A method and system for shortest path planning

Abstract

The present application provides a shortest route planning method and system, based on the AIS data of the target ship, and pre-processing the AIS data, then using a specific judgment method to screen the AIS data, then based on the pre-processed and screened data, using the grid method to construct a low-precision environment model, then using the ant colony algorithm to calculate the shortest path of the route in the low-precision environment model, and determining the shortest path grid range, reducing the unit grid width in the low-precision environment model to construct a high-precision environment model, and determining the shortest path grid range in the high-precision environment model according to the shortest path grid range in the low-precision environment model, finally using the A-star search algorithm to calculate the shortest path of the route in the shortest path grid range of the high-precision environment model, and using the Bezier curve method to optimize the shortest path of the route, obtaining the optimized shortest path of the route and displaying, which can greatly improve the precision of the environment model, and the time cost is low.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic route planning, and particularly relates to a shortest route planning method and system. BACKGROUND

[0002] With the rapid development of digitization and intelligence, realizing autonomous navigation of a ship has become an important development direction of the shipping industry, and automatically planning a route from an anchorage to another anchorage before sailing is a basic technology of autonomous navigation. The quality of the autonomously planned route has an important influence on the safety, economy and practicability of ship navigation. In the past research, many scholars have adopted different algorithms to automatically plan a route, such as dynamic programming algorithm, artificial potential field algorithm and clustering algorithm. However, the past research has some shortcomings. When searching for a globally shortest route, a high-precision environmental model is needed, and a too high environmental model will result in a too long time cost. When calculating a globally shortest route, a too high-precision environmental model will result in a local optimum in the search, and the obtained route may not be the shortest path. SUMMARY

[0003] To solve the problems of the route not being the shortest path and the time cost being too long in the process of automatically planning a route, the present application provides a shortest route planning method, constructs a low-precision environmental model based on AIS data and using a grid method, then constructs a high-precision environmental model, calculates a shortest path of a route by using an ant colony algorithm and an A-star search algorithm, optimizes the path by using a Bezier curve method, and outputs a route that conforms to actual navigation, which can greatly improve the precision of the environmental model and has a low time cost. The present application also relates to a shortest route planning system.

[0004] The technical scheme of the present application is as follows:

[0005] A shortest route planning method, characterized by comprising the following steps:

[0006] A data acquisition and processing step: acquiring AIS data of a target ship and pre-processing the AIS data;

[0007] A data screening step: comparing the draft and the speed in the pre-processed AIS data with preset threshold values respectively, and screening out AIS data with a draft greater than a preset draft threshold value and a speed greater than a preset speed threshold value;

[0008] A low-precision environmental model establishment step: constructing a low-precision environmental model based on the pre-processed and screened AIS data by using a grid method, calculating a shortest path of a route in the low-precision environmental model by using an ant colony algorithm, and determining a shortest path grid range;

[0009] The high-precision environment model establishing step: reducing the unit grid width in the low-precision environment model to below a grid threshold to construct a high-precision environment model, and determining the shortest path grid range in the high-precision environment model according to the shortest path grid range in the low-precision environment model;

[0010] The path calculation and optimization step: calculating the shortest path of the flight route in the shortest path grid range in the high-precision environment model by using the A-star search algorithm, and optimizing the shortest path of the flight route by using the Bezier curve method to obtain the optimized shortest path of the flight route and display the optimized shortest path of the flight route.

[0011] Preferably, in the path calculation and optimization step, the A-star search algorithm is used to calculate the shortest path of the flight route in the shortest path grid range in the high-precision environment model, and the calculation specifically includes:

[0012] S1: inserting the starting node of the grid into the open list of the A-star search algorithm;

[0013] S2: if the ending node of the grid exists in the open list, the path search is completed, and if the ending node of the grid does not exist in the open list, the path does not exist;

[0014] S3: selecting the node with the minimum estimated value in the open list as an active node, and adding the active node to the closed list of the A-star search algorithm;

[0015] S4: obtaining all directly accessible access nodes, then updating the open list, if the access node is in the closed list, it is not processed, if the access node is not in the open list, the access node is added to the open list, if the access node is in the open list, it is checked whether the estimated value of the access node is the minimum, if there is the minimum estimated value, the estimated value of the access node is updated;

[0016] S5: repeating steps S2 to S4 until the process is completed, and outputting each node of the shortest path.

