A method and system for cooperative deployment of multiple unmanned underwater vehicles by a host vehicle

Abstract

The application discloses a kind of main aircraft multi-point release's unmanned underwater vehicle cooperative deployment method and system, it is related to unmanned underwater vehicle field, including: receiving main aircraft starting position coordinates, UUV target position coordinates, main aircraft navigation speed, UUV navigation speed and UUV quantity, construct the two-dimensional cooperative deployment task space without geographical restriction, complete task basic information initialization;The application accurately captures UUV target position aggregation feature optimization release point selection, accurately calculates driving and navigation time in combination with main aircraft and UUV navigation characteristics and multiple influence factors, reasonably defines the search range of task time, simultaneously efficiently searches feasible path and dynamically adjusts parameter, iteratively compresses time interval to lock optimal total task time, distributes release point according to efficiency demand and eliminates redundant path, greatly improves deployment efficiency and path rationality, reduces task time consumption.

Description

Technical Field

[0001] This invention relates to the field of unmanned underwater vehicle technology, specifically to a method and system for the coordinated deployment of unmanned underwater vehicles using a multi-point deployment method by a main vehicle. Background Technology

[0002] Unmanned underwater vehicles (UUVs) are intelligent underwater devices that do not require human pilots. They can navigate in various waters using pre-programmed instructions or remote commands, and are capable of functions such as marine resource exploration, hydrological environment monitoring, and underwater target search and rescue. With their long endurance and high maneuverability, they can reach deep-sea areas inaccessible to humans and are widely used in marine scientific research, national defense, and marine ecological protection.

[0003] Patent application No. 202510672972.3 discloses a cooperative control method for unmanned underwater vehicles (UUVs) formations based on the MATD3 algorithm. This application aims to address the problem that "in the field of UUV formation control, traditional control methods are difficult to break free from their dependence on the dynamic model of the controlled object and the environmental disturbance model. Controllers based on linearized or nonlinear environmental disturbances can only meet control requirements under fixed conditions. When environmental or system parameters change, traditional controllers may need to be re-tuned or redesigned, lacking adaptive adjustment capabilities and making it difficult to meet the real-time control requirements in practical engineering tasks."

[0004] However, in a two-dimensional planar environment without geographical constraints, planning the optimal travel path for a single master UUV and the specific deployment points for multiple UUVs, so as to achieve the coordinated deployment of all UUVs and minimize the total mission time for them to reach their respective designated target locations, is a problem that has not yet been fully solved.

[0005] To address this, we propose a collaborative deployment method and system for unmanned underwater vehicles that utilizes a multi-point deployment system by a main vehicle. Summary of the Invention

[0006] In view of the above-mentioned shortcomings of the existing technology, the present invention provides a method and system for the collaborative deployment of unmanned underwater vehicles by multiple deployment points of the main vehicle, 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 collaborative deployment system for unmanned underwater vehicles (UUVs) that allows for multi-point deployment by a main vehicle, comprising: The initialization module receives the starting position coordinates of the main vehicle, the target position coordinates of the UUV, the main vehicle's speed, the UUV's speed, and the number of UUVs. It then constructs a geographically unrestricted two-dimensional collaborative deployment task space and completes the initialization of basic task information. The pre-calculation module analyzes the spatial clustering characteristics of the UUV target positions, determines the optimal clustering number and generates cluster centers, merges the UUV target positions to form a candidate deployment point set, and pre-calculates the travel time matrix between main vehicle points and the travel time matrix from candidate UUV points to the target point. The setting module calculates the minimum and maximum estimates of the total task time based on the initialization parameters and the pre-calculated time matrix. The system comprises the following modules: 1) Determine the initial search range for the total mission time; 2) A verification module, based on a set time limit, searches for the main vehicle's travel path that meets all UUV deployment requirements, simultaneously verifying the feasibility of deployment under this time limit; 3) A generation module, used to obtain the feasibility verification results, iteratively adjust the time search range based on the verification results, filter for the optimal total mission time, allocate the optimal deployment points based on the efficiency requirements of UUVs reaching the target, and determine the optimal travel path of the main vehicle after clearing redundant docking points; 4) A visualization module, used to output the optimal total mission time, algorithm computation time, main vehicle travel path, and a list of UUV deployments at each deployment point, simultaneously generating a static path diagram and a dynamic deployment animation. The initialization module is interconnected with the pre-calculation module via a local area network. The pre-calculation module is interconnected with the setting module via a local area network. The setting module is interconnected with the verification module via a local area network. The verification module is interconnected with the generation module via a local area network. The generation module is interconnected with the visualization module via a local area network.

