A ship path planning method and system based on a Beidou grid code

Abstract

This invention discloses a ship path planning method and system based on BeiDou grid codes, belonging to the field of path planning technology. The method includes acquiring planning data of the area to be planned for the ship, classifying the ship into priority levels, combining prior paths, priority levels, and planning indicators, generating a BeiDou grid code-based grid map through a real-world BeiDou grid code management module for global path planning, conducting collision risk analysis based on real-time sensing data from navigation and positioning systems, AIS, gyrocompass, logs, waves, and weather forecasts, evaluating the path planning effect based on the post-planning task completion efficiency, and adaptively adjusting the grid code level based on different indicators such as planning effect or energy efficiency. This method analyzes and mines real-time sensing data such as port information, ship route information, historical path planning data, and meteorological information to recommend routes, improve the safety, economy, and reliability of routes, and enhance the ship's intelligence level and operational efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of path planning, and particularly relates to a ship path planning method and system based on a Beidou grid code BACKGROUND

[0002] Intelligent shipping is an inevitable trend of future water transportation development, and is also the top priority of the development and research of China's shipping industry. With the gradual recovery of shipping demand and the application of autonomous ships, the path and navigation environment of ships will inevitably be more complex and diverse. The significance of ship path planning is to ensure the safety and efficient transportation of ships and goods. Choosing the right route can reduce the risk of natural disasters encountered by ships at sea, and also reduce the sailing time and cost.

[0003] The existing technology mainly relies on global path planning, local path planning or a combination of the two. Global path planning requires obtaining electronic charts and environmental charts, and then using path planning algorithms to optimize the path, which is only suitable for long-distance channel planning and dealing with static obstacles, and cannot respond to sudden situations (waves, marine animals, and the appearance of other ships). The existing local path planning on the sea mainly relies on algorithms such as A* and VFH* to handle local paths and obstacles, lacks a visual recognition system, and cannot efficiently and accurately respond to complex situations. At the same time, traditional path planning algorithms often use latitude and longitude coordinates to represent spatial information. However, latitude and longitude are described in degrees-minutes-seconds, which is complex and not conducive to fast calculation. Representing a planar space with latitude and longitude coordinates requires determining the latitude and longitude coordinates of several corner points and the center point of the space, which is not only complex and tedious but also requires verification of the accuracy of the calculation results. The Beidou grid code represents a region location with an integer number, and compared with the grid system currently used in the United States (MGRS, USNG, GOOGLE, etc.), the Beidou grid has technical advantages in the inheritance of existing spatial coordinate systems, the construction of a three-dimensional global space, and the binary integer coding of key technologies, mainly including: 1) maintaining high consistency with the commonly used latitude and longitude system of sea and land data, which can be directly converted; 2) a three-dimensional seamless and nested subdivision system with a globally unique code; 3) a geographic spatial coding algebraic calculation method based on binary operation mechanism, with high-efficiency binary calculation capability, realizing efficient coding operation of three-dimensional spatial information organization, storage, transmission, distribution, service, etc.

[0004] Therefore, in view of the needs of ship navigation safety and efficiency as well as the development of intelligent ships, this paper proposes a ship route planning method based on Beidou grid codes, which makes full use of the environmental information provided by electronic nautical charts. It integrates marine and land geospatial data with grid code information, and uses data mining technology to analyze historical data such as port information and ship information, as well as real-time sensing data such as meteorological information. It recommends routes according to different indicators, improves the safety, economy and reliability of routes, and further enhances the intelligence level and operational efficiency of ships in complex marine environments. Summary of the Invention

[0005] To address the aforementioned technical challenges, a ship path planning method and system based on BeiDou grid codes is proposed.

[0006] The technical solution of this invention is: a ship path planning method based on Beidou grid codes, comprising the following steps:

[0007] S1. Obtain planning data for the area to be planned for ships;

[0008] The planning data includes the starting point, the ending point, the ship's dimensions, and the ship's intended navigation depth.

[0009] Based on the stated start and end points, the data is converted into BeiDou grid code data to determine the ship's navigation area location information.

[0010] Based on the aforementioned starting and ending points, large-scale weather forecast data are determined;

[0011] S2. Priority levels are assigned based on different types of vessels. When multiple vessels are present, the vessel with the highest priority level is given priority in route planning.