[0017] Preferably, in the path calculation and optimization step, the Bezier curve method is used to optimize the shortest path of the flight route, and the optimization specifically includes:

[0018] Eliminating the redundant path nodes in the shortest path, extracting key points, connecting the key points into line segments that avoid the obstacle grid, and then performing smoothing processing on the inflection points between any two adjacent line segments by using the Bezier curve method to obtain the smoothed shortest path of the flight route.

[0019] Preferably, in the data screening step, the preset speed threshold is a certain multiple of the designed speed.

[0020] Preferably, the AIS data includes longitude, latitude, heading angle, draft, and designed speed, and the preprocessing includes eliminating AIS abnormal data.

[0021] The shortest route planning system is characterized in that it comprises data acquisition and processing module, data screening module, low-precision environment model establishment module, high-precision environment model establishment module and path calculation and optimization module connected in sequence,

[0022] The data acquisition and processing module acquires AIS data of the target ship and pre-processes the AIS data.

[0023] The data screening module compares the draft and the speed in the pre-processed AIS data with preset threshold values respectively, and screens out AIS data with draft greater than the preset draft threshold value and speed greater than the preset speed threshold value.

[0024] The low-precision environment model establishment module constructs a low-precision environment model by using a grid method based on the pre-processed and screened AIS data, calculates the shortest path of the route in the low-precision environment model by using an ant colony algorithm, and determines the shortest path grid range.

[0025] The high-precision environment model establishment module reduces the unit grid width in the low-precision environment model to below the grid threshold value to construct a high-precision environment model, and determines the shortest path grid range in the high-precision environment model according to the shortest path grid range in the low-precision environment model.

[0026] The path calculation and optimization module calculates the shortest path of the route in the shortest path grid range in the high-precision environment model by using an A-star search algorithm, optimizes the shortest path of the route by using a Bezier curve method, obtains the optimized shortest path of the route, and displays the optimized shortest path of the route.

[0027] Preferably, in the path calculation and optimization module, the calculation of the shortest path of the route in the shortest path grid range in the high-precision environment model by using the A-star search algorithm specifically comprises:

[0028] S1: inserting the starting node of the grid into the open list of the A-star search algorithm;

[0029] S2: if the ending node of the grid exists in the open list, completing the path search, and if the ending node of the grid does not exist in the open list, the path does not exist;

[0030] S3: selecting the node with the minimum estimated value in the open list as an active node, and adding the active node to the closed list of the A-star search algorithm;

[0031] S4: obtaining all accessible access nodes, then updating the open list, if the access node is in the closed list, not processing; if the access node is not in the open list, adding the access node to the open list; if the access node is in the open list, checking whether the estimated value of the access node is the minimum, if there is the minimum estimated value, updating the estimated value of the access node;

[0032] S5: repeating steps S2 to S4 until the process ends, outputting each node of the shortest path.

[0033] Preferably, in the path calculation and optimization module, the Bezier curve method is used to optimize the shortest path of the route, including:

[0034] The redundant path nodes in the shortest path are eliminated, the key points are extracted, the key points are connected into line segments that avoid the obstacle grid, and the Bezier curve method is used to smooth the inflection points between any two adjacent line segments to obtain the smooth shortest path of the route.

[0035] Preferably, the preset speed threshold is a certain multiple of the designed speed.

[0036] Preferably, the AIS data includes longitude, latitude, heading angle, draft and designed speed; and the preprocessing includes eliminating AIS abnormal data.

[0037] The present application has the following advantages:

[0038] The shortest route planning method provided by the present application is based on AIS data of a target ship, adopts a specific screening method to screen the AIS data, and adopts a grid method to construct a low-precision environment model to reduce the calculation time; then the ant colony algorithm is used to preliminarily determine the approximate range of the shortest route in the low-precision environment model to further reduce the length of the calculation time; then a high-precision environment model is constructed, and the search range is further narrowed in the high-precision environment model according to the approximate range, which can greatly improve the precision of the environment model and reduce the calculation time cost. Then the A-star search algorithm (also known as A* search algorithm or A* algorithm) is used to calculate the shortest path in the shortest path range of the high-precision environment model, and the route planning algorithm combining the ant colony algorithm and the A-star search algorithm can ensure global optimization. Finally, the Bezier curve method is used to optimize the key points of the shortest path to obtain the optimized shortest path and show the route that meets the actual navigation, and the optimization processing of the planned path can avoid the local optimal condition and ensure that the optimized route meets the actual route.

