Land-sea relay mixed path planning method, device and system

Abstract

The invention relates to the technical field of marine path planning, and particularly discloses a land-sea relay mixed path planning method, device and system, and the method comprises the steps: obtaining an impact factor inputted by a user; preprocessing the road network data, the route data and the ocean topographic data according to the influence factors; constructing a path topological relation according to the road section area data and constructing an ocean weight matrix according to the ocean grid area data; constructing a pre-path planning weight model according to the road attribute information, the weather forecast information, the path topological relation and the ocean weight matrix, and performing optimization processing on the pre-path planning weight model according to the first target cost function; constructing a real-time path planning weight model according to the pre-path planning result and the real-time road condition influence factor, and performing optimization processing on the real-time path planning weight model according to a second target cost function; and outputting a real-time path planning result. The land-sea relay mixed path planning method provided by the invention is high in safety and low in transportation cost.

Description

Technical Field

[0001] The present invention relates to the technical field of marine route planning, and in particular to a land-sea relay hybrid route planning method, a land-sea relay hybrid route planning device and a land-sea relay hybrid route planning system. Background Art

[0002] Traditional path planning algorithms are mainly based on road path calculations. Path planning needs to be performed on the basis of establishing a global road network. Under conditions such as special vehicles and meteorological environments, it is not possible to effectively meet user needs. In particular, it is often difficult to adjust the path in a dynamic environment, resulting in untimely updates of the driving route. The traditional sea route is manually predicted with low efficiency and accuracy. On the one hand, the safety of ocean transportation is greatly affected by meteorological factors, and the consideration of meteorological conditions is to protect ocean transportation. On the other hand, unlike land roads, ocean routes have no road network information or only a small amount of route information, which makes it difficult to fully guarantee the cost and safety of ocean transportation. Existing path planning methods cannot effectively consider the impact of the predicted meteorological environment, the mixed calculation of land and sea transportation, and the predicted transportation time, so it will bring unpredictable risks and timeliness problems. Finally, with the rapid development of technologies such as big data and AI, in the context of land-sea economic integration, land-sea relay transportation has gradually been valued by people, among which path planning plays an important role in land-sea relay transportation. Therefore, the future research and development direction of path planning should be committed to improving the accuracy, timeliness, universality, and globality of path planning, optimizing the adaptability of algorithm models, and providing users with various path requirements that can be met to the greatest extent by simply combining evaluation models.

[0003] Therefore, how to maximize traffic safety and reduce transportation time and cost has become a technical problem that needs to be urgently solved by technical personnel in this field. Summary of the invention

[0004] The present invention provides a land-sea relay hybrid path planning method, a land-sea relay hybrid path planning device and a land-sea relay hybrid path planning system, which solve the problems in the related art that the traditional path planning algorithm cannot meet the user needs and has low safety and high cost.

[0005] As a first aspect of the present invention, a land-sea relay hybrid path planning method is provided, which includes:

[0006] Obtaining the influencing factors input by the user, wherein the influencing factors include at least: a must-go path point, an avoidance area, and an area not considered in terms of direction;

[0007] Preprocessing the road network data, route data and ocean terrain data according to the influencing factors to obtain the road section area data and ocean grid area data associated with the influencing factors;

[0008] Constructing a path topology relationship based on the road section area data associated with the influencing factor and constructing an ocean weight matrix based on the ocean grid area data associated with the influencing factor;

[0009] Constructing a pre-path planning weight model according to road attribute information, weather forecast information, path topology relationship and ocean weight matrix, and optimizing the pre-path planning weight model according to a first target cost function to obtain a pre-path planning result, wherein the first target cost function is obtained by optimizing the cost function of the Dijkstra algorithm and the cost function of the A* algorithm;

[0010] A real-time path planning weight model is constructed according to the pre-path planning result and the real-time traffic condition influencing factor, and the real-time path planning weight model is optimized according to a second target cost function to obtain a real-time path planning result, wherein the real-time traffic condition influencing factor includes at least real-time traffic load information and real-time weather information, and the second target cost function is obtained by optimizing the cost function of the Dijkstra algorithm and the cost function of the A* algorithm based on motion constraints;

[0011] Output the real-time path planning result.

