search interval has an abnormal situation where the upper bound is less than the lower bound, the search interval will be reconstructed based on the median of the current search interval, and the reconstruction range will be the median plus or minus the preset interval offset.

[0014] Furthermore, the visualization module includes a visualization chart of the collaborative delivery path, comprising: Timeline visualization: Using time as the horizontal axis, it displays the time nodes from the ship's departure from the initial position, arrival at each candidate drop point, and completion of all drop tasks, as well as the time span for each UUV from the start of the drop to the target position; Spatial topology visualization: Mark the polygon boundaries and vertex coordinates of the safety zone, distinguish candidate deployment points, final selected deployment points, and UUV target points with different colors and sizes of markers, connect the ship's navigation path and the UUV's navigation path from the deployment point to the target point with arrowed line segments, and mark the UUV number and deployment timing corresponding to each deployment point. Performance index heat map: Against the background of the safe zone space, a heat distribution is generated based on the fitness value of each candidate deployment point, the corresponding UUV's travel time, and the energy cost of the ship to reach the point. The color depth maps the degree of quality of the index.

[0015] On the other hand, a path planning method for the coordinated deployment of multiple unmanned underwater vehicles within a safe zone includes: The system receives information on the safety zone, initial ship parameters, UUV target parameters, and navigation speed. It then completes the candidate deployment point set through adaptive grid partitioning and simultaneously calculates the navigation time matrix for both the ship and the UUV. Based on the navigation time matrix, it calculates the theoretical minimum and maximum values ​​of the total mission time to construct an initial search interval for the total mission time. Using the midpoint of the current search interval as a guess of the total mission time, it searches for feasible ship deployment paths using an improved genetic algorithm, verifying the deployment feasibility of all UUVs under this guess. Based on a preset fitness logic, it evaluates the merits of feasible paths, selecting the optimal deployment point sequence and corresponding deployment timing that can cover all UUV deployment needs. Based on the feasibility verification results and a dynamic shrinkage factor, it adjusts the search interval, iteratively executing the verification and decision-making process until the preset accuracy requirements are met and the optimal total mission time is locked. Finally, it outputs the optimal total mission time, ship navigation paths, and UUV deployment information, generating a visual chart of the collaborative deployment path that includes time axis correlation, spatial topology, and performance index heatmaps.

[0016] Compared with the known prior art, the technical solution provided by this invention has the following beneficial effects: This invention generates candidate deployment points by adaptively matching the safety zone boundary with UUV deployment requirements, accurately calculates the flight time matrix, rationally delineates the initial search interval, and efficiently explores feasible deployment paths using an optimized search algorithm. It then combines multi-dimensional fitness logic to select the optimal solution, dynamically adjusts the search interval and accuracy threshold, and balances task timeliness, full deployment, and deployment equilibrium. It quickly locks in the optimal total task time and presents key information in an intuitive visualization format, significantly improving the efficiency, accuracy, and practicality of path planning for multi-UUV collaborative deployment. It flexibly adapts to the priority requirements of different task scenarios, effectively avoids deployment conflicts and resource waste, and provides effective support for multi-UUV collaborative deployment within the vehicle's safety zone. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0018] Figure 1 A schematic diagram of a path planning system for the coordinated deployment of multiple unmanned underwater vehicles within a safe zone of an aircraft. Figure 2A flowchart illustrating a path planning method for the coordinated deployment of multiple unmanned underwater vehicles within a safe zone. Figure 3 This is a schematic diagram showing the system operation results in this aspect. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present 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 the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0020] The present invention will be further described below with reference to embodiments.