[0039] The present application also relates to a shortest route planning system, which corresponds to the shortest route planning method described above, and can be understood as a system for implementing the shortest route planning method described above, comprising sequentially connected data acquisition and processing module, data screening module, low-precision environment model establishing module, high-precision environment model establishing module and path calculation and optimization module, each module works in coordination with each other, the low-precision environment model is constructed based on AIS data, and then the high-precision environment model is constructed, the shortest path is calculated by using ant colony algorithm and A-star search algorithm, the path is optimized by using Bezier curve method, and the route meeting the actual navigation is output, so that the precision of the environment model can be greatly improved, and the time cost of calculation is low. BRIEF DESCRIPTION OF DRAWINGS

[0040] Figure 1 is a flow chart of the shortest route planning method of the present application.

[0041] Figure 2 is a preferred flow chart of the shortest route planning method of the present application.

[0042] Figure 3 is a principle diagram of the adjacent matrix of the present application.

[0043] Figure 4 is a schematic diagram of the shortest path grid range of the present application. DETAILED DESCRIPTION

[0044] The present application will be described below in combination with the drawings.

[0045] The present application relates to a shortest route planning method, a flow chart of the method is shown in Figure 1 , and comprises the following steps in sequence:

[0046] I. Data acquisition and processing step: acquiring AIS data of a target ship, and pre-processing the AIS data; specifically, as shown in the preferred flow chart Figure 2 , first, acquiring AIS data of a target ship, such as longitude, latitude, heading angle, draft and designed speed, inputting the start point and destination point, the draft and designed speed of the target ship, and eliminating abnormal data in the AIS data, the principles for determining the abnormal data are as follows:

[0047] Longitude <-180° or >180°

[0048] Latitude <-90° or >90°

[0049] Heading angle <0 or 360°

[0050] Draft <0 m

[0051] Speed <0 kn or 50 kn

[0052] II. Data screening step: comparing the draft and speed of the pre-processed AIS data with the preset threshold value, if the draft is greater than the preset draft threshold value and the speed is greater than the preset speed threshold value, then the AIS data of the draft and speed that meet the requirements is retained.

[0053] Specifically, the AIS data is screened, and the preset speed threshold value is preferably a certain multiple of the designed speed, such as 1 / 3 of the designed speed, that is, the AIS data is screened according to the draft and 1 / 3 of the designed speed of the target ship, for example, the draft is 10 m, and the designed speed is 14.5 kn, then the AIS data with a draft greater than 10 m and a speed greater than 4.84 kn is retained.

[0054] III. Low-precision environmental model establishment step: based on the pre-processed and screened AIS data, a low-precision environmental model is constructed by using the grid method, and the shortest path in the low-precision environmental model is calculated by using the ant colony algorithm, and the shortest path grid range is determined.

[0055] Specifically, the route of a ship in a certain specific navigation area can be simplified as the path between two positions of the ship on a plane. The grid method divides the provided navigation area map into a plurality of grids, and assigns a specified unit length to the side length of each grid. Then, according to the navigability of a single grid in the navigation area map, the grid with a free attribute is marked with white color and the obstacle grid is marked with black color. Using the grid, the route planning problem in a complex spatial navigation environment is converted into searching for a path between two grid center points in a two-dimensional grid that meets certain constraint conditions.

[0056] Assuming that the length of the working environment map is L and the width is N, the height and width of each grid are I, the environment can be divided into an information map of n*m cells, as shown in the following formula:

[0057] I = L / n = N / m (1)

[0058] In route planning, the AIS point data positioning free attribute grid is contained in the grid; if there is no AIS data, the grid is an obstacle grid. After evaluating the feasibility of the grid in the environmental space, the obstacle and free attribute grids are respectively retained in the infeasible region set and the feasible region set.