[0012] Further, the road network data, route data and ocean terrain data are preprocessed according to the influencing factors to obtain the road section area data and ocean grid area data associated with the influencing factors, including:

[0013] Classifying the road network data, route data and ocean terrain data into path planning types according to the influencing factors, wherein the path planning types include three types: road, ocean and road-ocean mixed paths;

[0014] Sorting the necessary path points according to the distance priority principle, and calculating the geographical range of all the sorted necessary path points;

[0015] Calculate the shortest distance from the necessary path point to the road area vector line within the geographical range of the necessary path point to determine a new path point located on the road area vector line;

[0016] Obtaining road section area data associated with the influencing factor according to the new path point;

[0017] The ocean grid area data associated with the influencing factor is determined according to the path planning type.

[0018] Further, calculating the shortest distance from the necessary path point to the road area vector line within the geographical range of the necessary path point to determine a new path point located on the road area vector line includes:

[0019] Determine the projection point of the current must-pass path point on the road area vector line within the geographical range of the must-pass path point;

[0020] Calculate the Euclidean distance from the current path point to the projection point;

[0021] Traverse all vector lines in the road area and select the shortest distance point as the new path point corresponding to the current must-go path point.

[0022] Further, constructing a path topology relationship according to the road section area data associated with the influencing factor and constructing an ocean weight matrix according to the ocean grid area data associated with the influencing factor, including:

[0023] Constructing a path topological relationship of road node lines regarding connectivity and directionality according to the road section area data associated with the influencing factors;

[0024] An ocean terrain weight matrix of terrain impact on ship navigability is constructed according to the ocean grid areas associated with the influencing factors.

[0025] Further, a pre-path planning weight model is constructed according to the road attribute information, the weather forecast information, the path topology relationship and the ocean weight matrix, and the pre-path planning weight model is optimized according to the first objective cost function to obtain a pre-path planning result, including:

[0026] Constructing a directed graph of vector data according to the road attribute information, the weather forecast information and the path topological relationship, and performing a tracking weight calculation on the directed graph of vector data according to the Dijkstra algorithm in the first objective cost function to obtain the shortest path in advance and the land route consumption cost of the land in each forecast time period;

[0027] A grid weight model is constructed based on weather forecast information and an ocean weight matrix, and the grid weight model is optimized according to the A* algorithm in the first objective cost function to obtain the ocean's pre-shortest path and ocean route consumption cost in each forecast time period.

[0028] Further, a real-time path planning weight model is constructed according to the pre-path planning result and the real-time traffic condition influencing factor, and the real-time path planning weight model is optimized according to the second objective cost function to obtain the real-time path planning result, wherein the real-time traffic condition influencing factor includes at least real-time traffic load information and real-time meteorological information, including:

[0029] Selecting the departure time corresponding to the minimum value of the total route consumption in each forecast time period in the pre-route planning result as the departure time for real-time path planning, wherein the total route consumption in each forecast time period includes the sum of the land route consumption cost and the sea route consumption cost in each forecast time period;

[0030] Calculate the shortest path from the land segment starting point to the land-sea transfer point in real time according to the real-time weather information, the real-time traffic load information and the Dijkstra algorithm in the second objective cost function;

[0031] Calculate the optimal path from the land-sea transfer point in the ocean section to the destination point in real time according to the real-time meteorological information and the A* algorithm based on motion constraints in the second objective cost function;

[0032] Obtaining a real-time path planning result based on the shortest path from the starting point of the land segment to the land-sea transfer point and the optimal path from the land-sea transfer point of the ocean segment to the destination point;

[0033] The real-time path planning result is updated according to the real-time weather information and the real-time traffic load information.