[0012] S3. Combining prior paths, priority levels, and planning indicators, global path planning is performed by generating a grid map based on the BeiDou grid code management module in the real environment.

[0013] The aforementioned prior path is an instance library established based on historical route planning results and actual navigation routes, which depends on the number and accuracy of historical instances;

[0014] The aforementioned grid map based on BeiDou grid code = spatial BeiDou grid code + BeiDou segmentation time code + object + attribute + associated information;

[0015] The global path planning method adopts an improved A* algorithm based on the BeiDou grid code grid map, which is based on priority level and planning indicators.

[0016] By combining prior paths and global path planning, an initial globally optimal path is obtained;

[0017] S4 transmits real-time sensing data from navigation and positioning systems, AIS, gyrocompass, log, wave data, and weather forecasts to the local dynamic path planning and analysis module.

[0018] S5. The local dynamic path planning and analysis module determines the grid code level and quickly generates a local real-time grid map from the grid code management module to perform collision risk analysis.

[0019] Based on real-time collected data and combined with the "BeiDou grid code grid map" data of the real environment, BeiDou grid location code positioning and matching are performed. Through the grid code hierarchy, the local environment is optimized, adjusted and refined, and the entity types in the grid are identified to generate a "local real-time grid map" for collision risk analysis.

[0020] S6. Establish memory data, classify danger levels based on memory data, and perform local path planning if there is a risk of collision;

[0021] The memory data consists of the characteristics, location, movement status, and danger level of non-target objects that were detected and avoided during the path planning at sea. The non-target objects are maritime objects that affect the path of the mission or the safety of the mission.

[0022] By combining the entity types in the local environment of the "local real-time mesh map", the system analyzes in real time whether the ship maintains a safe distance from the surrounding mesh and whether a collision will occur when it moves in the local environment. If a collision will occur, the system generates a local optimized path based on the "local real-time mesh map" and updates the memory data.

[0023] S7. Evaluate the effectiveness of path planning based on the efficiency of task completion after planning, and adaptively adjust the grid code level according to different indicators such as planning effectiveness or energy efficiency.

[0024] If the dangerous situation of the target is not improved, it is considered a poor plan. In this case, the accuracy, i.e. the grid code level, is increased in order to seek better results in the next planning.

[0025] If the current planning task completion efficiency exceeds the average of the historical task completion efficiency, it is considered well planned and the grid code level is reduced. If the current planning task completion efficiency is lower than the average of the historical task completion efficiency, it is considered poor planned and the accuracy, i.e., the grid code level, is increased.

[0026] If the planned indicators do not meet the requirements, the planning effect is considered poor, and the accuracy, i.e., the grid code level, is reduced.

[0027] Furthermore, the objects mentioned in step S3 include static obstacles and dynamic obstacles. Static obstacles include danger zones, restricted areas, islands, and areas that do not meet the set navigation depth of the vessel. Dynamic obstacles include vessels and navigation marks.

[0028] Furthermore, in step S3, the global path planning method based on the Beidou grid code grid map to improve A* has an evaluation function f(n) = g(n) + P × h(n).

[0029] Where, P = min(abs(x) curr -x souce ),abs(abs(y curr -y source ))+min(abs(x curr -x goal ),abs(abs(y curr -y goal ))

[0030] g(n) represents the actual cost from the initial node to the current node, P represents the changeover interference value, and the ship should sail in one direction as much as possible. Points with the same cost that are closer to the starting or ending point have higher priority. h(n) represents the estimated cost of the grid code of the current ship position and the grid code of the target point. (x curr ,y curr (x) represents the current BeiDou grid code. next ,y next ) represents the BeiDou grid code of the next node of the current node, (x goal ,y goal () is the BeiDou grid code of the target point;

[0031] The costs include Euclidean distance, Manhattan distance, diagonal distance, and Chebyshev distance; this method uses Manhattan distance.

[0032] g(n) = abs(x) curr -x next )+abs(y curr -y next )

[0033]

[0034] λ represents the threshold constant; ω1 and ω2 are weights, based on the original heuristic factor h(n). When they are higher than the threshold λ, the weights increase, and the algorithm searches faster; when they are lower than the threshold λ, the weights decrease, and the optimal path is considered first. The threshold is dynamically selected based on factors such as computer performance and map size, and the weight values ​​are also adjusted according to the size and complexity of the map. M is the number of static obstacles from the current node n to the target node, and N is the number of dynamic obstacles.