[0059] Chart represents the entire environment, Gridi,j represents all the information of a grid cell, i represents the horizontal direction, i.e. the east-west direction, j represents the vertical direction, i.e. the south-north direction, all grids are placed in the set P, n and m are the maximum values of i and j in the environmental space, and the expression is as shown in the following formula:

[0060] P = {(i,j) | 1≤i≤n, 1≤j≤m} (2)

[0061] Chart = { gridi,j | gridi,j = 0 or 1, (i, j) e P} (3)

[0062] Then, the grid method is used to build the environment model, and the grid width is set to 0.5°. Then, the ant colony algorithm is used to calculate the shortest path of the route in the low-precision environment model, that is, to search for the shortest path in the low-precision environment model and determine the approximate range of the shortest path. Specifically,

[0063] First, the adjacency matrix is constructed: the adjacency matrix conforming to the environment model is used to record the distance between adjacent navigable grids, as shown in equation (1). Second, as shown in equation (2), grid i-1,j+1 and grid i,j-1 are obstacles, so the ants in grid i,j can go to the other six free attribute grids. The distance d between two free attribute grids is calculated according to the following equation: Figure 3

[0064] d = 60 * arccos (sin (lat i,j ) * sin (lat i,j+1 ) + cos (lat i,j) * cos (lat i,j+1 ) * cos (lon i,j+1 - lon i,j )) (4)

[0065] The distance between the free attribute grid and the obstacle grid is 0. Then, the ant colony algorithm is used to calculate the shortest path of the route in the low-precision environment model.

[0066] The ant colony algorithm (ACA algorithm) is used to determine the shortest path in the low-precision environment model, and the main program is as follows:

[0067] I. Convert the specified starting point A and ending point B into their respective grid positions in the grid map, that is, the starting grid A and the ending grid B.

[0068] II. Initialize and set the parameters of each ant population.

[0069] III. Use the roulette method to select the next adjacent free attribute grid. The adjacent free attribute grid and its distance are determined according to the conclusion of the free attribute grid division model.

[0070] IV. Each ant moves along the pheromone trail until it reaches the ending point B.

[0071] V. Record the passable path and total distance of each grid.

[0072] VI. Update the pheromone. ​

[0073] VII. Determine whether the population has reached the maximum number of iterations.

[0074] VIII. Return the shortest path.

[0075] Finally, the grid range of the shortest path is determined; the free attribute grid range is further narrowed according to the calculated shortest path. The five grid nodes through which the shortest path passes, as shown in FIG. 6 (left), each have adjacent free attribute grids within the range determined by the shortest path; finally, the range shown in FIG. 6 (right) is obtained, with the white grid being the grid range selected by the shortest path, and the grids outside this range becoming obstacle grids. Figure 4 Figure 4 (Left) and (right) show the range determined by the shortest path.

[0076] Four, high-precision environment model establishment step: reduce the unit grid width in the low-precision environment model to below the grid threshold to construct a high-precision environment model, and determine the shortest path grid range in the high-precision environment model according to the shortest path grid range in the low-precision environment model.

[0077] Specifically, the width of the unit grid in the low-precision environment model is reduced to, for example, 0.0625°, so as to generate a more precise environment model according to the determined navigable range described above. On the basis of the high-precision environment model, the search range of the shortest path is further narrowed based on the approximate range of the shortest path determined above, and the navigable grid under the high-precision environment model is determined according to the grid range determined above; if the free attribute grid under the high-precision environment model is not within the grid range determined above, then the grid is an obstacle grid.

[0078] Five, path calculation and optimization step: the A-star search algorithm is used to calculate the shortest path of the route within the shortest path grid range in the high-precision environment model, and the Bezier curve method is used to optimize the shortest path of the route, to obtain the optimized shortest path of the route and display it.

[0079] Specifically, the A-star search algorithm calculates the minimum route according to the following formula:

[0080] F(n) = G(n) + H(n) (5)

[0081] G(n) = d(n, n-1) + G(n-1) (n > 1) (6)

[0082] ​G(n) is the real cost of moving between the start node of the current grid, H(n) represents the predicted moving cost between the current grid and the end grid, d(n, n-1) is the distance between the free attribute grids, corresponding to d in formula (4), G(n-1) is the straight-line distance from the previous grid to the start node, and n represents the number of steps of movement. d in formula (4) corresponds to H(n) in formula (5), that is, the moving distance represents the moving cost, and formula (5) is the heuristic function of the A-star search algorithm.