[0021] Example 1: This embodiment presents a path planning system for the coordinated deployment of multiple unmanned underwater vehicles within a safe zone, such as... Figure 1 As shown, it includes: The preprocessing module is used to receive information on the safety zone range, initial ship parameters, UUV target parameters, and navigation speed, generate a set of candidate deployment points within the safety zone, and estimate the navigation time matrix of the ship and the UUV. In the preprocessing module, during the stage of generating a set of candidate deployment points within the safe zone, the grid density is adaptively adjusted based on the polygonal boundary shape of the safe zone. The ratio of the safe zone area to the total number of UUVs to be deployed is used as the basic parameter for the grid density. Then, the set of candidate deployment points is completed based on the coordinates of the UUV target points. Mesh density is expressed based on the spacing between mesh nodes. , This refers to the spacing between grid nodes; This is the adaptive adjustment coefficient; The polygon area of ​​the safe zone; This represents the total number of UUVs to be deployed. The above formula combines the area of ​​the safe zone with the total number of UUVs to be deployed, and balances the impact of the two on the distribution of deployment points through square root operation. Then, an adaptive adjustment coefficient is introduced, which can be flexibly adjusted according to the different priorities of the task for path optimization accuracy and computational efficiency. This provides a scientific and reasonable grid foundation for the initial construction of the candidate deployment point set, and can also adapt to the core needs under different scenarios. It creates uniform and flexible initial distribution conditions for the subsequent completion of the deployment point set based on the coordinates of UUV target points. When supplementing the candidate deployment point set based on the UUV target point coordinates, a spatial topology judgment is performed on the target location coordinates of each UUV. If the target location is located inside or on the boundary of the safe zone, it is directly included in the candidate deployment point set. If the UUV target location is located outside the safe zone, the shortest distance from the target location to the boundary of the safe zone is calculated, and the projection point on the boundary line segment of the safe zone corresponding to the shortest distance is included as a supplementary deployment point in the candidate deployment point set. Finally, after traversing all UUV target points to complete the supplement, duplicate coordinate points in the candidate deployment point set are removed to obtain a discretized candidate deployment point set. The navigation time matrix includes the ship navigation time matrix. and UUV flight time matrix ,in The matrix is ​​(m+1)×(m+1) dimensional, where row index and column index 0 correspond to the initial position of the ship. The row index and column indexes 1~m correspond to the candidate delivery points respectively. Matrix elements This indicates that the ship starts from the position corresponding to index i (when i=0). When i≥1, it is Navigate to the position corresponding to index j (j=0 is...) When j≥1, it is The time, Tu, is m× A 3D matrix, where row indices 1 to m correspond to candidate delivery points. Column index 1~ Corresponding to number 1~ One UUV, matrix element This indicates that the q-th UUV is delivered from the candidate delivery point. sail to its target location The time, the elements of both matrices are calculated by dividing the Euclidean distance between the two points by the corresponding sailing speed; The projection point is obtained by drawing a perpendicular line from the target point to the boundary line segment. If the perpendicular line does not intersect the boundary line segment, the endpoint of the boundary line segment is taken as the supplementary projection point. The preset value range is [0.6, 1.4]. When prioritizing path optimization accuracy and allowing for reduced computational efficiency, the value is larger; when prioritizing computational efficiency and accepting a certain loss of accuracy, the value is smaller. The initialization module is used to receive the navigation time matrix and determine the initial search interval for the total mission time based on the matrix. When determining the initial search interval for the total task time in the initialization module, the following rules apply: Based on the flight time matrix generated by the preprocessing module, the theoretical minimum and maximum values ​​of the total mission time are calculated to form the initial search interval: ; In the formula: C represents the initial search range; C represents the set of candidate deployment points; and U represents the set of UUVs to be deployed. The time it takes for the ship to travel from its initial position to the candidate drop point c; For the i-th UUV, the distance from the candidate delivery point c to its target location is... The sailing time; In the above formula, the theoretical minimum total mission time is determined by selecting the combination of the shortest time from the main vehicle to the deployment point and the shortest time from the UUV to the target point for each UUV, and then taking the maximum value among them to ensure that the minimum deployment time requirement of all UUVs is covered; the theoretical maximum value is determined by superimposing the longest time from the main vehicle to the deployment point and the longest time from the UUV to the target point to define the extreme time range of the mission. The initial search interval formed by the two avoids the computational redundancy caused by the range being too wide, and prevents the optimal solution from being missed by the range being too narrow, providing an accurate and efficient starting point for subsequent iterative verification. The verification module is used to analyze the estimated total task time based on the current search interval, search for feasible ship deployment paths based on the estimated value, and simultaneously verify the deployment feasibility of all UUVs under the estimated value. When the verification module runs for the first time, it is based on the initial search range determined by the initialization module. The midpoint value is taken as the estimated total task time. During subsequent iterations and verifications, the estimated total task time is updated based on the midpoint value of the search interval dynamically adjusted by the optimization module according to the previous verification results. The verification module searches for feasible ship deployment paths using an improved genetic algorithm. For the current estimated total mission time, calculate the latest allowed time for the vessel to arrive at each candidate deployment point for each UUV to be deployed and each candidate deployment point. ,in This represents the estimated total task time. This indicates that the i-th UUV is delivered from the candidate delivery point. The time it takes to reach its target location; If the ship travels from its initial position to the candidate drop point via the preceding nodes in the path sequence... The cumulative time is less than Then determine the candidate delivery point. It can be used to deploy the i-th UUV; Traverse all candidate deployment points in the path sequence. If a feasible deployment point can be assigned to each UUV to be deployed, and all UUVs can be deployed... Once they reach their respective target positions, the current status is determined. If a feasible delivery path exists, the deployment is feasible; otherwise, the deployment is deemed infeasible. If the deployment is determined to be infeasible, the verification module outputs an infeasibility result to the optimization module, which then adjusts the current deployment. Set the new lower limit of the search range and re-trigger the verification module to update based on the adjusted new search range. Perform a feasibility assessment again; Regarding the configuration of the improved genetic algorithm: An improved genetic algorithm is used as the search engine for feasible ship deployment paths. The core parameters of the improved genetic algorithm are configured as follows: Encoding method: The ship's deployment path is encoded as an integer sequence, where each integer in the sequence corresponds to the index of a candidate deployment point in the candidate deployment point set; Selection strategy: An elite selection strategy is adopted, which retains a preset number of elite individuals with the best fitness in each generation of the population and directly enters the next generation. Crossover strategy: A single-point crossover method is adopted, randomly selecting the crossover point of two parent path sequences, and exchanging the gene fragments after the crossover point to generate the offspring path; Mutation strategy: Randomly swap the index positions of two candidate delivery points in the offspring path sequence with a preset mutation probability to maintain population diversity; The decision-making module is used to evaluate the merits of feasible paths based on