[0034] Furthermore, the first objective cost function is obtained by optimizing the cost function of the Dijkstra algorithm and the cost function of the A* algorithm, including:

[0035] Construct a pre-land path factor analysis matrix based on path topological relationships, road attribute information and weather forecast information;

[0036] Determine the first objective cost function influencing factor of the Dijkstra algorithm according to the pre-land path factor analysis matrix;

[0037] Constructing the first objective cost function of the Dijkstra algorithm according to the first objective cost function influencing factor of the Dijkstra algorithm;

[0038] Construct a pre-ocean grid factor analysis matrix based on the ocean weight matrix and weather forecast information;

[0039] Determine the first objective cost function influencing factor of the A* algorithm according to the pre-ocean grid factor analysis matrix;

[0040] The first objective cost function of the A* algorithm is constructed according to the first objective cost function influence factor of the A* algorithm.

[0041] Furthermore, the second objective cost function is obtained by optimizing the cost function of the Dijkstra algorithm and the cost function of the A* algorithm based on motion constraints, including:

[0042] Constructing a real-time land path factor analysis matrix according to the pre-path planning results, real-time road load information and real-time meteorological information;

[0043] Determine the second objective cost function influencing factor of the Dijkstra algorithm according to the real-time land path factor analysis matrix;

[0044] Constructing a second objective cost function of the Dijkstra algorithm according to the second objective cost function influencing factor of the Dijkstra algorithm;

[0045] Constructing a real-time ocean grid factor analysis matrix according to the pre-path planning results and real-time meteorological information;

[0046] Determining the ocean grid cost function influencing factor of the motion-constrained A* algorithm according to the real-time ocean grid factor analysis matrix;

[0047] Constructing an ocean grid cost function of the A* algorithm based on motion constraints according to the ocean grid cost function influencing factor of the A* algorithm;

[0048] Constructing a ship horizontal motion constraint cost function according to the ship motion model, wherein the ship horizontal motion constraint includes constraining the ship's forward direction;

[0049] A second objective cost function of the motion-constrained A* algorithm is obtained according to the ocean grid cost function and the ship horizontal motion constraint cost function.

[0050] As another aspect of the present invention, a land-sea relay hybrid path planning device is provided, which is used to implement the land-sea relay hybrid path planning method described above, and includes:

[0051] An acquisition module is used to acquire an influencing factor input by a user, wherein the influencing factor includes at least: a must-pass path point, an avoidance area, and an area not considered in terms of direction;

[0052] A preprocessing module, used for preprocessing the road network data, route data and ocean terrain data according to the influencing factors, and obtaining the road section area data and ocean grid area data associated with the influencing factors;

[0053] A topological relationship building module, used to build a path topological relationship according to the road section area data associated with the influencing factor and to build an ocean weight matrix according to the ocean grid area data associated with the influencing factor;

[0054] A pre-path planning module, used to construct a pre-path planning weight model according to road attribute information, weather forecast information, path topology relationship and ocean weight matrix, and optimize the pre-path planning weight model according to a first target cost function to obtain a pre-path planning result, wherein the first target cost function is obtained by optimizing the cost function of the Dijkstra algorithm and the cost function of the A* algorithm;

[0055] A real-time path planning module, configured to construct a real-time path planning weight model according to the pre-path planning result and the real-time traffic condition influencing factor, and optimize the real-time path planning weight model according to a second target cost function to obtain a real-time path planning result, wherein the real-time traffic condition influencing factor includes at least real-time traffic load information and real-time weather information, and the second target cost function is obtained by optimizing the cost function of the Dijkstra algorithm and the cost function of the A* algorithm based on motion constraints;

[0056] The output module is used to output the real-time path planning result.

[0057] As another aspect of the present invention, a land-sea relay hybrid path planning system is provided, which includes: a human-computer interaction device and the land-sea relay hybrid path planning device described above, wherein the human-computer interaction device is communicatively connected with the land-sea relay hybrid path planning device;

[0058] The human-computer interaction device is used to receive the influencing factors input by the user, and to display the pre-path planning results and the real-time path planning results to the user;

[0059] The land-sea relay hybrid path planning device is used to process data information input by a user, and perform pre-path planning and real-time path planning in combination with a target cost function.