[0083] The A-star search algorithm uses an open list to save nodes waiting for checking and uses a closed list to store nodes that have been explored before. The A-star search algorithm follows the following steps:

[0084] S1: Insert the start node S of the grid into the open list of the A-star search algorithm;

[0085] S2: If the end node E of the grid exists in the open list, complete the path search, and if the end node E of the grid does not exist in the open list, the path does not exist;

[0086] S3: Select the node u with the minimum estimated value F in the open list as the active node and add the active node to the closed list of the A-star search algorithm;

[0087] S4: Obtain all directly accessible access nodes v, then update the open list, if v is in the closed list, do not process; if v is not in the open list, add v to the open list, and its related F = G(u) + distance(u, v) + H(v); if v is in the open list, check whether the estimated value F of v is the minimum, if there is the minimum estimated value F, update the estimated value F of v;

[0088] S5: Repeat steps S2 to S4 until the process ends, and output each node of the shortest path.

[0089] Then, the path calculated by the A-star search algorithm includes many redundant turning points, which is quite different from the actual route. Therefore, the Bezier curve method is used to optimize the shortest path of the route, and the main steps are as follows:

[0090] First, eliminate redundant points and extract key points. Assume that the current path node and the previous and subsequent nodes are on the same straight line. In this case, delete the current node, update the path, and find that the subsequent nodes are all on the same line, and the last remaining node is the key point. In addition, extract the key points and ensure that the reconnected line segment does not pass through the obstacle grid. For example, after the first step, collect the key points (M1, M2,..., Mk), and connect M1 and M3 to generate a line segment l (line segment function: Ax+By+C=0). Determine whether the connected M1 and M3 pass through the obstacle grid N ((x, y), (-x, y), (x, -y), (-x, -y) are the four vertices of the obstacle grid N) using formulas (7) and (8). If it does not pass through the obstacle grid N, the P2 point can be deleted, the path can be updated, and the subsequence Mk and M1 can be continuously added until the trajectory point {Pi|i=1,2,3,...,n} passing through the obstacle grid is obtained.

[0091] (A(-x)+By+C)(Ax+B(-y)+C)>0 (7)

[0092] (A(-x)+B(-y)+C)(Ax+Bx+C)>0 (8)

[0093] Second, use the third-order Bezier curve method to smooth the processed route, as shown in formula (9). The coordinates of the control points pi to pi+3 are given, the first point Pi={xi,yi}, the second point Pi+1={xi+1,yi+1}, and so on. The parameter t is from 0 to 1, with a step size of 0.05, and the loop walks through 0, 0.05, 0.1, 0.15,..., 0.95, 1.

[0094] P=Pi*(1-t)3+3Pi+1*t(1-t)2+3Pi+2*t2(1-t)+Pi+3*t3 (9)

[0095] Embodiment:

[0096] For example, a 210,000-ton bulk carrier sets sail from Zhoushan (123.981 °E, 30.875 °N) to Tianjin Port (118.485 °E, 38.826 °N). The draft is 10 m, and the designed speed is 14.5 kn.

[0097] 1) Remove abnormal AIS data and filter AIS data;

[0098] The removal method is as described above, and the AIS data with a draft greater than 10 m and a speed greater than 4.84 kn is retained, and the AIS data with a longitude between (117, 124) and a latitude between (30, 41) is retained.

[0099] 2) construct a low-precision environment model;

[0100] The grid width is positioned at 0.5°, so that a grid map of 14*22 grids is obtained, and the starting point and the end point corresponding grids are determined.

[0101] 3) calculate the shortest path and the grid range of the shortest path of the low-precision environment model by using an ant colony algorithm;

[0102] 4) construct a high-precision environment model;

[0103] The grid width is positioned at 0.0625, so that a high-precision grid map of 128*176 grids is obtained.

[0104] 5) determine the shortest path range under the high-precision environment model;

[0105] According to the grid range obtained in step 3, the area of the range under the high-precision environment model is determined, and the starting point and the end point corresponding grids are determined.

[0106] 6) calculate the shortest path under the area of step 5 by using an A-star search algorithm;

[0107] 7) after obtaining the shortest path, eliminate redundant points, extract key points, and then smooth the route by using a Bezier curve;

[0108] 8) output the final route.

