Ship task adaptive planning system and method
Through deep reinforcement learning algorithms and intelligent decision-making support technology, an adaptive planning system for ship tasks was built, which solved the problem of difficulty in adapting to complex marine environments and experience inheritance of traditional methods, achieved efficient task planning and experience inheritance, and improved the combat of ship teams Performance and adaptability.
Patent Information
- Application Number
- CN202410991432.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-23
- Publication Date
- 2024-11-22
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Traditional ship mission planning methods are difficult to adapt to the complex and changeable marine environment and combat needs, and the inheritance of excellent experience poses the problems of personnel replacement, information islands and backward inheritance methods.
Deep reinforcement learning algorithms and intelligent decision support technology are adopted to build a ship task adaptive planning system including resource management, environmental perception, task planning, task evaluation, experience management and human-computer interaction modules to achieve efficient adaptive planning and experience inheritance.
It improves the consistency and accuracy of task execution, reduces the decision-making risks caused by human factors, ensures the effective inheritance of the captain's excellent experience, shortens the personnel adaptation period, and improves the combat effectiveness and adaptability of the ship team.
Smart Images

Figure CN119005734A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of ship mission planning, and in particular to a ship mission adaptive planning system and method. Background Art
[0002] With the rapid development of information technology, intelligent decision support systems have become one of the key technologies for modern ships to perform tasks. In this field, how to combine artificial intelligence algorithms and expert knowledge to achieve efficient planning of ship tasks and the inheritance of excellent experience has always been a hot topic and difficulty in research. Traditional task planning methods mainly rely on manual experience and rule judgment, which is difficult to adapt to the complex and changeable marine environment and combat needs. At the same time, the inheritance of excellent experience also faces problems such as personnel replacement, information islands and backward inheritance methods.
[0003] In response to the above challenges, in recent years, the application of artificial intelligence technology in different fields has gradually deepened, providing new solutions for ship mission planning and experience inheritance. In particular, the development of deep reinforcement learning algorithms enables intelligent agents to learn and optimize decision-making strategies through interaction with the environment, demonstrating strong adaptive capabilities and decision-making accuracy. Summary of the invention
[0004] The technical problem to be solved by the present invention is to provide a ship mission adaptive planning system and method, which uses deep reinforcement learning algorithms and intelligent decision support technology to achieve efficient adaptive planning of ship missions and intelligent inheritance of the captain's excellent experience. By integrating modules such as data perception, intelligent decision-making, experience management, and human-computer interaction, the present invention constructs a comprehensive and efficient ship mission adaptive planning system architecture, providing a new solution for the intelligent development of ships.
[0005] In order to solve the above technical problems, the first aspect of the embodiment of the present invention discloses a ship mission adaptive planning system, the system includes a resource management module, an environment perception module, a mission planning module, a mission evaluation module, an experience management module and a human-computer interaction module;
[0006] The resource management module is data-connected with the mission planning module to manage own resource information;
[0007] The environment perception module is data-connected with the human-computer interaction module and the task planning module to collect sea area data information in real time;
[0008] The task evaluation module is data-connected with the task planning module and the human-computer interaction module, and is used to perform task evaluation according to real-time data, and send the task evaluation result to the human-computer interaction module for display;
[0009] The experience management module is data-connected with the task planning module and the human-computer interaction module to manage experience information;
[0010] The human-computer interaction module is data-connected with the task planning module, the environment perception module, and the experience management module for information input and display.
[0011] As an optional implementation manner, in the first aspect of the embodiment of the present invention, the experience management module includes an experience knowledge base and an intelligent recommendation unit;
[0012] The experience knowledge base is used to store the digitized information of the captain’s experience;
[0013] The intelligent recommendation unit is used to receive instructions from the human-computer interaction module, provide simulation training for the new captain, and feed back the simulation training results of the new captain to the human-computer interaction module in real time.
[0014] As an optional implementation, in the first aspect of the embodiment of the present invention, the human-computer interaction module includes an input unit, a display unit, and a data interaction unit;
[0015] The input unit is data-connected with the display unit, and is used to receive and display user input information, and send the user input information to the task planning module;
[0016] The data interaction unit is data-connected with the display unit, the environment perception module, and the experience management module, and is used to interact with the environment perception module and the experience management module and send the data to the display unit for display.
[0017] A second aspect of an embodiment of the present invention discloses a method for adaptive planning of ship missions, the method comprising:
[0018] S1, obtaining ship mission adaptive planning data information, and sending the ship mission adaptive planning data information to the mission planning module;
[0019] S2, using a task planning module to process the ship task adaptive planning data information to obtain a ship task adaptive planning result; the ship task adaptive planning result includes a task allocation result and a path planning result;
[0020] S3, sending the ship mission adaptive planning result to the human-computer interaction module and the mission evaluation module;
[0021] S4, using the task evaluation module to evaluate the ship task adaptive planning result, obtain a task evaluation result, and send the task evaluation result to the human-computer interaction module;
[0022] S5, using the human-computer interaction module to display the ship mission adaptive planning result and the mission evaluation result.
[0023] As an optional implementation manner, in the second aspect of the embodiment of the present invention, the obtaining of ship mission planning data information includes:
[0024] S11, using a human-computer interaction module to obtain task description information;
[0025] S12, using the resource management module to obtain resource management information;
[0026] S13, using the experience management module to obtain historical experience information;
[0027] S14, using the environment perception module to obtain environmental information;
[0028] S15, integrating the task description information, the resource management information, the historical experience information and the environmental information to obtain ship task adaptive planning data information.