[0060] The land-sea relay hybrid path planning method provided by the present invention obtains the influence factors input by the user, and pre-processes the road network data, route data and ocean terrain data according to the influence factors, constructs the path topology relationship and the ocean weight matrix based on the pre-processed data, constructs the pre-path planning weight model according to the path topology relationship and the ocean weight matrix, and obtains the pre-path planning result after optimization processing based on the first target cost function, and constructs the real-time path planning weight model based on the pre-path planning result in combination with the real-time road condition influence factor, and obtains the real-time path planning result after optimization processing based on the second target cost function. This land-sea relay hybrid path planning method, according to different user needs, considers factors such as the location of the transport vehicle, real-time road conditions, road direction, transport conditions and meteorological environment, adopts two planning strategies of pre-path planning and real-time path planning to determine the target route and transport travel time, so that the safety of vehicle operation is maximized, and the combination of the two, whether it is land, sea or land-sea transshipment transportation, the land-sea relay path planning method combined with meteorological factors can effectively reduce transportation time and cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the present invention, but do not constitute a limitation of the present invention.

[0062] Figure 1 This is a flow chart of the land-sea relay hybrid path planning method provided by the present invention.

[0063] Figure 2 This is a specific architecture diagram of the land-sea relay hybrid path planning method combined with meteorology provided by the present invention.

[0064] Figure 3 A flow chart of a method for preprocessing road network data, route data and ocean terrain data according to influencing factors provided by the present invention.

[0065] Figure 4 A flow chart of a method for determining a new path point located on a road area vector line provided by the present invention.

[0066] Figure 5 A flow chart of the method for constructing path topological relationships and constructing an ocean weight matrix provided by the present invention.

[0067] Figure 6 A flow chart of the method for obtaining pre-path planning results provided by the present invention.

[0068] Figure 7 This is a cost factor analysis flow chart provided by the present invention.

[0069] Figure 8 A flow chart of the method for obtaining real-time path planning results provided by the present invention.

[0070] Fig. 9 A schematic diagram of the specific construction process of the ship horizontal motion constraint cost function provided by the present invention.

[0071] Fig.10 The directions of the ship's path lines before and after optimization provided by the present invention are more consistent with the ship's movement state at sea.

[0072] Fig.11 A schematic diagram of a travel time with recommended better road conditions for the advance route planning provided by the present invention.

[0073] Fig.12 The real-time path planning provided by the present invention is a path diagram that takes into account the real-time weather cost involved in the algorithm calculation.

[0074] Fig.13 This is a structural block diagram of the land-sea relay hybrid path planning device provided by the present invention.

[0075] Fig.14 This is a structural block diagram of the land-sea relay hybrid path planning system provided by the present invention. DETAILED DESCRIPTION

[0076] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0077] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment 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 making creative work should fall within the scope of protection of the present invention.

[0078] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so as to describe the embodiments of the present invention described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0079] In this embodiment, a land-sea relay hybrid path planning method is provided. Figure 1 is a flow chart of a land-sea relay hybrid path planning method provided according to an embodiment of the present invention. Figure 1 As shown, including:

[0080] S100, obtaining an influencing factor input by a user, wherein the influencing factor includes at least: a must-pass path point, an avoidance area, and an area not considered in terms of direction;

[0081] In an embodiment of the present invention, the user can input the necessary path points, avoidance areas, and areas not considering the direction in the path planning through the human-computer interaction device. It should be understood that the necessary path points can specifically include the starting point, the end point, and the passing point. The avoidance areas are specifically some prohibited areas. The areas not considering the direction are specifically some areas where the driving direction does not need to be considered, such as areas where fire trucks need to go through in reverse under special circumstances.