[0008] Furthermore, in the stage of determining the optimal number of clusters, the pre-computation module calculates the weighted intra-cluster sum of squares under different numbers of clusters k based on the UUV target location coordinate dataset: ; In the formula: The sum of squares within the weighted cluster; Let i be the target location coordinates of the i-th UUV; Let j be the set of UUV target points in the j-th cluster; The position weight of the i-th UUV; Let be the centroid coordinates of the j-th cluster; for arrive The Euclidean distance; Will In the curve that varies with k, the k value at which the rate of change of slope first falls below a preset threshold is determined as the optimal number of clusters.

[0009] Furthermore, the pre-calculation logic for the main vehicle's inter-point travel time matrix and the UUV candidate point-to-target point travel time matrix in the pre-calculation module is as follows: Travel time between points of the main aircraft ; UUV travel time ; In the formula: ( , ), ( , ) The coordinates of any two points in the main vehicle's path; , () represents the coordinates of the candidate delivery point; The stability coefficient of the main vehicle's speed; Main vehicle speed; ( , () represents the target location coordinates of the i-th UUV; The stability coefficient for UUV navigation speed; For UUV travel speed; The time matrix stores the travel time between all main vehicle points and the travel time between all candidate points and the UUV target point.

[0010] Furthermore, when the setting module calculates the minimum and maximum estimates of the total task time, it follows the following: ; In the formula: Number of UUVs; The starting position coordinates of the main vehicle; For the set of candidate delivery points; The travel time of the main vehicle from the starting point to the candidate point c; For the i-th UUV, from candidate point c to target point The sailing time; This represents the task complexity coefficient. The initial search range is .

[0011] Furthermore, the verification module uses a genetic algorithm to search for the main vehicle's travel path and verify its feasibility. The main vehicle's travel path is encoded into a sequence of candidate drop point indices, and a fitness function is constructed to evaluate the path's feasibility. ; In the formula: The fitness function value; , These are the weighting coefficients; The completion rate of UUV deployment corresponding to the path; This represents the actual time taken to complete all deployment tasks under this path; The theoretical shortest deployment time for all UUVs; This is the currently set time limit; when If the time limit is not less than the preset fitness threshold, the deployment of the task is deemed feasible under that time limit; otherwise, it is deemed infeasible.

[0012] Furthermore, the process by which the generation module iteratively adjusts the time search range based on the verification results includes: If deploying the task is feasible under the current time limit, record the actual completion time and update the new time limit to the actual completion time; if not, update the new time lower limit to the current time limit. After each iteration, the search interval width is calculated, and the difference between the new upper time limit and the new lower time limit is taken. The iteration continues when this difference is greater than a dynamic threshold, which is calculated using the following formula: ; In the formula: This is the initial threshold; The attenuation coefficient; This represents the current iteration number; When the search interval width does not exceed When the iteration stops, the new time limit is the optimal total task time.

[0013] Furthermore, the generation module allocates the optimal deployment point stage based on the efficiency requirements of the UUV reaching the target. For the i-th UUV, it iterates through all the docking points along the main vehicle's travel path. Calculate the UUV from Departure to destination Total time consumed: ; In the formula: From the origin point to the docking point of the main aircraft Travel time; For UUV from arrive The sailing time; The coefficient of coordination; This represents the average distance between the i-th UUV target point and other UUV target points; Number of UUVs; Choose to Minimum stop The optimal deployment point for the i-th UUV is determined by eliminating redundant docking points along the path that have not been assigned to any UUVs, thus obtaining the optimal travel path for the main vehicle.

[0014] Furthermore, when the visualization module generates a static path map, it marks the main vehicle's starting point, candidate deployment points, cluster centers, UUV target points, and the optimal driving path of the main vehicle with different identifiers, and simultaneously marks the UUV deployment list corresponding to each deployment point. The visualization module generates a dynamic deployment animation based on the timeline, proportionally dividing the optimal total task time. It uses a linear interpolation algorithm to generate the real-time position coordinates of the main vehicle and each UUV. The generation frequency of the animation frames is determined by the ratio of the algorithm calculation time to the optimal total task time. Each frame synchronously displays the current position of the main vehicle, the real-time navigation trajectory of the deployed UUVs, and the pending deployment status of the undeployed UUVs. The animation ends with a freeze frame displaying the optimal total task time, the total length of the main vehicle's travel path, and the deployment completion time of each UUV.

[0015] On the other hand, a method for the coordinated deployment of unmanned underwater vehicles by multiple deployment points of a main vehicle includes: The system acquires the starting coordinates of the main vehicle, the coordinates of the target UUV, its speed, and the number of UUVs to construct a geographically unrestricted two-dimensional collaborative deployment task space, completing the initialization of basic task information. It analyzes the spatial clustering characteristics of the UUV target locations, determines the optimal number of clusters and cluster centers, and merges candidate deployment points. It pre-calculates the travel time matrix between main vehicle points and the travel time matrix from candidate UUV points to the target point. Based on the initialization parameters and the pre-calculated time matrix, it calculates the minimum and maximum estimates of the total task time, defining the initial search range for the total task time. Based on the set time limit, it uses a genetic algorithm to search for main vehicle travel paths that meet the UUV deployment requirements. The feasibility of the deployment task is verified using a fitness function; if infeasible, the algorithm parameters are adjusted or the time range is broadened before re-verification. Based on the verification results, the time search range is iteratively adjusted to select the optimal total task time. The optimal deployment points are allocated according to the efficiency requirements for UUVs reaching the target, and redundant docking points are cleared to determine the optimal travel path for the main vehicle. The system outputs the optimal total task time, algorithm computation time, main vehicle travel path, and UUV deployment list, simultaneously generating a fully annotated static path map and a dynamic deployment animation displayed along a timeline.