[0029] As an optional implementation, in the second aspect of the embodiment of the present invention, the use of the task planning module to process the ship task adaptive planning data information to obtain the ship task adaptive planning result includes:
[0030] S21, using a task planning module, processing the ship task adaptive planning data information to obtain a training data set;
[0031] S22, using the training data set to train the ship mission adaptive planning model to obtain an optimized ship mission adaptive planning model;
[0032] S23, using the optimized ship mission adaptive planning model to process the ship mission adaptive planning task to be processed, to obtain a ship mission adaptive planning result.
[0033] As an optional implementation, in the second aspect of the embodiment of the present invention, the use of the task planning module to process the ship task adaptive planning data information to obtain a training data set includes:
[0034] S211, performing data cleaning on the ship mission adaptive planning data information to remove redundant information and format errors to obtain first preprocessed data information;
[0035] S212, performing data conversion on the first preprocessed data information to obtain second preprocessed data information;
[0036] S213, extracting key information from the second preprocessed data information to obtain third preprocessed data information;
[0037] S214, performing feature extraction on the third preprocessed data information to obtain feature parameter information;
[0038] S215, integrating the feature parameter information to obtain a training data set.
[0039] As an optional implementation manner, in the second aspect of the embodiment of the present invention, extracting key information from the second preprocessed data information to obtain third preprocessed data information includes:
[0040] S2131, processing the second preprocessed data information to obtain a word grid network;
[0041] The node set of the word grid network is N = {n0, n1, ...}, the arc set of the word grid network is E = {e0, e1, ...}, and each node n i ∈N contains a time stamp t(n i ), arc e u→v represents the arc connecting node u to node v;
[0042] S2132, processing the word grid network to obtain a keyword network;
[0043] The nodes in the keyword network are NS={N0, N1, ...}, Represents the node N in the keyword network i Connect to node N j A set of arcs;
[0044] S2133, extract key information from the keyword network to obtain third preprocessing data information.
[0045] As an optional implementation, in the second aspect of the embodiment of the present invention, the optimizing ship task adaptive planning model is used to process the ship task adaptive planning task to be processed to obtain the ship task adaptive planning result, including:
[0046] S231, using a human-computer interaction module to obtain a to-be-processed ship mission adaptive planning task;
[0047] S232, using the environment perception module to obtain real-time environment information;
[0048] S233, using the optimized ship task adaptive planning model, processing the to-be-processed ship task adaptive planning task and the real-time environmental information to obtain a task allocation result;
[0049] S234, performing path planning for the ship that performs the task according to the task allocation result to obtain a path planning result.
[0050] As an optional implementation, in the second aspect of the embodiment of the present invention, performing path planning for a ship performing a task according to the task allocation result to obtain a path planning result includes:
[0051] S2341, obtaining path planning initial information according to the task allocation result;
[0052] The path planning initial information includes the starting point S(x i ,y i ), target point G(x o ,y o ), the coordinates of the center of the obstacle (x ob1 ,y ob1 )、(x ob2 ,y ob2 )、(x ob3 ,y ob3 ), the radius of the obstacle r1, r2, r3, and the ocean current speed U monitored at the start time c and wind speed U a ;
[0053] S2342, using the path planning model, according to the ocean current speed U c and the wind speed U a , for the starting point S(x i ,y i ), target point G(x o ,y o ), the coordinates of the center of the obstacle (x ob1 ,y ob1 )、(x ob2 ,y ob2 )、(x ob3 ,y ob3 ) and the radius r1, r2, r3 of the obstacle to obtain the initial path;
[0054] S2343, the ship performing the mission sails along the initial path;
[0055] S2344, the ship continuously monitors the changes in sea current speed and wind speed during navigation;
[0056] S2345, when the ocean current speed and the wind speed change, the path planning model is used to calculate the current position information of the ship and the target point G(x o ,y o ), the coordinates of the center of the obstacle (x ob1 ,yob1 )、(x ob2 ,y ob2 )、(x ob3 ,y ob3 ) and the radius r1, r2, r3 of the obstacle are processed to obtain an updated path, and the ship navigates along the updated path;
[0057] S2346, determine whether the ship has reached the target point G(x o ,y o );
[0058] When the judgment result is yes, output the path planning result; when the judgment result is no, execute S2344.
[0059] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:
[0060] The present invention discloses a ship mission adaptive planning system and method, the system includes a resource management module, an environment perception module, a mission planning module, a mission evaluation module, an experience management module and a human-computer interaction module; through the interaction of various functional modules, ship mission adaptive planning is achieved, the consistency and accuracy of task execution are improved, and the decision-making risks caused by human factors are reduced. Through the experience management function, the present invention ensures that the captain's excellent experience is effectively passed on, shortens the adaptation period of personnel, and improves the overall combat effectiveness and adaptability of the ship team. The innovative architecture and algorithm application of the present invention provide a new solution for the intelligent development of ships, which has important value and practical application prospects. BRIEF DESCRIPTION OF THE DRAWINGS
[0061] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0062] Figure 1 It is a structural schematic diagram of a ship mission adaptive planning system disclosed in an embodiment of the present invention;
[0063] Figure 2 It is a flow chart of a method for adaptive planning of ship missions disclosed in an embodiment of the present invention;
[0064] Figure 3 It is a flow chart of another method for adaptive planning of ship tasks disclosed in an embodiment of the present invention. DETAILED DESCRIPTION
[0065] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0066] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, device, product or equipment that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or equipment.
[0067] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.
[0068] The present invention discloses a ship task adaptive planning system and method, the system includes a resource management module, an environment perception module, a task planning module, a task evaluation module, an experience management module and a human-computer interaction module; the resource management module is connected with the task planning module data for managing its own resource information; the environment perception module is connected with the human-computer interaction module and the task planning module data for real-time collection of sea area data information; the task evaluation module is connected with the task planning module and the human-computer interaction module data for performing task evaluation according to real-time data, and sending the task evaluation result to the human-computer interaction module for display; the experience management module is connected with the task planning module and the human-computer interaction module data for experience information management; the human-computer interaction module is connected with the task planning module, the environment perception module, and the experience management module data for information input and display. The method of the present invention can realize efficient adaptive planning of ship tasks. The following are detailed descriptions.