[0082] S200, pre-processing the road network data, route data and ocean terrain data according to the influencing factors to obtain the road section area data and ocean grid area data associated with the influencing factors;

[0083] In the embodiment of the present invention, data screening is performed on the road network data, the route data and the ocean terrain data to obtain the road section area data and the ocean grid area data associated with the influencing factors.

[0084] Specifically, it can be understood that the road section area data and ocean grid area data associated with the influencing factors are screened out from the huge road network data, route data and ocean terrain data through data processing.

[0085] S300, constructing a path topology relationship according to the road section area data associated with the influencing factor and constructing an ocean weight matrix according to the ocean grid area data associated with the influencing factor;

[0086] In an embodiment of the present invention, a path topology relationship (specifically, it can be understood as directed graph construction) is constructed based on the above-screened road section area data associated with the influencing factors to facilitate the determination of path nodes and driving routes; and an ocean weight matrix is ​​constructed based on the ocean grid area data associated with the influencing factors to facilitate the planning of ocean routes.

[0087] S400, constructing a pre-path planning weight model according to road attribute information, weather forecast information, path topology relationship and ocean weight matrix, and optimizing the pre-path planning weight model according to a first target cost function to obtain a pre-path planning result, wherein the first target cost function is obtained by optimizing the cost function of the Dijkstra algorithm and the cost function of the A* algorithm;

[0088] In an embodiment of the present invention, specific advance path planning can be performed based on road attribute information (such as road grade, road width, and weight, etc.), weather forecast information (such as weather conditions within 7 days), path topology relationships, and ocean weight matrices. That is, in an embodiment of the present invention, it can be specifically understood as planning future travel routes based on currently known information and forecasted information to obtain advance path planning results.

[0089] It should be understood that in the pre-path planning, the pre-path planning weight model needs to be optimized in combination with the first target cost function, so as to obtain the pre-shortest route with the goal of minimizing the overall route consumption cost.

[0090] S500, constructing a real-time path planning weight model according to the pre-path planning result and the real-time traffic condition influencing factor, and optimizing the real-time path planning weight model according to a second target cost function to obtain a real-time path planning result, wherein the real-time traffic condition influencing factor includes at least real-time traffic load information and real-time weather information, and the second target cost function is obtained by optimizing the cost function of the Dijkstra algorithm and the cost function of the A* algorithm based on motion constraints;

[0091] In an embodiment of the present invention, the departure time is determined based on the above-mentioned pre-path planning results, and a real-time path planning weight model is constructed based on the real-time road condition influencing factors at the current departure time, that is, adjustments are made on the pre-planned route based on the real-time road condition influencing factors to obtain the real-time path planning results. The real-time path planning results specifically need to optimize the real-time path planning weight model based on the second objective cost function, specifically to obtain the shortest land path and the optimal ocean path with the goal of still minimizing the overall consumption cost.

[0092] S600: Output the real-time path planning result.

[0093] In summary, the land-sea relay hybrid path planning method provided by the present invention obtains the influencing factors input by the user, and pre-processes the road network data, route data and ocean terrain data according to the influencing factors, constructs the path topology relationship and the ocean weight matrix based on the pre-processed data, constructs the pre-path planning weight model based on the path topology relationship and the ocean weight matrix, and obtains the pre-path planning result after optimization processing based on the first objective cost function, and constructs the real-time path planning weight model based on the pre-path planning result in combination with the real-time road condition influencing factors, and obtains the real-time path planning result after optimization processing based on the second objective cost function. This land-sea relay hybrid path planning method, according to different user needs, considers factors such as the location of the transport vehicle, real-time road conditions, road direction, transportation conditions and meteorological environment, adopts two planning strategies of pre-path planning and real-time path planning to determine the target route and transportation travel time, so that the safety of vehicle operation is maximized, and the combination of the two, whether it is land, sea or land-sea transshipment transportation, the land-sea relay path planning method combined with meteorological factors can effectively reduce transportation time and cost.