[0016] Compared with the known prior art, the technical solution provided by this invention has the following beneficial effects: This invention can accurately capture the clustering characteristics of UUV target locations to optimize the selection of deployment points. It combines the navigation characteristics of the main vehicle and UUVs with various influencing factors to accurately calculate travel and navigation times, reasonably define the task time search range, efficiently search feasible paths and dynamically adjust parameters, iteratively compress the time interval to lock the optimal total task time, allocate deployment points according to efficiency requirements and eliminate redundant paths, significantly improving deployment efficiency and path rationality, reducing task time, clearly presenting key deployment information and dynamic processes, and facilitating intuitive control of task progress. It provides reliable support for the collaborative deployment of UUVs with multiple deployment points and is adaptable to different distribution scenarios and task complexity, effectively improving deployment flexibility and practicality. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present 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 only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A schematic diagram of a collaborative deployment system for unmanned underwater vehicles that is deployed from multiple points by a main vehicle; Figure 2 A flowchart illustrating a collaborative deployment method for unmanned underwater vehicles using a multi-point deployment system from a main vehicle. Figure 3 This is a diagram showing the output results of the system in a real-world application scenario. 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 describes a collaborative deployment system for unmanned underwater vehicles that uses a main vehicle for multi-point deployment, such as... Figure 1 As shown, it includes: The initialization module is used to receive the starting position coordinates of the main vehicle, the target position coordinates of the UUV, the speed of the main vehicle, the speed of the UUV, and the number of UUVs, to construct a two-dimensional collaborative deployment task space without geographical restrictions, and to complete the initialization of basic task information. The pre-calculation module is used to analyze the spatial clustering characteristics of UUV target locations, determine the optimal number of clusters and generate cluster centers, merge UUV target locations to form a set of candidate deployment points, and pre-calculate the travel time matrix between main vehicle points and the travel time matrix from UUV candidate points to target points. In the stage of determining the optimal number of clusters, the pre-computation module calculates the weighted sum of squares within each cluster under different numbers of clusters k, based on the UUV target location coordinate dataset: ; In the formula: The sum of squares within the weighted cluster; Let i be the target location coordinates of the i-th UUV; Let j be the set of UUV target points in the j-th cluster; The position weight of the i-th UUV; Let be the centroid coordinates of the j-th cluster; for arrive The Euclidean distance; The above formula fully combines the spatial distribution characteristics of UUV target locations, assigns a positional weight related to the proportion of other UUVs within the surrounding preset range to each UUV target point, quantifies the degree of cluster aggregation of target points under different aggregation numbers, and then obtains the slope change rate by calculating the ratio of the difference in the sum of squares within the cluster corresponding to the number of adjacent clusters to the difference in the number of clusters. The number of clusters whose slope change rate is lower than the preset threshold for the first time is selected as the optimal value. This can not only accurately capture the spatial aggregation pattern of target points, but also avoid unreasonable subsequent deployment point settings due to inappropriate cluster numbers, thus providing support for the construction of candidate deployment point sets. Will In the curve that varies with k, the k value where the rate of change of slope first falls below the preset threshold is determined as the optimal number of clusters. The rate of change of slope is calculated by the ratio of the WCSS difference between adjacent k values ​​to the k difference. in, The number is determined by the proportion of other UUVs within a preset range around the target UUV; The pre-calculation logic for the main vehicle's inter-point travel time matrix and the UUV candidate point-to-target point travel time matrix in the pre-calculation module is as follows: Travel time between points of the main aircraft ; UUV travel time ; In the formula: ( , ), ( , ) The coordinates of any two points in the main vehicle's path; , () represents the coordinates of the candidate delivery point; The stability coefficient of the main vehicle's speed; Main vehicle speed; ( , () represents the target location coordinates of the i-th UUV; The stability coefficient for UUV navigation speed; For UUV travel speed; The formula for calculating the travel time between points of the main vehicle is based on the Euclidean distance between any two points in the main vehicle's travel path. Considering that changes in the main vehicle's load and fluctuations in the power system's operating conditions will affect the stability of the travel speed in actual navigation, a speed stability coefficient that is positively correlated with it and whose value is limited to a reasonable range is introduced. The distance and speed stability correction factor are combined and divided by the main vehicle's travel speed to truly reflect the actual time spent on different travel segments, providing accurate data support that fits the actual operating conditions for subsequent path planning and total mission time estimation. The formula for calculating UUV flight time is based on the Euclidean distance between the candidate drop point and the UUV target point. It incorporates a speed stability coefficient that is positively correlated with the output fluctuations of the propulsion system and the operational status fluctuations of the mission payload and is within the range of UUV flight characteristics. After correcting the distance for the impact of speed stability, it is divided by the UUV flight speed. This fully takes into account various fluctuation factors in the actual flight of the UUV, ensuring the accuracy of the flight time estimation and providing a reference for drop point allocation and mission time planning. The time matrix stores the travel time between all main vehicle points and the travel time between all candidate points and the UUV target point; in, The preset value range is [0.01, 0.15], and its value is positively correlated with the load variation of the main vehicle and the fluctuation of the power system operating conditions. The value range of black is [0.008, 0.12], and its value is positively correlated with the output fluctuation amplitude of the UUV propulsion system and the degree of fluctuation of the mission payload operating status; The configuration module is used to calculate the minimum and maximum estimates of the total task time based on the initialization parameters and the pre-calculated time matrix, and to determine the initial search range of the total task time. When setting the module to calculate the minimum and maximum estimates of the total task time, the following rules apply: ; In the formula: Number of UUVs; The starting position coordinates of the main vehicle; For the set of candidate delivery points; The travel time of the main vehicle from the starting point to the candidate point c; For the i-th UUV, from candidate point c to target point The sailing time; This represents the task complexity coefficient. The above formula fully considers the shortest possible time for a single UUV deployment and the extreme time for all UUV deployments. It introduces a task complexity coefficient and dynamically adjusts its value according to the uniformity of the distribution of UUV target points. The more uniform the distribution, the larger the coefficient, and the more uneven the distribution, the smaller the coefficient. By reasonably weighting and correcting the time consumption of different dimensions, a reasonable initial time search range is constructed. This avoids the optimal solution being missed due to an overly narrow range, and also prevents the computational burden from being increased due to an overly wide range. The initial search range is ; in, ∈[0.6, 0.9], the higher the uniformity of the UUV target point distribution, the larger the value; the more uneven the UUV target point