[0035] D(n) collision risk value:

[0036]

[0037] Where θ is the scaling factor, d is the shortest distance between the current node n and the obstacle in the path, and D is the range of impact of the obstacle collision risk;

[0038] The normalized conversion coefficient for converting safety (planning indicators) into safety (planning indicators) costs.

[0039] Furthermore, the collision risk analysis described in step S5 uses a collision risk value D(n).

[0040]

[0041] Furthermore, the memory data mentioned in step S6 consists of the characteristics, location, movement status, and hazard level of non-target objects that were detected and avoided during the path planning at sea. These non-target objects are maritime objects that affect the path of the mission or the safety of the user.

[0042] Furthermore, step S6 classifies the risk level based on the memory data, specifically including the following steps:

[0043] If a target terrain or object matching the memory data is detected, a higher processing priority is set.

[0044] Classify objects in the memory data that have previously posed a threat to one's safety into different levels of danger.

[0045] Classify the danger level of targets that were previously avoided in the path planning in the memory data;

[0046] If the target that was dodged was stationary or slow-moving, the memory path is used to determine whether there is a conflict between the memory path and the current best global path. If there is a conflict, the dodged target is classified as the first danger level.

[0047] If the target that was dodged is moving, then the dodged target is classified as the second danger level.

[0048] Another aspect of the present invention provides a ship path planning system based on BeiDou grid codes, comprising: a priori path generation module, a BeiDou grid code management module, a real-time data acquisition module, a local real-time grid map construction module, a path planning template, a path collision risk analysis module, and a user information interaction module.

[0049] (1) Prior path generation module: extract historical route planning results, ship historical operation data, AIS historical trajectory data, etc., perform digital and standardized data preprocessing, establish an instance library through trajectory clustering, navigation experience, etc., and obtain prior paths for different types of ships.

[0050] (2) The BeiDou grid code management module manages the basic data of BeiDou grid location codes, and encodes the real environment, such as ships, dangerous areas, restricted navigation areas, and islands, based on BeiDou grid codes to generate a "basic BeiDou grid base map," i.e., BeiDou grid map data, which is provided to various modules for querying and access. The BeiDou grid map = spatial BeiDou grid code + BeiDou subdivision time code + object + attribute + associated information.

[0051] (3) Real-time data acquisition module, which collects data from shipboard equipment in real time. Shipboard equipment includes navigation and positioning system, AIS, gyrocompass, and speedometer. The data collected by the real-time data acquisition module also includes wave and weather forecast data. The collected data is used for the path planning module.

[0052] (4) The local real-time grid map construction module includes performing Beidou grid location code positioning and matching based on real-time collected data and real-world environment grid map data, optimizing, adjusting and refining the local environment through grid code hierarchy, identifying the entity types in the grid, generating a "local real-time grid map" and passing it to the path planning module.

[0053] (5) Path planning template, including a global path planning module and a local dynamic path planning analysis module:

[0054] The global path planning model uses BeiDou grid map data, combined with prior paths, priority levels, and planning indicators, to optimize and adjust the environment through a grid code hierarchy. It generates global paths based on improved A* and other algorithms, taking into account planar obstacles that have a significant impact on the global planning effect, such as danger zones, no-navigation zones, and islands.

[0055] The local dynamic path planning and analysis module optimizes and adjusts the environment based on the local real-time grid map data from the local real-time grid map construction module through a grid code hierarchy, and generates local paths based on improved A* and other algorithms to achieve path optimization in the local environment.

[0056] (6) The path collision risk analysis module includes combining the entity types in the environment, calculating the closest distance between the current node and the obstacle in the path, and analyzing in real time whether the ship maintains a safe distance from the surrounding obstacles and whether a collision will occur when it moves.

[0057] (7) User information interaction module: obtain planning data of the area to be planned, including starting point, ending point, ship size, and ship set navigation depth; convert the starting point and ending point into Beidou grid code data to determine the ship navigation area location information; determine large-scale weather forecast data based on the starting point and ending point.