[0069] Embodiment 1
[0070] See also Figure 1 , Figure 1 : is a schematic diagram of the structure of a ship mission adaptive planning system disclosed in an embodiment of the present invention. Figure 1The described ship mission adaptive planning system is applied to the field of ship mission planning technology to realize ship mission adaptive planning, improve the consistency and accuracy of mission execution, and reduce the decision-making risk caused by human factors, which is not limited in the embodiments of the present invention. Figure 1 As shown, the ship mission adaptive planning system includes a resource management module, an environmental perception module, a mission planning module, a mission evaluation module, an experience management module and a human-computer interaction module; this system adopts an advanced distributed architecture, integrates various functional modules, and realizes efficient information processing and intelligent decision support.
[0071] The resource management module is data-connected with the mission planning module to manage own resource information;
[0072] The environment perception module is data-connected with the human-computer interaction module and the task planning module to collect sea area data information in real time;
[0073] The task evaluation module is data-connected with the task planning module and the human-computer interaction module, and is used to perform task evaluation according to real-time data, and send the task evaluation result to the human-computer interaction module for display;
[0074] The experience management module is data-connected with the task planning module and the human-computer interaction module to manage experience information;
[0075] The human-computer interaction module is data-connected with the task planning module, the environment perception module, and the experience management module for information input and display.
[0076] Optionally, the experience management module includes an experience knowledge base, an intelligent recommendation unit and a simulation training unit;
[0077] The experience knowledge base is used to store digital information of the captain's experience, including the operation experience and decision-making records of excellent captains;
[0078] The intelligent recommendation unit is used to receive instructions from the human-computer interaction module, provide simulation training for the new captain, and feed back the simulation training results of the new captain to the human-computer interaction module in real time.
[0079] The simulation training unit is based on real historical cases and provides a virtual training environment where new captains can improve their skills through simulation training.
[0080] The experience management module has a real-time feedback function, which collects the operation data of new captains in actual tasks, updates the experience knowledge base and intelligent recommendation unit in real time, and ensures the dynamic inheritance of experience.
[0081] Optionally, the human-computer interaction module includes an input unit, a display unit, and a data interaction unit;
[0082] The input unit is data-connected with the display unit, and is used to receive and display user input information, and send the user input information to the task planning module;
[0083] The data interaction unit is data-connected with the display unit, the environment perception module, and the experience management module, and is used to interact with the environment perception module and the experience management module and send the data to the display unit for display.
[0084] It can be seen that the present invention discloses a ship mission adaptive planning system and method, the system includes a resource management module, an environmental perception module, a mission planning module, a mission evaluation module, an experience management module and a human-computer interaction module; through the interaction of various functional modules, ship mission adaptive planning is realized, the consistency and accuracy of task execution are improved, and the decision-making risks caused by human factors are reduced. Through the experience management function, the present invention ensures that the captain's excellent experience can be effectively passed on, shortens the adaptation period of personnel, and improves the overall combat effectiveness and adaptability of the ship team. The innovative architecture and algorithm application of the present invention provide new solutions for the intelligent development of ships, which has important value and practical application prospects.
[0085] Embodiment 2
[0086] See also Figure 2 , Figure 2 1 is a flow chart of a method for adaptive planning of ship missions disclosed in an embodiment of the present invention. Figure 2 The described ship mission adaptive planning method is applied to the field of ship mission planning technology to achieve ship mission adaptive planning, improve the consistency and accuracy of task execution, and reduce the decision-making risk caused by human factors, which is not limited in the embodiments of the present invention. Figure 2 As shown, the ship mission adaptive planning method includes:
[0087] S1, obtaining ship mission adaptive planning data information, and sending the ship mission adaptive planning data information to the mission planning module;
[0088] S2, using a task planning module to process the ship task adaptive planning data information to obtain a ship task adaptive planning result; the ship task adaptive planning result includes a task allocation result and a path planning result;
[0089] S3, sending the ship mission adaptive planning result to the human-computer interaction module and the mission evaluation module;
[0090] S4, using the task evaluation module to evaluate the ship task adaptive planning result, obtain a task evaluation result, and send the task evaluation result to the human-computer interaction module;
[0091] S5, using the human-computer interaction module to display the ship mission adaptive planning result and the mission evaluation result.
[0092] Optionally, the obtaining of ship mission planning data information includes:
[0093] S11, using a human-computer interaction module to obtain task description information;
[0094] Task description information includes: task objectives, task priorities, and task execution conditions;
[0095] S12, using the resource management module to obtain resource management information;
[0096] Resource management information includes the status of own resources, ship type, number, location, fuel and ammunition;
[0097] S13, using the experience management module to obtain historical experience information;
[0098] S14, using the environment perception module to obtain environmental information;
[0099] Environmental information includes sea environment data and enemy dynamic information;
[0100] S15, integrating the task description information, the resource management information, the historical experience information and the environmental information to obtain ship task adaptive planning data information.
[0101] The integration is to construct ship mission adaptive planning data information:
[0102] Ship mission adaptive planning data information = {mission description information, resource management information, historical experience information, environmental information}
[0103] Optionally, the utilizing task planning module to process the ship task adaptive planning data information to obtain the ship task adaptive planning result includes:
[0104] S21, using a task planning module, processing the ship task adaptive planning data information to obtain a training data set;
[0105] S22, using the training data set to train the ship mission adaptive planning model to obtain an optimized ship mission adaptive planning model;
[0106] S23, using the optimized ship mission adaptive planning model to process the ship mission adaptive planning task to be processed, to obtain a ship mission adaptive planning result.