distribution, the smaller the value. The verification module searches for the main vehicle's travel path that meets all UUV deployment requirements based on the set time limit, and simultaneously verifies the feasibility of the deployment task under this time limit. The verification module uses a genetic algorithm to search for the main vehicle's travel path and verify its feasibility. The main vehicle's travel path is encoded into a sequence of candidate drop point indices, and a fitness function is constructed to evaluate the path's feasibility. ; In the formula: The fitness function value; , These are the weighting coefficients; The completion rate of UUV deployment corresponding to the path; This represents the actual time taken to complete all deployment tasks under this path; The theoretical shortest deployment time for all UUVs; This is the currently set time limit; The above formula simultaneously considers the UUV deployment completion rate and the degree of matching between the actual task time and the theoretical minimum time, as well as the set time limit. By setting reasonable weight coefficients, it highlights the core position of the deployment completion rate, while balancing the rationality of the time consumption, ensuring that the selected main vehicle travel path can meet the deployment requirements and control the time cost. When the deployment task is determined to be infeasible, the crossover probability and mutation probability of the genetic algorithm are dynamically adjusted according to the difference between the current fitness function value and the preset benchmark value. When the number of iterations has not reached the upper limit, the algorithm's global search capability is enhanced to avoid local optima. When the number of iterations reaches the upper limit, the results are fed back and the time range is relaxed. This ensures the effectiveness of the path search and improves the flexibility of the task feasibility verification. when If the time limit is not less than the preset fitness threshold, the task deployment is deemed feasible under this time limit; otherwise, it is deemed infeasible. in, , All are positive numbers, and , The sum is 1, and ∈[0.6,0.8]; If the decision is deemed infeasible, the following further steps are performed: First, determine if the current iteration count of the genetic algorithm has reached the preset maximum iteration count. If not, dynamically adjust the crossover and mutation probabilities of the genetic algorithm and search again. The adjustment formulas for the crossover and mutation probabilities are as follows: ; In the formula: To adjust the crossover probability and mutation probability; These represent the initial crossover probability and the initial mutation probability. The preset fitness baseline value; After adjustment, the selection, crossover, and mutation operations of the genetic algorithm are re-executed, and the fitness function value is recalculated to determine feasibility. If the current iteration number has reached the preset maximum iteration number, the current genetic algorithm search is terminated, and the infeasibility result is reported to the setting module. The setting module then expands the time search range according to preset rules and enters the next round of verification process. The generation module is used to obtain the feasibility verification results, iteratively adjust the time search range based on the verification results, filter the optimal total task time, allocate the optimal deployment point based on the efficiency requirements of UUV reaching the target, and determine the optimal driving path of the main vehicle after clearing redundant docking points. The process by which the generation module iteratively adjusts the time search range based on the verification results includes: If deploying the task is feasible under the current time limit, record the actual completion time and update the new time limit to the actual completion time; if not, update the new time lower limit to the current time limit. After each iteration, the search interval width is calculated, and the difference between the new upper time limit and the new lower time limit is taken. The iteration continues when this difference is greater than the dynamic threshold. The formula for calculating the dynamic threshold is: ; In the formula: This is the initial threshold; The attenuation coefficient; This represents the current iteration number; The above formula introduces an initial threshold, a decay coefficient, and the current iteration number, so that the criterion for determining the width of the search interval gradually decays with the iteration process. The threshold is larger in the early iterations, which can quickly narrow the time search range. The threshold gradually decreases in the later iterations to ensure that the optimal total task time is accurately locked. This dynamic adjustment method with iteration takes into account search efficiency and avoids the problems of insufficient search or over-computation caused by a fixed threshold. When the search interval width does not exceed When the time limit is reached, the iteration stops, and the new time limit at this point is the optimal total task time. The generation module allocates the optimal deployment point based on the efficiency requirements of the UUV reaching the target. For the i-th UUV, it iterates through all the docking points along the main vehicle's travel path. Calculate the UUV from Departure to destination Total time consumed: ; In the formula: From the origin point to the docking point of the main aircraft Travel time; For UUV from arrive The sailing time; The coefficient of coordination; This represents the average distance between the i-th UUV target point and other UUV target points; Number of UUVs; The above formula not only includes the travel time of the main vehicle from the starting point to the docking point and the travel time of the UUV from the docking point to the target point, but also incorporates the average distance between the UUV target point and all other UUV target points as a coordination consideration factor. The coordination coefficient is adjusted according to the density of the target point distribution. The denser the distribution, the larger the coefficient, and vice versa. By comprehensively balancing the individual travel time and the overall coordination requirements, the docking point with the minimum overall travel time is selected as the optimal deployment point. At the same time, redundant docking points without assigned UUVs are eliminated to ensure the deployment efficiency of each UUV and make the main vehicle's travel path simpler and more reasonable. Choose to Minimum stop As the optimal deployment point for the i-th UUV, redundant docking points in the path that are not assigned to any UUVs are eliminated to obtain the optimal driving path for the main vehicle. in, The preset value range is [0.1, 0.5]. The denser the distribution of UUV target points, the larger the value, and vice versa. The visualization module is used to output the optimal total mission time, algorithm calculation time, main vehicle travel path and UUV deployment list at each deployment point, and simultaneously generate static path map and dynamic deployment animation. When the visualization module generates a static path map, it marks the main vehicle's starting point, candidate deployment points, cluster centers, UUV target points, and the optimal driving path of the main vehicle with different identifiers, and simultaneously marks the UUV deployment list corresponding to each deployment point. The visualization module generates dynamic deployment animations by proportionally dividing the optimal total task time based on the timeline. A linear interpolation algorithm is used to generate the real-time position coordinates of the main vehicle and each UUV. The generation frequency of the animation frames is determined by the ratio of the algorithm calculation time to the optimal total task time. Each frame synchronously displays the current position of the main vehicle, the real-time navigation trajectory of the deployed UUVs, and the pending deployment status of the undeployed UUVs. The animation ends with a freeze frame displaying the optimal total task time, the total length of the main vehicle's travel path, and the deployment completion time of each UUV. The initialization module interacts with the pre-calculation module via the local area network. The pre-calculation module interacts with the setting module via the local area network. The setting module interacts with the verification module via the local area network. The verification module interacts with the generation module via the local area network. The generation module interacts with the visualization module via the local area network.