[0109] The application also relates to a shortest route planning system, which corresponds to the shortest route planning method described above and can be understood as a system for implementing the method, and the system comprises data acquisition and processing modules, a data screening module, a low-precision environment model establishment module, a high-precision environment model establishment module and a path calculation and optimization module connected in sequence, and specifically,

[0110] The data acquisition and processing module acquires AIS data of a target ship and pre-processes the AIS data;

[0111] The data screening module compares the draft and the speed in the pre-processed AIS data with preset threshold values respectively, and screens out AIS data with a draft greater than a preset draft threshold value and a speed greater than a preset speed threshold value;

[0112] The low-precision environment model establishment module constructs a low-precision environment model by using a grid method based on the pre-processed and screened AIS data, calculates the shortest route of the low-precision environment model by using an ant colony algorithm, and determines the grid range of the shortest route;

[0113] The high-precision environment model establishing module reduces the unit grid width in the low-precision environment model to below a grid threshold to construct a high-precision environment model, and determines the shortest path grid range in the high-precision environment model according to the shortest path grid range in the low-precision environment model;

[0114] The path calculation and optimization module calculates the shortest path of the flight route in the shortest path grid range in the high-precision environment model by using the A-star search algorithm, and optimizes the shortest path of the flight route by using the Bezier curve method to obtain an optimized shortest path of the flight route and display the optimized shortest path of the flight route.

[0115] Preferably, in the path calculation and optimization module, the calculation of the shortest path of the flight route in the shortest path grid range in the high-precision environment model by using the A-star search algorithm specifically includes:

[0116] S1: inserting a starting node S of the grid into an open list of the A-star search algorithm;

[0117] S2: if an ending node E of the grid exists in the open list, completing the path search, and if the ending node E of the grid does not exist in the open list, the path does not exist;

[0118] S3: selecting a node u with the minimum estimated value F in the open list as an active node, and adding the active node to a closed list of the A-star search algorithm;

[0119] S4: obtaining all directly accessible access nodes v, then updating the open list, if v is in the closed list, not processing; if v is not in the open list, adding v to the open list; if v is in the open list, checking whether the estimated value F of v is the minimum, if there is the minimum estimated value F, updating the estimated value F of v;

[0120] S5: repeating steps S2 to S4 until the process ends, and outputting each node of the shortest path.

[0121] Preferably, in the path calculation and optimization module, the optimization of the shortest path of the flight route by using the Bezier curve method includes:

[0122] Eliminating redundant path nodes in the shortest path, extracting key points, connecting the key points into line segments that do not pass through (avoid) obstacle grids, and then performing smoothing processing on the inflection points between any two adjacent line segments by using the Bezier curve method to obtain a smoothed shortest path of the flight route.

[0123] Preferably, the preset speed threshold is a certain multiple of the designed speed.

[0124] Preferably, the AIS data includes longitude, latitude, heading angle, draft, and designed speed, and the preprocessing includes eliminating AIS abnormal data.

[0125] The application provides an objective and scientific shortest route planning method and system, constructs a low-precision environment model based on AIS data and by using a grid method, further constructs a high-precision environment model, calculates a shortest path of a route by using an ant colony algorithm and an A-star search algorithm, optimizes the path by using a Bezier curve method, and outputs a route conforming to actual navigation, which can greatly improve the precision of the environment model and has low time cost.

[0126] It should be noted that the above specific embodiments can enable those skilled in the art to have a more comprehensive understanding of the present application, but in no way limit the present application. Therefore, although the present application has been described in detail with reference to the drawings and examples, those skilled in the art should understand that the present application can still be modified or replaced by equivalents, in short, all technical solutions and improvements which do not deviate from the spirit and scope of the present application should be covered in the protection scope of the patent of the present application.

Claims

1. A shortest route planning method, characterized in that, Includes the following steps: Data acquisition and processing steps: Acquire the AIS data of the target vessel and preprocess the AIS data; Data filtering steps: Compare the draft and speed in the preprocessed AIS data with preset thresholds respectively, and filter out AIS data with a draft greater than the preset draft threshold and a speed greater than the preset speed threshold. The steps for establishing a low-precision environment model are as follows: Based on the preprocessed and screened AIS data, a low-precision environment model is constructed using the raster method. Then, the ant colony algorithm is used to calculate the shortest path of the flight route in the low-precision environment model and determine the raster range of the shortest path. Steps for establishing a high-precision environment model: Reduce the unit grid width in the low-precision environment model to below the grid threshold to construct a high-precision environment model, and determine the shortest path grid range in the high-precision environment model based on the shortest path grid range in the low-precision environment model. Path calculation and optimization steps: The shortest path of the route within the shortest path grid range in the high-precision environment model is calculated using the A* search algorithm, and the shortest path of the route is optimized using the Bézier curve method to obtain the optimized shortest path of the route and display it.