[0107] The ship mission adaptive planning model is a deep reinforcement learning algorithm that trains the agent to learn and optimize mission planning. The algorithm can combine the captain's historical decision data and real-time environmental information to generate an efficient and adaptable mission deployment plan.
[0108] Deep reinforcement learning algorithms include:
[0109] State space definition: Define the state space of the task allocation problem, including ship position, task type, environmental conditions, etc.
[0110] Action space definition: Define possible actions, such as dispatching ships to perform specific tasks, adjusting task priorities, etc.
[0111] Reward function design: Design a reward function to give rewards or penalties based on factors such as task completion, resource consumption, and enemy response.
[0112] The deep reinforcement learning algorithm may be PPO (Proximal Policy Optimization) or A2C (Advantage Actor-Critic), which is not limited in this embodiment.
[0113] Optionally, the utilizing the mission planning module to process the ship mission adaptive planning data information to obtain a training data set includes:
[0114] S211, performing data cleaning on the ship mission adaptive planning data information to remove redundant information and format errors to obtain first preprocessed data information;
[0115] S212, performing data conversion on the first preprocessed data information to obtain second preprocessed data information; the data conversion is to convert the first preprocessed data information into structured data, a table or a database format, to facilitate subsequent processing;
[0116] S213, extracting key information from the second preprocessed data information to obtain third preprocessed data information;
[0117] S214, performing feature extraction on the third preprocessed data information to obtain feature parameter information;
[0118] The specific method is to perform vectorization processing on the third preprocessed data information, convert it into a vector form, and normalize the elements in the vector to 1, so as to obtain characteristic parameter information;
[0119] S215, integrating the feature parameter information to obtain a training data set.
[0120] Optionally, extracting key information from the second preprocessed data information to obtain third preprocessed data information includes:
[0121] S2131, processing the second preprocessed data information to obtain a word grid network;
[0122] The node set of the word grid network is N = {n0, n1, ...}, the arc set of the word grid network is E = {e0, e1, ...}, and each node n i ∈N contains a time stamp t(n i ), arc e u→v represents the arc connecting node u to node v;
[0123] S2132, processing the word grid network to obtain a keyword network;
[0124] The nodes in the keyword network are NS={N0, N1, ...}, Represents the node N in the keyword network i Connect to node N j A set of arcs;
[0125] S2133, extract key information from the keyword network to obtain third preprocessing data information.
[0126] The specific method of steps S2131 to S2133 is:
[0127] The second preprocessed data information is subjected to a lattice network construction, which is a prior art in the art. The set of nodes for constructing the lattice network is N = {n0, n1, ...}, the set of arcs is E = {e0, e1, ...}, and each node n i ∈N contains a time stamp t(n i ), arc e u→v represents the arc connecting node u to node v. Let the nodes in the keyword network be NS = {N0, N1, ...}, Represents the node N in the keyword network i Connect to node N j A set of arcs.
[0128] 1. Any n i ∈N i , n j ∈N i , by t(n i )<t(n j ) we can get i≤j.
[0129] 2. Any n i ∈N i , n j ∈N i , by t(n i )=t(n j ) we can get i=j.
[0130] 3. Any e u→v ∈E, if for u∈N i and v∈N j , e u→v Equivalent to a group Then i≤m≤n≤j. (Here n=m+1, e u→v The two endpoints belong to N m and N n . )
[0131] The word grid network is converted into a keyword network, and the generation algorithm is:
[0132] 1. Convert keywords into corresponding syllable strings: K1…K M (M is the number of syllables of the keyword). Traverse all nodes N in the Lattice network and find the node n that matches K1. k , set N k1 .
[0133] 2. As the end point, search for the nearest node n in reverse time order. k-1 If n k-1 With N k1 There is no arc between them, then n k-1 Also belongs to N k1 , continue to search k-1 Otherwise, the search stops.
[0134] 3. As the starting point, search for similar nodes n backward in chronological order k+1 If n k+1 With N k1 There is no arc between them, then n k+1 belong Continue Search k+1 Otherwise, the search stops.
[0135] 4. For e k→k+1 belong n k+1 Belong to N k2 .
[0136] 5. All included in The arc e k→k+1Form a keyword confusion network.
[0137] 6. For each arc If u belongs to N s , n i Belong to N t When t=s+1, belong otherwise, belong
[0138]
[0139]
[0140] Where w(l) and w(e) are the words corresponding to arcs l and e, sim(·,·) refers to the similarity between two words. It refers to the arc and e overlap in normalized time. In addition, each node n in the Lattice i ∈N contains the time stamp t(n i ) will be used as a constraint to determine the termination of the search.
[0141] The data normalization method is used to confirm the keywords. First, the linguistic probability likelihood score of each candidate is normalized:
[0142]
[0143] Among them, Value is the set of linguistic score values in the keyword network; minValue and maxValue are the minimum and maximum values in the set respectively; x is the score before normalization; y is the score after normalization. After that, the sum of the scores is calculated and marked as the final score of the arc in the keyword network.
[0144] In each keyword network, select the arc candidate with the highest score and determine whether it is a keyword. If it is, output the keyword and time tag; if not, skip and enter the next keyword network. In this way, the keyword detection is completed.
[0145] Optionally, the optimizing ship mission adaptive planning model is used to process the ship mission adaptive planning task to be processed to obtain the ship mission adaptive planning result, including:
[0146] S231, using a human-computer interaction module to obtain a to-be-processed ship mission adaptive planning task;
[0147] S232, using the environment perception module to obtain real-time environment information;
[0148] S233, using the optimized ship task adaptive planning model, processing the to-be-processed ship task adaptive planning task and the real-time environmental information to obtain a task allocation result;
[0149] S234, performing path planning for the ship that performs the task according to the task allocation result to obtain a path planning result.