[0022] In this embodiment, the initialization module receives the starting position coordinates of the main vehicle, the target position coordinates of the UUV, the main vehicle's speed, the UUV's speed, and the number of UUVs. It constructs a geographically unrestricted two-dimensional collaborative deployment task space and completes the initialization of basic task information. The pre-calculation module then analyzes the spatial clustering characteristics of the UUV target positions, determines the optimal clustering number, generates cluster centers, merges the UUV target positions to form a candidate deployment point set, and pre-calculates the travel time matrix between main vehicle points and the travel time matrix from UUV candidate points to the target point. The setting module further calculates the minimum and maximum estimated values ​​of the total task time based on the initialization parameters and the pre-calculated time matrix. The estimation value determines the initial search range for the total mission time. The time limit set by the verification module is verified. The main vehicle travel path that meets the deployment requirements of all UUVs is searched. The feasibility of the deployment mission under this time limit is verified simultaneously. The feasibility verification result is then obtained by the generation module. Based on the verification result, the time search range is iteratively adjusted to select the optimal total mission time. The optimal deployment point is allocated based on the efficiency requirements of UUVs to reach the target. After clearing redundant docking points, the optimal travel path of the main vehicle is determined. Finally, the optimal total mission time, algorithm calculation time, main vehicle travel path and UUV deployment list of each deployment point are output through the visualization module. Static path map and dynamic deployment animation are generated simultaneously.