preset fitness logic, and to select the optimal sequence of delivery points and delivery timing that can cover all UUV delivery needs. The fitness logic is preset in the decision-making module as follows: ; In the formula: This represents the path fitness value. As weight; UUV deployment completion rate; This is the upper bound of the current search interval; This represents the actual total task time corresponding to the current path; This represents the total number of candidate delivery points. The number of UUVs allocated to the p-th candidate delivery point; The number of UUVs is evenly distributed among the delivery points; The above formula comprehensively considers three core indicators: UUV deployment completion rate, deviation between actual total task time and reference time, and UUV distribution balance of candidate deployment points. It adapts to different task requirements through weight allocation. When the task emphasizes full deployment, the weight of deployment completion rate can be increased; when focusing on timeliness, the weight of time deviation can be increased; and when focusing on deployment balance, the weight of distribution balance can be increased. This multi-dimensional integrated evaluation method can comprehensively reflect the overall performance of the path. The decision-making module's logic for selecting feasible paths is as follows: First, filter out The set of paths with a value of 1 is selected, and then fitness values ​​are chosen from this set. Paths exceeding a preset fitness threshold; if the number of paths meeting the condition is not less than a preset number, then select one of them. The longest path is selected as the optimal path. If the number of paths that meet the conditions is less than the preset number, the preset fitness threshold is lowered by a preset ratio and the search is repeated until the optimal delivery point sequence and delivery timing are determined. in, All are positive numbers, and their sum is 1. This is especially important when the task has a high requirement for full UUV deployment and cannot tolerate any deployment omissions. The larger the value, the better when the task allows for delayed or temporary deployment of some non-core UUVs, or focuses more on other optimization objectives. The smaller the value, the more sensitive the task is to the total task time, and the more urgent the need to quickly deploy UUVs to adapt to the dynamic response requirements of the battlefield. A larger value indicates a more relaxed task time constraint, a greater focus on deployment balance, or a higher priority for full deployment over timeliness. The smaller the value, the better, especially in scenarios where the maximum number of candidate delivery points is limited, and it is necessary to avoid excessive congestion at a single delivery point leading to decreased delivery efficiency or deployment conflicts. A larger value indicates that the deployment points have sufficient resources and no load pressure, and the focus is more on full deployment or time optimization. The smaller the value; The optimization module is used to dynamically adjust the search range based on the feasibility verification results, control the verification module to run repeatedly until the preset accuracy requirements are met and the optimal total task time is locked. During the optimization module's dynamic adjustment of the search range, the preset accuracy requirement is dynamically updated with each iteration: ; The above formula causes the absolute accuracy threshold in the iteration process to decay exponentially with each iteration. The accuracy requirement is low in the initial stage to speed up the convergence speed, and the accuracy is gradually increased in the subsequent stages to approach the optimal solution. The decay coefficient can be flexibly set according to the different requirements of the task for convergence speed and computational accuracy. This dynamic accuracy control method avoids the low computational efficiency caused by high initial accuracy and prevents the deviation of the optimal value caused by low accuracy in the later stages, thus ultimately achieving a dynamic balance between convergence speed and computational accuracy. And the iteration termination condition also satisfies: and ; In the formula: The absolute precision threshold for the k-th iteration; The absolute precision threshold for the k-th iteration; This is the accuracy attenuation coefficient; , These are the upper and lower bounds of the search interval in the k-th iteration, respectively; To preset the relative accuracy threshold; The iteration termination condition adopts dual constraints of absolute precision and relative precision. The absolute precision threshold decays exponentially with the number of iterations, and the precision decay coefficient can be flexibly adapted to the requirements of convergence speed and computational accuracy. The relative precision threshold can be compatible with task scenarios of different numerical scales, avoiding the deviation of precision judgment due to the difference in time consumption. When the iteration meets the termination condition, the optimal total task time is locked. If the termination condition is not met even after the preset maximum number of iterations is reached, the median of the current search interval is taken as the optimal total task time, and the accuracy failure prompt message is recorded and output to the visualization module simultaneously. in, ∈[0.01,0.1], when the task requires prioritizing improving the iteration convergence speed, is sensitive to the computational time of the algorithm, and allows for acceptable accuracy to gradually approach the optimal value. The larger the value, the more critical it is when the task requires extremely high precision in calculating the optimal total task time, accepts a longer iteration cycle, and needs to slowly increase precision to accurately lock in the optimal value. The smaller the value; When the optimization module adjusts the search interval, it uses a dynamic shrinkage factor based on the feasibility verification results. ; The above formula calculates the shrinkage magnitude based on the average fitness difference between the optimal feasible path and all feasible paths in the current verification round. When there is a high-quality feasible path, the shrinkage intensity is adjusted by the fitness difference to make the search interval shrink more accurately towards the optimal solution. When there is no feasible path, a preset benchmark shrinkage factor is used. This method makes the interval shrinkage magnitude directly related to the path quality and avoids the blindness of a fixed shrinkage ratio. This is the fitness value of the optimal feasible path in the current validation round; This is the average fitness value of all feasible paths in the current validation round; if there are no feasible paths in the current validation round, then... = =0, contractility factor Then take the preset baseline shrinkage factor; When the verification result is feasible, the upper bound of the new search interval is updated to... When the verification result is infeasible, the lower bound of the new search interval is updated to... ; After each update, if the new search interval has an abnormal situation where the upper bound is less than the lower bound, the search interval will be reconstructed based on the median of the current search interval, and the reconstruction range will be the median plus or minus the preset interval offset. The visualization module is used to output the optimal total mission time, ship navigation path and UUV deployment information, and generate a visualization chart of the collaborative deployment path; The visualization module includes the following charts showing the collaborative delivery path: Timeline visualization: Using time as the horizontal axis, it displays the time nodes from the ship's departure from the initial position, arrival at each candidate drop point, and completion of all drop tasks, as well as the time span for each UUV from the start of the drop to the target position; Spatial topology visualization: Mark the polygon boundaries and vertex coordinates of the safety zone, distinguish candidate deployment points, final selected deployment points, and UUV target points with different colors and sizes of markers, connect the ship's navigation path and the UUV's navigation path from the deployment point to the target point with arrowed line segments, and mark the UUV number and deployment timing corresponding to each deployment point. Performance index heat map: Against the background of the safe zone space, a heat distribution is generated based on the fitness value of each candidate deployment point, the corresponding UUV's travel time, and the energy cost of the ship to reach the point. The color depth maps the degree of quality of the index. The preprocessing module interacts with the initialization module and the verification module via a wireless network. The verification module interacts with the decision module via a wireless network. The decision module interacts with the optimization module via a wireless network. The optimization module interacts with the visualization module via a wireless network.