[0058] The beneficial effects of this invention are:

[0059] This invention, based on the BeiDou grid code technology system, analyzes historical navigation data such as port information and ship information, as well as real-time perceived data such as meteorological information, in ship route planning, and performs local route optimization. This invention employs the BeiDou grid code technology system, utilizing the efficient binary computing capabilities of integer coding to achieve efficient coded operations for the organization, storage, transmission, distribution, and service of three-dimensional spatial information. The navigation route planning results and / or actual navigation routes are stored in data form as a route planning example library, which is more accurate and closely reflects the usual practices of crew members. Based on practical considerations, these can be directly selected as preliminary routes, avoiding disagreements with ship navigators and ensuring smooth ship operations. This invention can recommend routes based on different indicators, providing a safe, economical, and effective solution for dealing with complex and urgent maritime traffic situations, serving maritime safety assurance, and realizing the intelligent and scientific nature of maritime safety assurance. Attached Figure Description

[0060] Figure 1 This is a flowchart of a ship path planning method based on BeiDou grid codes according to the present invention.

[0061] Figure 2 This is a system composition diagram of a ship routing system based on BeiDou grid codes according to the present invention. Detailed Implementation

[0062] 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 embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0063] like Figure 1 , Figure 2 As shown, this invention discloses a ship path planning method based on BeiDou grid codes, comprising the following steps:

[0064] S1. Obtain planning data for the area to be planned for ships;

[0065] The planning data includes the starting point, the ending point, the ship's dimensions, and the ship's intended navigation depth.

[0066] Based on the stated start and end points, the data is converted into BeiDou grid code data to determine the ship's navigation area location information.

[0067] Based on the stated starting and ending points, large-scale weather forecast data are determined.

[0068] S2. Priority levels are assigned based on different types of vessels. When multiple vessels are present, the vessel with the highest priority level is given priority in route planning.

[0069] S3. Combining prior paths, priority levels, and planning indicators, the BeiDou grid code management module generates a grid map based on the BeiDou grid code for global path planning in the real environment.

[0070] The aforementioned prior path is an instance library established based on historical route planning results and actual navigation routes, which depends on the number and accuracy of historical instances.

[0071] The aforementioned grid map based on BeiDou grid code = spatial BeiDou grid code + BeiDou segmentation time code + object + attribute + associated information.

[0072] The objects mentioned include static obstacles and dynamic obstacles. Static obstacles include danger zones, restricted areas, islands, and areas that do not meet the set navigation depth of ships. Dynamic obstacles include ships and navigation marks.

[0073] The global path planning method adopts an improved A* algorithm based on the BeiDou grid code grid map, which is based on priority level and planning indicators.

[0074] The evaluation function of the improved A* algorithm is f(n) = g(n) + P × h(n).

[0075] Where, P = min(abs(x) curr -x souce ),abs(abs(y curr -y source ))+min(abs(x curr -x goal ),abs(abs(y curr -y goal ))

[0076] g(n) represents the actual cost from the initial node to the current node, P represents the changeover interference value, and the ship should sail in one direction as much as possible. Points with the same cost that are closer to the starting point or the ending point have higher priority. h(n) represents the estimated cost of the grid code of the current ship position and the grid code of the target point. The cost includes Euclidean distance, Manhattan distance, diagonal distance, and Chebyshev distance. This algorithm uses Manhattan distance.

[0077] g(n) = abs(x) curr -x next )+abs(y curr -y next )

[0078]

[0079] λ represents the threshold constant; ω1 and ω2 are weights, based on the original heuristic factor h(n). When they are higher than the threshold λ, the weights increase, resulting in a faster search speed; when they are lower than the threshold λ, the weights decrease, prioritizing the optimal path. The threshold is dynamically selected based on factors such as computer performance and map size, and the weight values ​​are also adjusted according to the size and complexity of the set map. M is the number of static obstacles from the current node n to the target node, and N is the number of dynamic obstacles.

[0080] D(n) collision risk value:

[0081]

[0082] Where θ is the scaling factor, d is the shortest distance between the current node n and the obstacle in the path, and D is the range of impact of the obstacle collision risk.

[0083] The conversion factor is a normalized conversion coefficient for converting a safety (indicator) into a safety (indicator) cost.

[0084] (x curr ,y curr (x) represents the current BeiDou grid code. next ,y next ) represents the BeiDou grid code of the next node of the current node, (x goal ,y goal () is the Beidou grid code where the target point is located.