[0023] In the above embodiments, the system accurately adapts to the distribution characteristics of UUV targets, rationally plans the driving of the main vehicle and the deployment of UUVs, significantly shortens the total mission time, ensures deployment feasibility, and clearly presents the driving path, deployment list and dynamic process, adapting to different mission complexities and target distribution situations, effectively improving the efficiency, accuracy and operability of collaborative deployment of unmanned underwater vehicles.

[0024] Application example: In a marine survey mission, a single main submersible was required to coordinate the deployment of six unmanned underwater vehicles (UUVs) to complete a survey of a specific sea area. The staff first input the core parameters into the system: the main submersible's initial position coordinates were (150, 150) meters, the target position coordinates of the six UUVs were (0, 100) meters, (-50, 100) meters, (0, 0) meters, (80, -60) meters, (-100, -80) meters, and (120, 50) meters respectively, the main submersible's speed was 3 meters per second, and the UUVs' speed was 2 meters per second.

[0025] After receiving the parameters, the system initialization module constructs a geographically unrestricted two-dimensional collaborative deployment task space and completes the basic information initialization. The pre-calculation module first performs K-Means clustering on the UUV target locations, determining the optimal cluster size to be 2 using the elbow rule, generating two cluster centers at (-25, -25) meters and (90, 30) meters. These cluster centers are then merged with six UUV target locations to form eight candidate deployment points. Subsequently, the pre-calculation obtains the travel time matrix between the main vehicle points and the travel time matrix from the UUV candidate points to the target points. For example, the travel time of the main vehicle from the starting point to (-25, -25) meters is 92 seconds, and the travel time of the UUV from (-25, -25) meters to (0, 100) meters is 68 seconds.

[0026] The configuration module combines initialization parameters and pre-calculated time matrix, considering a task complexity coefficient of 0.8 (UUV target points are relatively evenly distributed), and calculates the minimum estimated total task time of 280 seconds and the maximum estimated total time of 420 seconds, determining the initial search range as [280, 420] seconds.

[0027] The verification module uses a genetic algorithm to search for the main vehicle's travel path. Initially, a time limit of 350 seconds (the median of the search range) is set, and a fitness function (weighting coefficients α=0.7, β=0.3) is constructed for feasibility evaluation. If the calculated fitness function value is lower than the preset threshold and the genetic algorithm reaches its maximum iteration count, the time limit is deemed infeasible, and the result is reported back to the setting module, which adjusts the search range to [350, 420] seconds. The median value of 385 seconds is then used as the time limit again, and the crossover and mutation probabilities of the genetic algorithm are dynamically adjusted before a new search is performed. This time, the fitness function value meets the requirements, and the search is deemed feasible. The actual completion time of 320 seconds is recorded, and the search limit is updated to 320 seconds.

[0028] The generation module continuously iterates and adjusts the search range, calculating the search interval width each time. When the interval width is less than the dynamic threshold (the initial threshold is calculated using a decay coefficient) after the 5th iteration, the iteration stops, and the optimal total task time is determined to be 295 seconds. Subsequently, an optimal deployment point is assigned to each UUV. For example, after UUV-0 traverses the main vehicle's path docking points, (-25, -25) meters is selected as the deployment point (with the shortest overall time). At the same time, redundant docking points for unassigned UUVs are eliminated. Finally, the optimal travel path of the main vehicle is determined to be: starting point (150, 150) meters → deployment point (-25, -25) meters → deployment point (90, 30) meters.

[0029] Finally, the visualization module outputs the following results: the optimal total mission time is 295 seconds, the algorithm calculation time is 6.8 seconds, the main vehicle's travel path and the deployment list of each drop point (UUV-0, UUV-1, and UUV-2 are deployed at drop point (-25, -25) meters; UUV-3, UUV-4, and UUV-5 are deployed at drop point (90, 30) meters), and simultaneously generates a clearly marked static path map and a dynamic animation showing the entire deployment process, providing intuitive guidance for mission execution.