[0022] In this embodiment, the preprocessing module receives information on the safe zone range, initial ship parameters, UUV target parameters, and navigation speed, generates a set of candidate deployment points within the safe zone, estimates the navigation time matrix of the ship and UUV, and the initialization module receives the navigation time matrix after initialization. Based on the matrix, it determines the initial search interval for the total mission time. The verification module further analyzes the estimated total mission time based on the current search interval, searches for feasible ship deployment paths based on the estimated values, and simultaneously verifies the deployment feasibility of all UUVs under the estimated values. Then, the decision module evaluates the merits of feasible paths according to preset fitness logic, selects the optimal deployment point sequence and deployment timing that can cover all UUV deployment needs, and the optimization module dynamically adjusts the search interval by simultaneously obtaining the feasibility verification results. It controls the verification module to run repeatedly until the preset accuracy requirements are met and the optimal total mission time is locked. Finally, the visualization module outputs the optimal total mission time, ship navigation path, and UUV deployment information, generating a visual chart of the collaborative deployment path.

[0023] In the above embodiments, the system can accurately adapt to the safe zone environment and task requirements, efficiently plan the deployment path and timing, shorten the total task time, ensure full deployment of UUVs and balanced load at deployment points, adapt to different scenario priorities, intuitively present relevant information, and improve the efficiency, reliability and flexibility of multi-UUV collaborative deployment.

[0024] Application example: To conduct a coordinated deployment of multiple unmanned underwater vehicles (UUVs) in a nearshore polygonal safety zone, 10 UUVs need to be deployed from the main vehicle to a designated area to complete a marine environmental monitoring task. The polygonal area of ​​the safety zone is 50 square kilometers. The initial position coordinates of the main vehicle are (120.5°E, 30.2°N). The main vehicle's speed is assumed to be 20 knots, and the UUVs' speed is assumed to be 10 knots. Of the 10 UUVs, 7 are located within the safety zone, and 3 are located outside the safety zone. The optimal deployment path and timing need to be planned.