[0150] Optionally, in step S233, obtaining the task allocation result is achieved by using a task allocation method, specifically:
[0151] Step 1: Set the cost-benefit function and constraints for ship task allocation.
[0152] Step 2: Set the relevant parameters of PSO (particle swarm) algorithm and FSA (simulated annealing) algorithm, including the number of particles and the maximum number of iterations.
[0153] Step 3: Randomly generate a particle swarm in the search space and assign a position and velocity to each particle.
[0154] Step 4: Use the PSO algorithm to perform a global search, update the position and velocity of the particles with the nonlinear inertia factor, and record the overall optimal solution and the individual optimal solution.
[0155] The nonlinear inertia factor is:
[0156]
[0157] Where ω is the nonlinear inertia factor, ω max The value of is 0.95, ω min The value of is 0.6. t represents the number of iterations, t max Indicates the maximum number of iterations;
[0158] Step 5: Use the FSA algorithm to adjust the position of the overall optimal solution of the current iteration number to find a new solution.
[0159] Step 6: Compare the new solution with the optimal solution obtained by the PSO algorithm. If the new solution is better than the current optimal solution, the new solution will be used as the optimal solution for the current number of iterations. Otherwise, the new solution will be accepted with a certain probability.
[0160] Step 7: Determine whether the maximum number of iterations has been reached or other termination conditions have been met.
[0161] Step 8: If the termination condition is not met, return to Step 4 to continue iterating; otherwise, output the ship task allocation result.
[0162] Optionally, performing path planning for a ship that performs a task according to the task allocation result to obtain a path planning result includes:
[0163] S2341, obtaining path planning initial information according to the task allocation result;
[0164] The path planning initial information includes the starting point S(x i ,y i ), target point G(x o ,y o ), the coordinates of the center of the obstacle (x ob1 ,y ob1 )、(x ob2 ,y ob2 )、(x ob3 ,y ob3 ), the radius of the obstacle r1, r2, r3, and the ocean current speed U monitored at the start time c and wind speed U a ;
[0165] S2342, using the path planning model, according to the ocean current speed U c and the wind speed U a , for the starting point S(x i ,y i ), target point G(x o ,y o ), the coordinates of the center of the obstacle (x ob1 ,y ob1 )、(x ob2 ,y ob2 )、(x ob3 ,y ob3 ) and the radius r1, r2, r3 of the obstacle to obtain the initial path;
[0166] The specific method is:
[0167] 1) Initialize the particle population;
[0168] 2) Calculate the forbidden angle of each particle;
[0169] 3) Determine whether the particle is feasible based on the particle's forbidden angle and the ship's turning radius. If feasible, the particle's fitness value takes the total length of the path, and the individual optimal solution and the population optimal solution are updated; if not feasible, the particle's fitness value takes positive infinity, and the individual optimal solution and the population optimal solution are updated;
[0170] 4) Update the particle's velocity and position;
[0171] 5) Determine whether the number of iterations or accuracy requirements have been met. If yes, end; if not, proceed to step 2).
[0172] After the above steps, the initial path is obtained.
[0173] In order to safely and effectively avoid obstacles on the navigation path, it is necessary to calculate the no-entry angles of the path points on the circles of each dimension. The specific no-entry angles are set according to the actual obstacles and are not limited in this embodiment.
[0174] S2343, the ship performing the mission sails along the initial path;
[0175] S2344, the ship continuously monitors the changes in sea current speed and wind speed during navigation;
[0176] S2345, when the ocean current speed and the wind speed change, the path planning model is used to calculate the current position information of the ship and the target point G(x o ,y o ), the coordinates of the center of the obstacle (x ob1 ,y ob1 )、(x ob2 ,y ob2 )、(x ob3 ,y ob3 ) and the radius r1, r2, r3 of the obstacle are processed to obtain an updated path, and the ship navigates along the updated path;
[0177] S2346, determine whether the ship has reached the target point G(x o ,y o );
[0178] When the judgment result is yes, output the path planning result; when the judgment result is no, execute S2344.
[0179] Optionally, the method of performing path planning for the ship performing the task according to the task allocation result and obtaining the path planning result is based on a combination of reinforcement learning and a classical algorithm:
[0180] (1) Environmental modeling: Use real-time environmental data to build a sea area map, including obstacles, enemy locations, and safe areas.
[0181] (2) Path search algorithm: Combined with A* algorithm or Dijkstra algorithm, a preliminary path is generated.
[0182] (3) Reinforcement learning optimization: The DRL algorithm is used to further optimize the path, making it more flexible to cope with dynamically changing environments.
[0183] It can be seen that the present invention discloses a ship mission adaptive planning system and method, the system includes a resource management module, an environmental perception module, a mission planning module, a mission evaluation module, an experience management module and a human-computer interaction module; through the interaction of various functional modules, ship mission adaptive planning is realized, the consistency and accuracy of task execution are improved, and the decision-making risks caused by human factors are reduced. Through the experience management function, the present invention ensures that the captain's excellent experience can be effectively passed on, shortens the adaptation period of personnel, and improves the overall combat effectiveness and adaptability of the ship team. The innovative architecture and algorithm application of the present invention provide new solutions for the intelligent development of ships, which has important value and practical application prospects.
[0184] Embodiment 3
[0185] See also Figure 3 , Figure 3 1 is a flow chart of another method for adaptive planning of ship missions disclosed in an embodiment of the present invention. Figure 3 The described ship mission adaptive planning method is applied to the field of ship mission planning technology to achieve ship mission adaptive planning, improve the consistency and accuracy of task execution, and reduce the decision-making risk caused by human factors, which is not limited in the embodiments of the present invention. Figure 3 As shown, the ship mission adaptive planning method includes:
[0186] 1. Overall design
[0187] System architecture: This system adopts an advanced distributed architecture, integrating modules such as data collection and processing, intelligent decision-making, experience management and human-computer interaction to achieve efficient information processing and intelligent decision-making support.