[0030] See Figure 3 As shown, this is a diagram displaying the results output of the system in a real-world marine survey scenario. It intuitively presents the core data and specific deployment details after the system's solution. The diagram clearly shows key information such as algorithm computation time, optimal total task time, and ship travel path and UUV deployment allocation. Example 2: At the implementation level, based on Example 1, this example refers to... Figure 2 A more detailed description is provided of the collaborative deployment system for unmanned underwater vehicles that involves multi-point deployment of a main vehicle in Example 1: A method for the coordinated deployment of unmanned underwater vehicles using a multi-point deployment system by a main vehicle includes: Obtain the starting position coordinates of the main vehicle, the target position coordinates of the UUV, the speed of travel and the number of UUVs, construct a two-dimensional collaborative deployment task space without geographical restrictions, and complete the initialization of basic task information; Analyze the spatial clustering characteristics of UUV target locations, determine the optimal number of clusters and cluster centers, merge candidate deployment points, and pre-calculate the travel time matrix between main vehicle points and the travel time matrix from UUV candidate points to target points. Based on the initialization parameters and the pre-calculated time matrix, the minimum and maximum estimates of the total task time are calculated, and the initial search range of the total task time is defined. Based on the set time limit, a genetic algorithm is used to search for the main vehicle's travel path that meets the UUV deployment requirements. The feasibility of the deployment task is verified by the fitness function. If it is not feasible, the algorithm parameters are adjusted or the time range is relaxed and then verified again. Based on the verification results, the time search range is iteratively adjusted, the optimal total mission time is selected, the optimal deployment point is allocated according to the efficiency requirements of the UUV to reach the target, and redundant docking points are cleared to determine the optimal driving path of the main vehicle. Output the optimal total mission time, algorithm calculation time, main vehicle travel path and UUV deployment list, and simultaneously generate a fully annotated static path map and a dynamic deployment animation displayed on the timeline.

[0031] In summary, the system and method described in the above embodiments can accurately capture the clustering characteristics of UUV target locations to optimize the selection of deployment points. By combining the navigation characteristics of the main vehicle and UUVs and various influencing factors, the system can accurately calculate travel and navigation times, reasonably define the task time search range, efficiently search for feasible paths and dynamically adjust parameters, iteratively compress the time interval to lock the optimal total task time, allocate deployment points according to efficiency requirements and eliminate redundant paths, significantly improving deployment efficiency and path rationality, reducing task time, clearly presenting key deployment information and dynamic processes, facilitating intuitive control of task progress, providing reliable support for collaborative deployment of UUVs at multiple deployment points, and adapting to different distribution scenarios and task complexity, effectively improving deployment flexibility and practicality.

[0032] 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 host vehicle multi-point launched uncrewed submersible cooperative deployment system, characterized by, include: The initialization module is used to receive the starting position coordinates of the main vehicle, the target position coordinates of the UUV, the speed of the main vehicle, the speed of the UUV, and the number of UUVs, to construct a two-dimensional collaborative deployment task space without geographical restrictions, and to complete the initialization of basic task information. The pre-computation module is used to analyze the spatial clustering characteristics of UUV target locations, determine the optimal number of clusters and generate cluster centers, merge UUV target locations to form a set of candidate deployment points, and pre-compute the travel time matrix between main vehicle points and the travel time matrix from UUV candidate points to target points. The configuration module is used to calculate the minimum and maximum estimates of the total task time based on the initialization parameters and the pre-calculated time matrix, and to determine the initial search range of the total task time. The verification module searches for the main vehicle's travel path that meets all UUV deployment requirements based on the set time limit, and simultaneously verifies the feasibility of the deployment task under this time limit. The generation module is used to obtain the feasibility verification results, iteratively adjust the time search range based on the verification results, filter the optimal total task time, allocate the optimal deployment point based on the efficiency requirements of UUV reaching the target, and determine the optimal driving path of the main vehicle after clearing redundant docking points. The visualization module is used to output the optimal total mission time, algorithm calculation time, main vehicle travel path, and UUV deployment list at each deployment point, and simultaneously generate static path maps and dynamic deployment animations.

2. The system of claim 1, wherein: In the stage of determining the optimal number of clusters, the pre-computation module calculates the weighted intra-cluster sum of squares for different numbers of clusters k based on the UUV target location coordinate dataset: ； In the formula: The sum of squares within the weighted cluster; Let i be the target location coordinates of the i-th UUV; Let j be the set of UUV target points in the j-th cluster; The position weight of the i-th UUV; Let be the centroid coordinates of the j-th cluster; for arrive The Euclidean distance; Will In the curve that varies with k, the k value at which the rate of change of slope first falls below a preset threshold is determined as the optimal number of clusters.

3. The collaborative deployment system for unmanned underwater vehicles with multi-point deployment of a main vehicle according to claim 1, characterized in that, The pre-calculation logic for the main vehicle's inter-point travel time matrix and the UUV candidate point to target point travel time matrix in the pre-calculation module is as follows: Main vehicle inter-point travel time ; UUV transit time ; In the formula: ( , ), ( , ) The coordinates of any two points in the main vehicle's path; , () represents the coordinates of the candidate delivery point; The stability coefficient of the main vehicle's speed; Main vehicle speed; ( , () represents the target location coordinates of the i-th UUV; The stability coefficient for UUV navigation speed; For UUV travel speed; The time matrix stores the travel time between all main vehicle points and the travel time between all candidate points and the UUV target point.

4. The system of claim 1, wherein, When the setting module calculates the minimum and maximum estimates of the total task time, it follows the following rules: ； In the formula: For the number of UUVs; The starting position coordinates of the main vehicle; For the set of candidate delivery points; The travel time of the main vehicle from the starting point to the candidate point c; For the i-th UUV, from candidate point c to target point The sailing time; This represents the task complexity coefficient. then the initial search range is .