[0025] First, the system preprocessing module is activated to receive the aforementioned safe zone range, initial parameters of the main vehicle, UUV target parameters, and assumed flight speed information. When generating the candidate deployment point set, the ratio of the safe zone area to the total number of UUVs (5 square kilometers / UUV) is used as the basic parameter for grid density. An adaptive adjustment coefficient k=1.0 is selected (balancing optimization accuracy and computational efficiency), resulting in a grid node spacing of approximately 2.24 kilometers. The safe zone is then divided into grids according to this density. Subsequently, spatial topology judgment is performed on the UUV target points. Target points within the 7 safe zones are directly included in the candidate deployment point set. For the 3 target points outside the safe zones, the shortest distance to the safe zone boundary is calculated, and the corresponding boundary projection points are obtained through vertical projection (for one target point, the vertical projection has no intersection point, so the endpoint of the boundary line segment is used). These are added to the candidate deployment point set, and duplicate coordinates are removed, ultimately resulting in 20 discretized candidate deployment points. Simultaneously, the system calculates and generates a navigation time matrix: the main vehicle's navigation time matrix is ​​21×21 dimensions, in which the main vehicle takes 30 minutes to travel from its initial position to the 5th candidate drop point and 25 minutes to travel to the 12th candidate drop point; the UUV's navigation time matrix is ​​20×10 dimensions, in which the 3rd UUV takes 32 minutes to travel from the 8th candidate drop point to its target position and the 7th UUV takes 28 minutes to travel from the 15th candidate drop point to its target position.

[0026] Next, the initialization module calculates the initial search interval based on the flight time matrix: the theoretical minimum total mission time is 65 minutes (that is, the largest value among the shortest values ​​of "time from the main vehicle to the drop point + time from the UUV to the target point" for each UUV), and the theoretical maximum value is 95 minutes (that is, the sum of the longest time from the main vehicle to all candidate drop points, 50 minutes, and the longest time from all UUVs to the target point, 45 minutes). Therefore, the initial search interval is determined to be [65 minutes, 95 minutes].

[0027] When the verification module runs for the first time, the midpoint of the initial search interval, 80 minutes, is taken as the estimated total task time. An improved genetic algorithm is used to search for feasible deployment paths for the main UUV. The algorithm configuration is as follows: the deployment path is encoded as an integer sequence of candidate deployment point indices; an elite selection strategy is used to retain the top 5 individuals in each generation; offspring are generated by exchanging parent gene fragments through single-point crossover; and two deployment point indices in the offspring path are randomly swapped with a mutation probability of 0.05. For each UUV and candidate deployment point, the latest allowed time is calculated. For example, the latest allowed time for the first UUV to depart from the third candidate deployment point is 80 minutes minus its travel time of 32 minutes, resulting in 48 minutes. It is determined that the cumulative time for the main UUV to travel from the initial position to the third candidate deployment point via the preceding nodes is 42 minutes, which is less than 48 minutes. Therefore, this deployment point can be used to deploy the first UUV. After traversing all candidate deployment points, all 10 UUVs can be assigned to feasible deployment points and can all reach the target position within 80 minutes. The deployment is deemed feasible under the current estimated value.

[0028] In the decision-making module, the preset fitness weights are ω1=0.4, ω2=0.3, and ω3=0.3 (task priority ensures full deployment, while also considering timeliness and deployment balance). The fitness value of the current feasible path is calculated as follows: UUV deployment completion rate R=1, the upper bound of the current search interval is 95 minutes, the actual total task time for the path is 78 minutes, the average number of UUVs allocated to candidate deployment points is 0.5, and the third indicator after calculating the deviation in the number of UUVs allocated to each deployment point is 0.9, resulting in a final fitness value of approximately 0.89. After filtering out the set of paths with R=1, the preset fitness threshold is 0.85. Eight paths meet the criteria, and the path with the highest fitness value of 0.89 is selected as the candidate optimal path.

[0029] The optimization module dynamically adjusts the search interval, with a preset accuracy decay coefficient α=0.05. The absolute accuracy threshold for the first iteration is approximately 4.76 minutes. Based on the current validation results, the shrinkage factor λ≈0.54 is calculated. Due to deployment feasibility, the upper bound of the new search interval is updated to 73.1 minutes, and the new search interval is [65 minutes, 73.1 minutes]. After three iterations, the search interval difference is 3 minutes, and the relative difference is 4%, both meeting the preset accuracy requirements. The optimal total task time is locked at approximately 69 minutes.