[0085] By combining prior paths and global path planning, an initial globally optimal path is obtained.

[0086] S4, real-time sensing data from navigation and positioning systems, AIS, gyrocompass, odometer, wave data, and weather forecasts are transmitted to the local dynamic path planning and analysis module.

[0087] S5. The local dynamic path planning and analysis module determines the grid code level and quickly generates a local real-time grid map from the grid code management module to perform collision risk analysis.

[0088] Based on real-time collected data and combined with the "BeiDou grid code-based grid map" data provided by the BeiDou grid location code management module in the real environment, BeiDou grid location code positioning and matching are performed. Through the grid code hierarchy, the local environment is optimized, adjusted and refined, and the entity types in the grid are identified to generate a "local real-time grid map" for collision risk analysis.

[0089]

[0090] S6. Establish memory data, classify the danger level based on the memory data, and if there is a risk of collision, perform local path planning based on the danger level.

[0091] The memory data consists of the characteristics, location, movement status, and danger level of non-target objects that were detected and avoided during the path planning at sea. The non-target objects are maritime objects that affect the path of the mission or the safety of the mission.

[0092] As a preferred technical solution, the specific steps include:

[0093] If a target terrain or object matching the memory data is detected, a higher processing priority is set.

[0094] Classify objects in the memory data that have previously posed a threat to one's safety into different levels of danger.

[0095] Classify the danger level of targets that were previously avoided in the path planning in the memory data;

[0096] If the target that was dodged was stationary or slow-moving, the memory path is used to determine whether there is a conflict between the memory path and the current best global path. If there is a conflict, the dodged target is classified as the first danger level.

[0097] If the target that was dodged is moving, then the dodged target is classified as the second danger level.

[0098] By combining the entity types in the local environment of the "local real-time mesh map", the system analyzes in real time whether the ship maintains a safe distance from the surrounding mesh and whether a collision will occur when it moves in the local environment. If a collision will occur, the system generates a local optimized path based on the "local real-time mesh map" and updates the memory data.

[0099] S7. Evaluate the effectiveness of path planning based on the efficiency of task completion after planning, and adaptively adjust the grid code level according to different indicators such as planning effectiveness or energy efficiency.

[0100] If the danger situation of the target does not improve, that is, the danger level of the target does not decrease, it is considered a poor plan. In this case, the accuracy, i.e. the grid code level, is increased in order to seek better results in the next planning.

[0101] If the current task completion efficiency exceeds the average of the historical task completion efficiency, it is considered well planned and the grid code level is reduced. If the current task completion efficiency is lower than the average of the historical task completion efficiency, it is considered poorly planned and the accuracy, i.e., the grid code level, is increased.

[0102] If the planned indicators do not meet the requirements, the planning effect is considered poor, and the accuracy, i.e., the grid code level, is reduced.

[0103] This invention also provides a ship path planning system based on BeiDou grid codes, including: a priori path generation module, a BeiDou grid code management module, a real-time data acquisition module, a local real-time grid map construction module, a path planning template, a path collision risk analysis module, and a user information interaction module.

[0104] (1) Prior path generation module: extract historical route planning results, ship historical operation data, AIS historical trajectory data, etc., perform digital and standardized data preprocessing, establish an instance library through trajectory clustering, navigation experience, etc., and obtain prior paths for different types of ships.

[0105] (2) The Beidou grid code management module manages the basic data of Beidou grid location code, and encodes the real environment such as ships, dangerous areas, restricted areas, and islands based on Beidou grid code to generate "basic Beidou grid base map" or Beidou grid map data, which is provided to various modules for query and call.

[0106] The aforementioned BeiDou grid map = spatial BeiDou grid code + BeiDou segmentation time code + object (static obstacles, ships, and other dynamic obstacles, etc.) + attribute + associated information.

[0107] The obstacles are divided into three categories: points, lines, and areas. Point obstacles mainly include various isolated reefs, shipwrecks, navigational aids, shallow spots, etc.; line obstacles mainly include safety isobaths, land boundaries, and channel boundaries; and area obstacles mainly include danger zones, no-navigation zones, and islands.

[0108] (3) Real-time data acquisition module, which collects data from shipboard equipment in real time. Shipboard equipment includes navigation and positioning system, AIS, gyrocompass, and speedometer. The data collected by the real-time data acquisition module also includes wave and weather forecast data. The collected data is used for the path planning module.