5. The system of claim 1, wherein, The verification module uses a genetic algorithm to search for the main vehicle's travel path and verify its feasibility. The main vehicle's travel path is encoded into a sequence of candidate drop point indices, and a fitness function is constructed to evaluate the path's feasibility. ； In the formula: The fitness function value; , These are the weighting coefficients; The completion rate of UUV deployment corresponding to the path; This represents the actual time taken to complete all deployment tasks under this path; The theoretical shortest deployment time for all UUVs; This is the currently set time limit; when If the time limit is not less than the preset fitness threshold, the deployment of the task is deemed feasible under that time limit; otherwise, it is deemed infeasible.

6. The system of claim 1, wherein: The process by which the generation module iteratively adjusts the time search range based on the verification results includes: If deploying the task is feasible under the current time limit, record the actual completion time and update the new time limit to the actual completion time; if not, update the new time lower limit to the current time limit. After each iteration, the search interval width is calculated, and the difference between the new upper time limit and the new lower time limit is taken. The iteration continues when this difference is greater than a dynamic threshold, which is calculated using the following formula: ； In the formula: is an initial threshold value; is a decay coefficient; is the current iteration number; When the search interval width does not exceed the iteration is stopped, and the new upper time limit at this time is the optimal total task time.

7. The mother vehicle multi-point launched UUV cooperative deployment system of claim 6, wherein, The generation module allocates the optimal deployment point stage based on the efficiency requirements of the UUV reaching the target. For the i-th UUV, it iterates through all the docking points along the main vehicle's travel path. Calculate the UUV from Departure to destination Total time consumed: ； In the formula: From the origin point to the docking point of the main aircraft Travel time; For UUV from arrive The sailing time; The coefficient of coordination; This represents the average distance between the i-th UUV target point and other UUV target points; For the number of UUVs; selecting to cause minimum stop point as the optimal drop-off point for the ith UUV, while eliminating redundant stop points in the path that are not assigned any UUV, to obtain the optimal travel path for the host vehicle.

8. The system of claim 1, wherein: When the visualization module generates a static path map, it marks the main vehicle's starting point, candidate deployment points, cluster centers, UUV target points, and the optimal driving path of the main vehicle with different identifiers, and simultaneously marks the UUV deployment list corresponding to each deployment point. The visualization module generates a dynamic deployment animation based on the timeline, proportionally dividing the optimal total task time. It uses a linear interpolation algorithm to generate the real-time position coordinates of the main vehicle and each UUV. The generation frequency of the animation frames is determined by the ratio of the algorithm calculation time to the optimal total task time. Each frame synchronously displays the current position of the main vehicle, the real-time navigation trajectory of the deployed UUVs, and the pending deployment status of the undeployed UUVs. The animation ends with a freeze frame displaying the optimal total task time, the total length of the main vehicle's travel path, and the deployment completion time of each UUV.

9. The system of claim 1, wherein, The initialization module is interconnected with the pre-calculation module via a local area network. The pre-calculation module is interconnected with the setting module via a local area network. The setting module is interconnected with the verification module via a local area network. The verification module is interconnected with the generation module via a local area network. The generation module is interconnected with the visualization module via a local area network.

10. A method for cooperative deployment of multiple unmanned underwater vehicles from a mother vehicle, the method being a method for implementing a system for cooperative deployment of multiple unmanned underwater vehicles from a mother vehicle according to any one of claims 1-9, characterized in that, include: Obtain the starting position coordinates of the main vehicle, the target position coordinates of the UUV, the speed of travel and the number of UUVs, construct a two-dimensional collaborative deployment task space without geographical restrictions, and complete the initialization of basic task information; Analyze the spatial clustering characteristics of UUV target locations, determine the optimal number of clusters and cluster centers, merge candidate deployment points, and pre-calculate the travel time matrix between main vehicle points and the travel time matrix from UUV candidate points to target points. Based on the initialization parameters and the pre-calculated time matrix, the minimum and maximum estimates of the total task time are calculated, and the initial search range of the total task time is defined. Based on the set time limit, a genetic algorithm is used to search for the main vehicle's travel path that meets the UUV deployment requirements. The feasibility of the deployment task is verified by the fitness function. If it is not feasible, the algorithm parameters are adjusted or the time range is relaxed and then verified again. Based on the verification results, the time search range is iteratively adjusted, the optimal total mission time is selected, the optimal deployment point is allocated according to the efficiency requirements of the UUV to reach the target, and redundant docking points are cleared to determine the optimal driving path of the main vehicle. Output the optimal total mission time, algorithm calculation time, main vehicle travel path and UUV deployment list, and simultaneously generate a fully annotated static path map and a dynamic deployment animation displayed on the timeline.

Images