[0030] Finally, the visualization module outputs the following results: the optimal total mission time is 69 minutes, and the main vehicle's flight path is: initial position → 5th candidate drop point → 12th candidate drop point → 8th candidate drop point → ..., with corresponding drop times of 15 minutes, 30 minutes, 42 minutes, etc. The generated collaborative drop path visualization chart clearly shows the process of the main vehicle departing at 0 minutes, each drop time node, and completing all drops in 55 minutes, while also presenting the flight time span of each UUV. The spatial topology chart marks the safety zone boundary and vertex coordinates, with small gray dots representing candidate drop points, large red dots representing the final selected drop point, and blue triangles representing UUV target points. Arrowed line segments connect the flight paths of the main vehicle and UUVs, and the UUV numbers and drop times are marked. The performance index heatmap uses the safety zone as the background, with selected drop points appearing in darker colors due to their high adaptability, short flight time, and low energy consumption, while edge candidate drop points are in lighter colors, intuitively reflecting the performance of each drop point.

[0031] See Figure 3 As shown in the diagram, this is a schematic representation of the actual operation results of the path planning system. It visually presents the core output data and result format after the system completes the solution. It clearly shows that the system's calculation time was 9.88 seconds, the estimated shortest total task time was 195.27 seconds, and the starting point coordinates of the main vessel and the specific coordinates of each drop point are clearly given. The system has also automatically saved the results as image files with specified names. This result is the final presentation of the optimal solution after the system has undergone collaborative iterative calculations through modules such as preprocessing, initialization, verification, decision-making, and optimization, meeting the accuracy requirements. It demonstrates the system's core performance in rapidly solving and accurately outputting ship navigation paths and drop point parameters.

[0032] Example 2: At the implementation level, based on Example 1, this example refers to... Figure 2 A further detailed description is provided of the path planning system for the coordinated deployment of multiple unmanned underwater vehicles within a safe zone, as described in Example 1: A path planning method for the coordinated deployment of multiple unmanned underwater vehicles within a safe zone of a vehicle includes: The system receives information on the safety zone range, initial ship parameters, UUV target parameters, and navigation speed. It completes the candidate deployment point set through adaptive grid partitioning and simultaneously calculates the navigation time matrix of the ship and the UUV. The theoretical minimum and maximum values ​​of the total mission time are calculated based on the flight time matrix to construct the initial search interval for the total mission time. Using the midpoint of the current search interval as the estimated total task time, we searched for feasible ship deployment paths using an improved genetic algorithm and verified the deployment feasibility of all UUVs under this estimated value. Based on the preset fitness logic, the merits of feasible paths are evaluated, and the optimal sequence of delivery points and corresponding delivery times that can cover all UUV delivery needs are selected. Based on the feasibility verification results and the dynamic shrinkage factor, the search range is adjusted, and the verification and decision-making process is executed cyclically until the preset accuracy requirements are met and the optimal total task time is locked. Output the optimal total mission time, ship navigation path and UUV deployment information, and generate a collaborative deployment path visualization chart that includes time axis correlation, spatial topology and performance index heatmap.

[0033] In summary, the system in the above embodiments generates candidate deployment points by adaptively matching the safety zone boundary with UUV deployment requirements, accurately calculates the flight time matrix, rationally delineates the initial search interval, efficiently explores feasible deployment paths with an optimized search algorithm, selects the optimal solution by combining multi-dimensional fitness logic, dynamically adjusts the search interval and accuracy threshold, takes into account task timeliness, full deployment and deployment balance, quickly locks the optimal total task time, and presents key information in an intuitive visualization form. This significantly improves the efficiency, accuracy and practicality of path planning for multi-UUV collaborative deployment, flexibly adapts to the priority requirements of different task scenarios, effectively avoids deployment conflicts and resource waste, and provides effective support for multi-UUV collaborative deployment within the vehicle's safety zone.

[0034] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. 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. Such modifications or substitutions will 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 system for the coordinated deployment of multiple unmanned underwater vehicles within a safe zone of a submersible, characterized in that, include: The preprocessing module is used to receive information on the safety zone range, initial ship parameters, UUV target parameters, and navigation speed, generate a set of candidate deployment points within the safety zone, and estimate the navigation time matrix of the ship and the UUV. The initialization module is used to receive the navigation time matrix and determine the initial search interval for the total mission time based on the matrix. The verification module is used to analyze the estimated total task time based on the current search interval, search for feasible ship deployment paths based on the estimated value, and simultaneously verify the deployment feasibility of all UUVs under the estimated value. The decision-making module is used to evaluate the merits of feasible paths based on preset fitness logic, and to select the optimal sequence of delivery points and delivery timing that can cover all UUV delivery needs. The optimization module is used to dynamically adjust the search range based on the feasibility verification results, control the verification module to run repeatedly until the preset accuracy requirements are met and the optimal total task time is locked. The visualization module is used to output the optimal total mission time, ship navigation path and UUV deployment information, and generate a visualization chart of the collaborative deployment path.