[0109] (4) The local real-time grid map construction module includes performing Beidou grid location code positioning and matching based on real-time collected data and real-world environment grid map data. It optimizes, adjusts and refines the local environment through grid code hierarchy, identifies the entity types in the grid, and generates a "local real-time grid map" which is then passed to the path planning module.

[0110] (5) The path planning template includes a global path planning module and a local dynamic path planning analysis module:

[0111] The global path planning model uses BeiDou grid map data from the BeiDou grid code management module, combined with prior paths, priority levels, and planning indicators, to optimize and adjust the environment through a grid code hierarchy. It generates global paths based on improved A* and other algorithms, taking into account planar obstacles that have a significant impact on the global planning effect, such as danger zones, no-navigation zones, or islands.

[0112] The local dynamic path planning and analysis module optimizes and adjusts the environment based on the local real-time grid map data from the local real-time grid map construction module through a grid code hierarchy, and generates local paths based on improved A* and other algorithms to achieve path optimization in the local environment.

[0113] (6) The path collision risk analysis module includes combining the entity types in the environment, calculating the closest distance between the current node and the obstacle in the path, and analyzing in real time whether the ship maintains a safe distance from the surrounding obstacles and whether a collision will occur when it moves.

[0114] (7) User information interaction module: obtain planning data of the area to be planned, including starting point, ending point, ship size, and ship set navigation depth; convert the starting point and ending point into Beidou grid code data to determine the ship navigation area location information; determine large-scale weather forecast data based on the starting point and ending point.

[0115] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.

Claims

1. A ship path planning method based on BeiDou grid codes, characterized in that, Includes the following steps: S1. Obtain planning data for the area to be planned by the ship: The planning data includes the starting point, the ending point, the ship size, and the ship's set navigation depth; Based on the starting point and the ending point, convert the data into Beidou grid code data to determine the ship's navigation area location information; Based on the starting point and the ending point, determine large-scale weather forecast data; S2. Priority levels are assigned based on different types of vessels. When multiple vessels are present, the vessel with the highest priority level is given priority in route planning. S3. Combining prior paths, priority levels, and planning indicators, global path planning is performed by generating a grid map based on the BeiDou grid code management module in the real environment. The prior path is an instance library established based on historical route planning results and actual navigation routes, which depends on the number and accuracy of historical instances; the global path planning method adopts an improved A* algorithm based on Beidou grid code grid map, which is based on priority level and planning index; by combining the prior path and global path planning, an initial global optimal path is obtained. S4, navigation and positioning system, AIS, gyrocompass, log, wave and weather forecast real-time sensing data are transmitted to the local dynamic path planning and analysis module; S5. The local dynamic path planning and analysis module determines the grid code level and quickly generates a local real-time grid map from the grid code management module to perform collision risk analysis. Based on the real-time collected data and combined with the real-world environment "based on Beidou grid code grid map" data, Beidou grid location code positioning and matching are performed. The local environment is optimized, adjusted and refined through the grid code level, and the entity types in the grid are identified to generate a "local real-time grid map" for collision risk analysis. S6. Establish memory data, classify the danger level according to the memory data, and perform local path planning if there is a collision risk; combine the entity types in the local environment of the "local real-time grid map" to analyze in real time whether the ship maintains a safe distance from the surrounding grid and whether a collision will occur when it moves in the local environment. If a collision will occur, combine the "local real-time grid map" to generate a local optimized path and update the memory data. S7. Evaluate the effectiveness of path planning based on the efficiency of task completion after planning, and adaptively adjust the grid code level according to the planning effect or the effect of different energy efficiency indicators.

2. The ship path planning method based on Beidou grid code according to claim 1, characterized in that: The grid map based on BeiDou grid code mentioned in step S3 = spatial BeiDou grid code + BeiDou segmentation time code + object + attribute + associated information.