[0188] Deep reinforcement learning algorithm: Using deep reinforcement learning algorithm, the agent is trained to learn and optimize the task planning strategy. The algorithm can combine the captain's historical decision data and real-time environmental information to generate efficient and adaptable task deployment solutions.
[0189] Adaptive mission planning: The system can dynamically adjust mission planning based on real-time collected data, such as the sea environment, enemy dynamics and own resources, to ensure that the ship can perform the optimal mission deployment in different scenarios.
[0190] Task description module: includes detailed information of the task, such as task objectives, task priorities, task execution conditions, etc.
[0191] Resource Management Module: Manage the status of your own resources, such as ship type, quantity, location, fuel and ammunition.
[0192] Environmental perception module: collects sea environment data, enemy dynamics and other information in real time as input for mission planning.
[0193] Task evaluation module: Evaluate the feasibility and priority of tasks based on real-time data and dynamically adjust task plans.
[0194] Task allocation algorithm:
[0195] The task allocation algorithm is based on the deep reinforcement learning (DRL) method and mainly includes the following steps:
[0196] State space definition: Define the state space of the task allocation problem, including ship position, task type, environmental conditions, etc.
[0197] Action space definition: Define possible actions, such as dispatching ships to perform specific tasks, adjusting task priorities, etc.
[0198] Reward function design: Design a reward function to give rewards or penalties based on factors such as task completion, resource consumption, and enemy response.
[0199] Algorithm selection: Use deep reinforcement learning algorithms such as PPO (Proximal Policy Optimization) or A2C (Advantage Actor-Critic) for training to optimize task allocation strategies.
[0200] Path planning: Path planning is based on the combination of reinforcement learning and classical algorithms:
[0201] Environmental modeling: Use real-time environmental data to build a sea area map, including obstacles, enemy locations, and safe areas.
[0202] Path search algorithm: Combined with A* algorithm or Dijkstra algorithm, a preliminary path is generated.
[0203] Reinforcement learning optimization: Use the DRL algorithm to further optimize the path to make it more flexible to cope with dynamically changing environments.
[0204] Intelligent inheritance of experience: By building an experience knowledge base and intelligent recommendation system, the captain's excellent experience can be digitally stored and intelligently inherited. New captains can quickly master job skills and tactical strategies through systematic learning of historical cases, simulation training and real-time feedback.
[0205] The experience intelligent inheritance system mainly includes the experience knowledge base, intelligent recommendation system and simulation training module:
[0206] Experience knowledge base: collect and digitally store the operational experience and decision-making records of excellent captains.
[0207] Intelligent recommendation system: Analyzes the learning needs and progress of new captains, and recommends personalized learning resources and simulation training tasks.
[0208] Simulation Training: Based on real historical cases, it provides a virtual training environment where new captains can improve their skills.
[0209] Real-time feedback: Collect operational data from new captains during actual missions, update the knowledge base and recommendation system in real time, and ensure dynamic inheritance of experience.
[0210] Human-computer interaction interface: A user-friendly human-computer interaction interface is designed to support natural interaction between the captain and the system. The interface provides intuitive mission planning display, experience learning progress feedback, and intelligent decision support functions to improve user experience and decision-making efficiency.
[0211] Interaction with other modules: The human-computer interaction interface serves as a bridge between the user and the system. Its main functions include:
[0212] Mission planning display: Displays the mission allocation results and path planning diagram for the captain's reference.
[0213] Experience Learning Progress Feedback: Displays new captains’ learning progress and recommended learning resources.
[0214] Intelligent Decision Support: Provide real-time decision suggestions and receive feedback and adjustment suggestions from the captain.
[0215] Data interaction: Obtain real-time data from the data collection and processing module, obtain task planning results from the intelligent decision-making module, and exchange learning and training data with the experience management module.
[0216] 2. Implementation Method
[0217] 1. Data collection and processing
[0218] Data collection: Before the rescue mission begins, the system collects deployment mission information, personnel information, etc.
[0219] Data preprocessing: convert the format and extract features of the collected data.
[0220] Data storage: Store the preprocessed data in the database.
[0221] 2. Intelligent decision-making and task planning
[0222] Requesting experience knowledge: Based on the current rescue mission requirements, the system requests historical experience and decision records related to similar rescue missions from the experience management component.
[0223] Mission planning generation: The intelligent decision-making component uses deep reinforcement learning models (such as PPO) to combine real-time data and retrieved empirical knowledge to generate mission planning strategies. For example, the system recommends the most suitable rescuer deployment action plan for each person.
[0224] Task evaluation: The system evaluates the generated task plan to ensure the feasibility and optimality of the task. The evaluation criteria include task completion time, resource consumption, and task success rate.
[0225] 3. Task display and feedback collection
[0226] Mission display: The user interaction component displays the mission planning strategy to the captain and operators through an intuitive graphical interface, such as personnel planning, mission planning, etc.
[0227] Feedback collection: After reviewing the mission plan, the captain and operators provide feedback, such as adjusting personnel, adding spare resources, etc. The system collects these feedbacks and records them.
[0228] 4. Feedback processing and knowledge updating
[0229] Feedback processing: The user interaction component passes the feedback information from the captain and operators to the intelligent decision-making component and the experience management component. The intelligent decision-making component adjusts the mission planning based on the feedback, such as recalculating the rescue personnel arrangement.
[0230] Knowledge update: The experience management component updates the experience knowledge base based on the feedback from the captain and operators. For example, the successful experience and feedback records of this rescue mission are stored in the knowledge base for reference in similar missions in the future.
[0231] 5. Real-time adjustment and optimization
[0232] Real-time monitoring: During the task execution process, the system continuously monitors real-time data, such as task execution status, personnel status, etc.