2. The path planning system for the coordinated deployment of multiple unmanned underwater vehicles within a safe zone according to claim 1, characterized in that, In the stage of generating a candidate deployment point set within the safe zone, the preprocessing module adaptively adjusts the grid division density based on the polygonal boundary shape of the safe zone, uses the ratio of the safe zone area to the total number of UUVs to be deployed as the basic parameter of the grid density, and then completes the candidate deployment point set based on the coordinates of the UUV target points. The grid division density is expressed based on the grid node spacing. , This refers to the spacing between grid nodes; This is the adaptive adjustment coefficient; The polygon area of ​​the safe zone; This represents the total number of UUVs to be deployed. When supplementing the candidate deployment point set based on the UUV target point coordinates, a spatial topology judgment is performed on the target location coordinates of each UUV. If the target location is located inside or on the boundary of the safe zone, it is directly included in the candidate deployment point set. If the UUV target location is located outside the safe zone, the shortest distance from the target location to the boundary of the safe zone is calculated, and the projection point on the boundary line segment of the safe zone corresponding to the shortest distance is included as a supplementary deployment point in the candidate deployment point set. Finally, after traversing all UUV target points to complete the supplement, duplicate coordinate points in the candidate deployment point set are removed to obtain a discretized candidate deployment point set. The navigation time matrix includes the ship navigation time matrix. and UUV flight time matrix ,in The matrix is ​​(m+1)×(m+1) dimensional, where row index and column index 0 correspond to the initial position of the ship. The row index and column indexes 1~m correspond to the candidate delivery points respectively. Matrix elements This represents the time it takes for a ship to travel from position i to position j, where Tu is m × A 3D matrix, where row indices 1 to m correspond to candidate delivery points. Column index 1~ Corresponding to number 1~ One UUV, matrix element This indicates that the q-th UUV is delivered from the candidate delivery point. sail to its target location The time is calculated by dividing the Euclidean distance between the two points by the corresponding sailing speed.

3. The path planning system for the coordinated deployment of multiple unmanned underwater vehicles within a safe zone according to claim 1, characterized in that, When determining the initial search interval for the total task time in the initialization module, the following applies: Based on the flight time matrix generated by the preprocessing module, the theoretical minimum and maximum values ​​of the total mission time are calculated to form the initial search interval: ； In the formula: C represents the initial search range; C is the set of candidate delivery points. U represents the set of UUVs to be deployed; The time it takes for the ship to travel from its initial position to the candidate drop point c; For the i-th UUV, the distance from the candidate delivery point c to its target location is... The sailing time.

4. The path planning system for the coordinated deployment of multiple unmanned underwater vehicles within a safe zone according to claim 1, characterized in that, When the verification module is run for the first time, it is based on the initial search interval determined by the initialization module. The midpoint value is taken as the estimated total task time. During subsequent iterations and verifications, the estimated total task time is updated based on the midpoint value of the search interval dynamically adjusted by the optimization module according to the previous verification results. The verification module searches for feasible ship deployment paths using an improved genetic algorithm. For the current estimated total mission time, calculate the latest allowed time for the vessel to arrive at each candidate deployment point for each UUV to be deployed and each candidate deployment point. ,in This represents the estimated total task time. This indicates that the i-th UUV is delivered from the candidate delivery point. The time it takes to reach its target location; If the ship travels from its initial position to the candidate drop point via the preceding nodes in the path sequence... The cumulative time is less than Then the candidate delivery point is determined. It can be used to deploy the i-th UUV; Traverse all candidate deployment points in the path sequence. If a feasible deployment point can be assigned to each UUV to be deployed, and all UUVs can be deployed... Once they reach their respective target positions, the current status is determined. If a feasible delivery path exists, the deployment is feasible; otherwise, the deployment is deemed infeasible. If the deployment is determined to be infeasible, the verification module outputs an infeasibility result to the optimization module, which then adjusts the current deployment. Set the new lower limit of the search range and re-trigger the verification module to update based on the adjusted new search range. Then, a feasibility assessment was performed again.

5. A path planning system for the coordinated deployment of multiple unmanned underwater vehicles within a safe zone, as described in claim 1, is characterized in that... The fitness logic preset in the decision-making module is as follows: ； In the formula: This represents the path fitness value. As weight; UUV deployment completion rate; This is the upper bound of the current search interval; This represents the actual total task time corresponding to the current path; This represents the total number of candidate delivery points. The number of UUVs allocated to the p-th candidate delivery point; The number of UUVs is evenly distributed among the delivery points; The decision-making module uses the following logic to select feasible paths: First, filter out The set of paths with a value of 1 is selected, and then fitness values ​​are chosen from this set. Paths exceeding a preset fitness threshold; if the number of paths meeting the condition is not less than a preset number, then select one of them. The longest path is selected as the optimal path. If the number of paths that meet the criteria is less than the preset number, the preset fitness threshold is lowered by a preset ratio and the search is repeated until the optimal delivery point sequence and delivery timing are determined.