2. The shortest route planning method according to claim 1, characterized in that, In the path calculation and optimization steps, the A* search algorithm is used to calculate the shortest path of the airway within the shortest path grid range in the high-precision environmental model. Specifically, this includes: S1: Insert the starting node of the raster into the open list of the A* search algorithm; S2: If the end node of the grid exists in the open list, the path search is completed; if the end node of the grid does not exist in the open list, the path does not exist. S3: Select the node with the smallest estimated value from the open list as the active node, and add the active node to the closed list of the A* search algorithm; S4: Obtain all directly accessible nodes, then update the open list. If an access node is in the closed list, do not process it; if an access node is not in the open list, add it to the open list; if an access node is in the open list, check if the estimated value of the access node is the minimum. If the minimum estimated value is found, update the estimated value of the access node. S5: Repeat steps S2 to S4 until the process ends, and output each node of the shortest path.

3. The shortest route planning method according to claim 1, characterized in that, The path calculation and optimization steps include optimizing the shortest path of the air route using the Bézier curve method, which includes: Eliminate redundant path nodes in the shortest path, extract key points, connect the key points into line segments that avoid the obstacle grid, and then use the Bézier curve method to smooth the inflection points between any two adjacent line segments to obtain a smooth shortest route.

4. The shortest route planning method according to claim 1, characterized in that, In the data filtering step, the preset speed threshold is a certain multiple of the design speed.

5. The shortest route planning method according to claim 1, characterized in that, The AIS data includes longitude, latitude, heading, draft, and design speed; the preprocessing includes removing anomalous AIS data.

6. A shortest route planning system, characterized in that, It includes, in sequence, a data acquisition and processing module, a data filtering module, a low-precision environment model building module, a high-precision environment model building module, and a path calculation and optimization module. The data acquisition and processing module acquires the AIS data of the target vessel and preprocesses the AIS data. The data filtering module compares the draft and speed in the preprocessed AIS data with preset thresholds, and filters out AIS data with a draft greater than the preset draft threshold and a speed greater than the preset speed threshold. The low-precision environment model building module uses a raster method to construct a low-precision environment model based on preprocessed and screened AIS data. Then, it uses an ant colony algorithm to calculate the shortest path of the flight route in the low-precision environment model and determines the raster range of the shortest path. The high-precision environment model building module reduces the unit grid width in the low-precision environment model to below the grid threshold to build a high-precision environment model, and determines the shortest path grid range in the high-precision environment model based on the shortest path grid range in the low-precision environment model. The path calculation and optimization module uses the A* search algorithm to calculate the shortest path of the flight route within the shortest path grid range in the high-precision environment model, and then uses the Bézier curve method to optimize the shortest path of the flight route, and displays the optimized shortest path of the flight route.

7. The shortest route planning system according to claim 6, characterized in that, In the path calculation and optimization module, the shortest path of the airway within the shortest path grid range in the high-precision environmental model is calculated using the A* search algorithm, specifically including: S1: Insert the starting node of the raster into the open list of the A* search algorithm; S2: If the end node of the grid exists in the open list, the path search is completed; if the end node of the grid does not exist in the open list, the path does not exist. S3: Select the node with the smallest estimated value from the open list as the active node, and add the active node to the closed list of the A* search algorithm; S4: Obtain all directly accessible nodes, then update the open list. If an access node is in the closed list, do not process it; if an access node is not in the open list, add it to the open list; if an access node is in the open list, check if the estimated value of the access node is the minimum. If the minimum estimated value is found, update the estimated value of the access node. S5: Repeat steps S2 to S4 until the process ends, and output each node of the shortest path.

8. The shortest route planning system according to claim 6, characterized in that, The path calculation and optimization module uses the Bézier curve method to optimize the shortest path of the air route, including: Eliminate redundant path nodes in the shortest path, extract key points, connect the key points into line segments that avoid the obstacle grid, and then use the Bézier curve method to smooth the inflection points between any two adjacent line segments to obtain a smooth shortest route.

9. The shortest route planning system according to claim 6, characterized in that, The preset speed threshold is a certain multiple of the design speed.

10. The shortest route planning system according to claim 6, characterized in that, The AIS data includes longitude, latitude, heading, draft, and design speed; the preprocessing includes removing anomalous AIS data.

Images