[0233] Dynamic adjustment: Based on real-time monitoring data, the intelligent decision-making component adjusts task planning at any time.
[0234] Continuous optimization: After the task is completed, the system summarizes and analyzes the entire task execution process to further optimize the task planning strategy and experience knowledge base.
[0235] Figure 2 Participants include:
[0236] Ship Mission Adaptive Planning System: Represents the entire system, but there is no direct activity in this diagram, instead the working of the system is shown through the interaction of other components.
[0237] People: Represents the human users who interact with the system.
[0238] Data collection and processing: responsible for collecting and processing historical experience data and deployment table data from different structures, etc.
[0239] Intelligent decision-making: Utilize the information provided by data perception and the knowledge in the experience management module to formulate mission planning strategies.
[0240] Experience management: Manage the experience knowledge base in the system and provide the required knowledge and experience for intelligent decision-making.
[0241] User interaction: Responsible for presenting the planning plan to the captain and receiving his feedback and operation information.
[0242] Figure 2 The interactive process:
[0243] The process starts with the data acquisition and processing component, which collects real-time data and sends it to the intelligent decision-making component.
[0244] The intelligent decision-making component requests the experience management component to retrieve relevant experience knowledge.
[0245] The experience management component returns the experience knowledge to the intelligent decision-making component, which uses this knowledge and data to generate task planning strategies.
[0246] The generated planning strategies are presented to personnel through user interaction components.
[0247] After viewing the planning scheme, personnel provide feedback and operational information, which is passed back to the system through the user interaction component.
[0248] The user interaction component sends the captain's feedback and operation information to the experience management component to update the knowledge base.
[0249] The experience management component updates the knowledge base based on the feedback and uses the updated knowledge for future decisions of the system.
[0250] In addition, the user interaction component also transmits the captain's operations and feedback to the intelligent decision-making component to further adjust and optimize the planning strategy.
[0251] Figure 2 Hardware foundation of the system: Considering the limitations of domestic computers, we will make full use of their computing resources. As computer hardware is upgraded, we will tune the algorithm to match higher processing power and storage capacity. In addition, we will ensure the scalability of the system architecture so that more hardware resources can be easily integrated in the future.
[0252] Figure 2 The algorithm models involved in the system include:
[0253] Deep reinforcement learning algorithms: Use complex deep reinforcement learning algorithms, such as PPO (Proximal Policy Optimization) or A2C (Advantage Actor-Critic), to handle high-dimensional state space and action space.
[0254] Hierarchical Reinforcement Learning (HRL): HRL is introduced to handle long-term planning and task decomposition, enabling the system to cope with complex tasks more effectively.
[0255] Simulators and data augmentation: Build simulators to reproduce historical scenarios and use data augmentation techniques to expand the dataset and improve the generalization ability of the model.
[0256] Personalized recommendation system: Combined with personnel data, a recommendation system is implemented to provide personalized decision support.
[0257] Specific examples
[0258] Suppose a ship needs to perform a complex rescue mission involving multiple areas and different threat levels. The system first learns how to rescue effectively from historical experience data and trains in simulation scenarios. Then, combined with the current staffing and skill level, the system recommends the most suitable rescuer deployment action plan for each person. Personnel can provide feedback based on actual conditions, and the system continuously optimizes the recommended strategy based on this feedback.
[0259] This system uses deep reinforcement learning algorithms to achieve efficient adaptive planning of ship missions, improve the consistency and accuracy of task execution, and reduce decision-making risks caused by human factors. Through the intelligent inheritance of experience function, this system ensures that the captain's excellent experience can be effectively inherited, shortens the adaptation period of personnel, and improves the overall combat effectiveness and adaptability of the ship team. The innovative architecture and algorithm application of this system provide new solutions for the intelligent development of ships, which has important value and practical application prospects. At the same time, the system also helps to improve the level of informatization and combat capabilities, and contribute to safeguarding national maritime rights and interests and regional stability.
[0260] The device embodiments described above are only illustrative, wherein the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, i.e., they may be located in one place, or they may be distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative work.
[0261] Through the specific description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the above technical solution can be essentially or partly contributed to the prior art in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable rewritable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.
[0262] Finally, it should be noted that the ship mission adaptive planning system and method disclosed in the embodiments of the present invention disclose only the preferred embodiments of the present invention, which are only used to illustrate the technical solution of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, it should be understood by those skilled in the art that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features therein can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A ship mission adaptive planning system, It is characterized in that The system includes a resource management module, an environment perception module, a task planning module, a task evaluation module, an experience management module and a human-computer interaction module; The resource management module is data-connected with the mission planning module to manage own resource information; The environment perception module is data-connected with the human-computer interaction module and the task planning module to collect sea area data information in real time; The task evaluation module is data-connected with the task planning module and the human-computer interaction module, and is used to perform task evaluation according to real-time data, and send the task evaluation result to the human-computer interaction module for display; The experience management module is data-connected with the task planning module and the human-computer interaction module for managing experience information; The human-computer interaction module is data-connected with the task planning module, the environment perception module, and the experience management module for information input and display.
2. The ship mission adaptive planning system according to claim 1, It is characterized in that The experience management module includes an experience knowledge base and an intelligent recommendation unit; The experience knowledge base is used to store the digitized information of the captain’s experience; The intelligent recommendation unit is used to receive instructions from the human-computer interaction module, provide simulation training for the new captain, and feed back the simulation training results of the new captain to the human-computer interaction module in real time.
3. The ship mission adaptive planning system according to claim 1, It is characterized in that The human-computer interaction module includes an input unit, a display unit, and a data interaction unit; The input unit is data-connected with the display unit, and is used to receive and display user input information, and send the user input information to the task planning module; The data interaction unit is data-connected with the display unit, the environment perception module, and the experience management module, and is used to interact with the environment perception module and the experience management module and send the data to the display unit for display.