6. The path planning system for the coordinated deployment of multiple unmanned underwater vehicles within a safe zone according to claim 1, characterized in that, During the process of dynamically adjusting the search interval by the optimization module, the preset accuracy requirement is dynamically updated with each iteration: ； And the iteration termination condition also satisfies: and ; In the formula: The absolute precision threshold for the k-th iteration; The absolute precision threshold for the k-th iteration; This is the accuracy attenuation coefficient; , These are the upper and lower bounds of the search interval in the k-th iteration, respectively; To preset the relative accuracy threshold; When the iteration meets the termination condition, the optimal total task time is locked. If the termination condition is not met even after the preset maximum number of iterations is reached, the median of the current search interval is taken as the optimal total task time, and the error message indicating that the accuracy is not up to standard is recorded and output to the visualization module simultaneously.

7. A path planning system for the coordinated deployment of multiple unmanned underwater vehicles within a safe zone, as described in claim 6, is characterized in that... When adjusting the search interval, the optimization module uses a dynamic shrinkage factor based on feasibility verification results. ; The This is the fitness value of the optimal feasible path in the current validation round; This is the average fitness value of all feasible paths in the current validation round; if there are no feasible paths in the current validation round, then... = =0, contractility factor Then take the preset baseline shrinkage factor; When the verification result is feasible, the upper bound of the new search interval is updated to... When the verification result is infeasible, the lower bound of the new search interval is updated to... ; After each update, if the new search interval has an abnormal situation where the upper bound is less than the lower bound, the search interval will be reconstructed based on the median of the current search interval, and the reconstruction range will be the median plus or minus the preset interval offset.

8. A path planning system for the coordinated deployment of multiple unmanned underwater vehicles within a safe zone, as described in claim 1, is characterized in that... The visualization module includes the following visual charts showing the collaborative delivery path: Timeline visualization: Using time as the horizontal axis, it displays the time nodes from the ship's departure from the initial position, arrival at each candidate drop point, and completion of all drop tasks, as well as the time span for each UUV from the start of the drop to the target position; Spatial topology visualization: Mark the polygon boundaries and vertex coordinates of the safety zone, distinguish candidate deployment points, final selected deployment points, and UUV target points with different colors and sizes of markers, connect the ship's navigation path and the UUV's navigation path from the deployment point to the target point with arrowed line segments, and mark the UUV number and deployment timing corresponding to each deployment point. Performance index heat map: Against the background of the safe zone space, a heat distribution is generated based on the fitness value of each candidate deployment point, the corresponding UUV's travel time, and the energy cost of the ship to reach the point. The color depth maps the degree of quality of the index.

9. A path planning system for the coordinated deployment of multiple unmanned underwater vehicles within a safe zone, as described in claim 1, is characterized in that... The preprocessing module is interconnected with the initialization module and the verification module via a wireless network. The verification module is interconnected with the decision module via a wireless network. The decision module is interconnected with the optimization module via a wireless network. The optimization module is interconnected with the visualization module via a wireless network.

10. A path planning method for the coordinated deployment of multiple unmanned underwater vehicles within a safe zone of a vehicle, the method being an implementation method of the path planning system for the coordinated deployment of multiple unmanned underwater vehicles within a safe zone of a vehicle as described in any one of claims 1-9, characterized in that... include: The system receives information on the safe zone range, initial ship parameters, UUV target parameters, and navigation speed. It completes the candidate deployment point set through adaptive grid partitioning and simultaneously calculates the navigation time matrix of the ship and the UUV. The theoretical minimum and maximum values ​​of the total mission time are calculated based on the navigation time matrix to construct the initial search interval for the total mission time. Using the midpoint of the current search interval as the estimated total task time, we searched for feasible ship deployment paths using an improved genetic algorithm and verified the deployment feasibility of all UUVs under this estimated value. Based on the preset fitness logic, the merits of feasible paths are evaluated, and the optimal sequence of delivery points and corresponding delivery times that can cover all UUV delivery needs are selected. Based on the feasibility verification results and the dynamic shrinkage factor, the search range is adjusted, and the verification and decision-making process is executed cyclically until the preset accuracy requirements are met and the optimal total task time is locked. Output the optimal total mission time, ship navigation path and UUV deployment information, and generate a collaborative deployment path visualization chart that includes time axis correlation, spatial topology and performance index heatmap.

Images