3. The ship path planning method based on Beidou grid code according to claim 1, characterized in that: The improved A* algorithm based on the BeiDou grid code grid map described in step S3 has an evaluation function f(n) = g(n) + P × h(n), where P = min (abs(x)). curr -x souce ),abs(abs(y curr -y source ))+min(abs(x curr -x goal ),abs(abs(y curr -y goal )); g(n) represents the actual cost from the initial node to the current node, P represents the changeover interference value, and points traveling in one direction have higher priority for the same cost as those closer to the starting or ending point. h(n) represents the estimated cost of the grid code of the current ship position and the grid code of the target point. (x curr ,y curr (x) represents the current BeiDou grid code. next ,y next ) represents the BeiDou grid code of the next node of the current node, (x goal ,y goal () is the BeiDou grid code of the target point; The cost includes Manhattan distance; g(n)=abs(x curr -x next )+abs(y curr -y next ) λ represents the threshold constant; ω1 and ω2 are weights, based on the original heuristic factor h(n). When they are higher than the threshold λ, the weights increase, and the algorithm searches faster; when they are lower than the threshold λ, the weights decrease, and the optimal path is considered first. The threshold is dynamically selected based on computer performance and map size, and the weights are also adjusted according to the size and complexity of the map. M is the number of static obstacles from the current node n to the target node, and N is the number of dynamic obstacles. D(n) collision risk value: Where θ is the scaling factor, d is the shortest distance between the current node n and the obstacle in the path, and D is the range of impact of the obstacle collision risk; The conversion factor is a normalized conversion coefficient for converting security into security cost.

4. The ship path planning method based on Beidou grid code according to claim 1, characterized in that: The memory data mentioned in step S6 consists of the characteristics, location, movement status, and danger level of non-target objects that were detected and avoided during the path planning at sea. The non-target objects are sea objects that affect the path of the mission or the safety of the mission.

5. The ship path planning method based on Beidou grid code according to claim 1, characterized in that: Step S6, which involves classifying the risk level based on the memory data, includes the following steps: If a target terrain or object matching the memory data is detected, a higher processing priority is set. Classify objects in the memory data that have previously posed a threat to one's safety into different levels of danger. Classify the danger level of targets that were previously avoided in the path planning in the memory data: If the target that was dodged was stationary or slow-moving, the memory path is used to determine whether there is a conflict between the memory path and the current best global path. If there is a conflict, the dodged target is classified as the first danger level. If the target that was dodged is moving, then the dodged target is classified as the second danger level.

6. A system for implementing the method of any one of claims 1-5, namely, a ship path planning system based on BeiDou grid codes, characterized in that, include: The prior path generation module extracts historical route planning results, historical ship operation data, and historical AIS trajectory data, performs digital and standardized data preprocessing, and establishes an instance library through trajectory clustering and navigation experience to obtain prior paths for different types of ships. The BeiDou grid code management module manages the basic data of BeiDou grid location codes, encodes the real environment (ships, dangerous areas, restricted areas, islands) based on BeiDou grid codes, generates BeiDou grid map data, and provides it for query and access by various modules. The real-time data acquisition module collects data from shipboard equipment, including navigation and positioning systems, AIS, gyrocompass, and logs. The data collected by the real-time data acquisition module also includes wave and weather forecast data. The collected data is used by the path planning module. The local real-time grid map construction module includes matching BeiDou grid location codes based on real-time collected data and grid map data, optimizing, adjusting and refining the local environment through grid code hierarchy, identifying entity types in the grid, and generating a "local real-time grid map". The path planning template includes a global path planning module and a local dynamic path planning analysis module. The global path planning module, based on BeiDou grid map data and combined with prior paths, priority levels, and planning indicators, optimizes and adjusts the environment through a grid code hierarchy. It generates a global path based on an improved A* algorithm, considering planar obstacles that significantly impact the global planning effect, specifically danger zones, no-navigation zones, or islands. The local dynamic path planning analysis module, based on local real-time grid map data from the local real-time grid map construction module, optimizes and adjusts the environment through a grid code hierarchy. It generates a local path based on an improved A* algorithm, achieving path optimization in local environments. The path collision risk analysis module includes combining the entity types in the environment and calculating the closest distance between the current node and obstacles in the path to analyze in real time whether the ship maintains a safe distance from surrounding obstacles and whether a collision will occur when it moves. The user information interaction module obtains planning data for the area to be planned, including the starting point, ending point, ship size, and set navigation depth. Based on the starting point and ending point, the data is converted into Beidou grid code data to determine the ship's navigation area location information. Based on the starting point and ending point, large-scale weather forecast data is determined.

Images