4. A ship mission adaptive planning method, It is characterized in that The method is applied to the ship mission adaptive planning system according to any one of claims 1 to 3, and the method comprises: S1, obtaining ship mission adaptive planning data information, and sending the ship mission adaptive planning data information to the mission planning module; S2, using a task planning module to process the ship task adaptive planning data information to obtain a ship task adaptive planning result; the ship task adaptive planning result includes a task allocation result and a path planning result; S3, sending the ship mission adaptive planning result to the human-computer interaction module and the mission evaluation module; S4, using the task evaluation module to evaluate the ship task adaptive planning result, obtain a task evaluation result, and send the task evaluation result to the human-computer interaction module; S5, using the human-computer interaction module to display the ship mission adaptive planning result and the mission evaluation result.
5. The ship mission adaptive planning method according to claim 4, It is characterized in that The obtaining of ship mission planning data information includes: S11, using a human-computer interaction module to obtain task description information; S12, using the resource management module to obtain resource management information; S13, using the experience management module to obtain historical experience information; S14, using the environment perception module to obtain environmental information; S15, integrating the task description information, the resource management information, the historical experience information and the environmental information to obtain ship task adaptive planning data information.
6. The method for adaptive ship mission planning according to claim 4, It is characterized in that The task planning module is used to process the ship task adaptive planning data information to obtain the ship task adaptive planning result, including: S21, using a task planning module, processing the ship task adaptive planning data information to obtain a training data set; S22, using the training data set to train the ship mission adaptive planning model to obtain an optimized ship mission adaptive planning model; S23, using the optimized ship mission adaptive planning model to process the ship mission adaptive planning task to be processed, to obtain a ship mission adaptive planning result.
7. The ship mission adaptive planning method according to claim 6, It is characterized in that The mission planning module is used to process the ship mission adaptive planning data information to obtain a training data set, including: S211, performing data cleaning on the ship mission adaptive planning data information to remove redundant information and format errors to obtain first preprocessed data information; S212, performing data conversion on the first preprocessed data information to obtain second preprocessed data information; S213, extracting key information from the second preprocessed data information to obtain third preprocessed data information; S214, performing feature extraction on the third preprocessed data information to obtain feature parameter information; S215, integrating the feature parameter information to obtain a training data set.
8. The method for adaptive ship mission planning according to claim 7, It is characterized in that The step of extracting key information from the second preprocessed data information to obtain third preprocessed data information includes: S2131, processing the second preprocessed data information to obtain a word grid network; The node set of the word grid network is N = {n 0 ,n 1 ,…}, the arc set of the word grid network is E = {e 0 ,e 1 ,…}, each node n i ∈N contains a time stamp t(n i ), arc e u→v represents the arc connecting node u to node v; S2132, processing the word grid network to obtain a keyword network; The nodes in the keyword network are NS={N 0 ,N 1 ,…}, Represents the node N in the keyword network i Connect to node N j A set of arcs; S2133, extract key information from the keyword network to obtain third preprocessing data information.
9. The method for adaptive ship mission planning according to claim 6, It is characterized in that The method of using the optimized ship mission adaptive planning model to process the ship mission adaptive planning task to be processed to obtain the ship mission adaptive planning result includes: S231, using a human-computer interaction module to obtain a to-be-processed ship mission adaptive planning task; S232, using the environment perception module to obtain real-time environment information; S233, using the optimized ship task adaptive planning model, processing the to-be-processed ship task adaptive planning task and the real-time environmental information to obtain a task allocation result; S234, performing path planning for the ship that performs the task according to the task allocation result to obtain a path planning result.
10. The ship mission adaptive planning method according to claim 9, It is characterized in that The step of performing path planning for a ship that performs a task according to the task allocation result to obtain a path planning result includes: S2341, obtaining path planning initial information according to the task allocation result; The path planning initial information includes the starting point S(x i ,y i ), target point G(x o ,y o ), the coordinates of the center of the obstacle (x ob1 ,y ob1 )、(x ob2 ,y ob2 )、(x ob3 ,y ob3 ), the radius of the obstacle r 1 、r 2 、r 3 , the ocean current speed U monitored at the start time c and wind speed U a ; S2342, using the path planning model, according to the ocean current speed U c and the wind speed U a , for the starting point S(x i ,y i ), target point G(x o ,y o ), the coordinates of the center of the obstacle (x ob1 ,y ob1 )、(x ob2 ,y ob2 )、(x ob3 ,y ob3 ) and the radius r of the obstacle 1 、r 2 、r 3 Processing is performed to obtain the initial path; S2343, the ship performing the mission sails along the initial path; S2344, the ship continuously monitors the changes in sea current speed and wind speed during navigation; S2345, when the ocean current speed and the wind speed change, the path planning model is used to calculate the current position information of the ship and the target point G(x o ,y o ), the coordinates of the center of the obstacle (x ob1 ,y ob1 )、(x ob2 ,y ob2 )、(x ob3 ,y ob3 ) and the radius r of the obstacle 1 、r 2 、r 3 Processing is performed to obtain an updated path, and the ship sails along the updated path; S2346, determine whether the ship has reached the target point G(x o ,y o ); When the judgment result is yes, output the path planning result; when the judgment result is no, execute S2344.
Citation Information
Patent Citations
Unmanned ship global path planning method and electronic device
CN109931943A
Ship route planning method fusing ship driver experience
CN112185171A
Unmanned ship navigation autonomous decision-making system and autonomous navigation method
CN117193321A
Ship navigation method, terminal equipment and readable storage medium
CN117433540A
Shipborne unmanned commanding and dispatching and big data airborne monitoring intelligent system
CN117784795A