[0094] In the embodiment of the present invention, the specific architecture of the land-sea relay hybrid path planning method combined with meteorology is as follows: Figure 2 As shown, it mainly includes data preprocessing, conditional input factor calculation, algorithm cost function generation and hybrid path algorithm output. The main difference between the land road network algorithm is that the meteorological environment and attribute fields (turning radius, load-bearing, etc.) participate in the cost weight calculation. The ocean grid algorithm mainly generates a weight matrix from basic ocean data, meteorological data, etc., and uses the ship's maritime motion model to improve the search trajectory of the A* algorithm to meet the maritime navigation planning route. In order to ensure the complexity and timeliness of different algorithm calculations, the technology of the present invention uses parallel processing of large-scale road network data, meteorological data, ocean grid data and internal cost influencing factor calculations.

[0095] Specifically, the road network data, route data and ocean terrain data are preprocessed according to the influencing factors to obtain the road section area data and ocean grid area data associated with the influencing factors, such as Figure 3As shown, including:

[0096] S210, classifying the road network data, route data and ocean terrain data into path planning types according to the influencing factors, wherein the path planning types include three types: road, ocean and road-ocean mixed paths;

[0097] In the embodiment of the present invention, the path planning type is divided into road, ocean or mixed path mode according to the input necessary path point sequence including the starting point and the ending point.

[0098] S220, sorting the necessary path points according to the distance priority principle, and calculating the geographical range of all sorted necessary path points;

[0099] Specifically, the necessary path points are sorted according to distance priority, and the geographical ranges of all the sorted necessary path points are calculated.

[0100] S230, calculating the shortest distance from the necessary path point to the road area vector line within the geographical range of the necessary path point to determine a new path point located on the road area vector line;

[0101] It should be understood that for the necessary path points, since they may not be located on the road area vector line, that is, the necessary path points input by the user may be located on a non-actual road, when performing path planning, it is necessary to select points on the road area vector line to determine them as new path points, and the principle of selecting new path points is the shortest distance to the current necessary path points.

[0102] More specifically, the shortest distance from the necessary path point to the road area vector line is calculated within the geographical range of the necessary path point to determine a new path point located on the road area vector line, such as Figure 4 As shown, including:

[0103] S231, determining the projection point of the current must-pass path point on the road area vector line within the geographical range of the must-pass path point;

[0104] S232, calculating the Euclidean distance from the current necessary path point to the projection point;

[0105] S233, traverse all vector lines in the road area, and select the shortest distance point as the new path point corresponding to the current must-pass path point.

[0106] Specifically, the new path point is calculated, that is, the shortest distance point from the current must-go path point P to the road area vector line. The calculation steps are as follows:

[0107] a. Define vectors separately and

[0108] b. Calculate the projection point O of P on line segment AB, vector projection formula Where t is the ratio from P to AB;

[0109] c. Determine the projection position: If t<0, the nearest point is A; if t>1, the nearest point is B; otherwise, point O is on AB, then the nearest point is

[0110] d. According to the Euclidean distance from point P to point O;

[0111] e. Traverse all vector lines in the area and take the point with the minimum distance as the new input path point.

[0112] It should be noted that there is a mixed type in which the shortest distance between the road and ocean route points and the port transfer point is determined as the nearest land-sea relay point.

[0113] S240, obtaining road section area data associated with the influencing factor according to the new path point;

[0114] In the embodiment of the present invention, data may be screened according to the new path point to obtain the road section area data associated with the influencing factor.

[0115] S250. Determine ocean grid area data associated with the influencing factor according to the path planning type.

[0116] It should be understood that the ocean grid area data associated with the influencing factors are screened and determined according to the path planning type.

[0117] In an embodiment of the present invention, a path topology relationship is constructed based on the road segment area data associated with the influencing factor, and an ocean weight matrix is ​​constructed based on the ocean grid area data associated with the influencing factor, such as Figure 5 As shown, including:

[0118] S310, constructing a path topological relationship of road node lines regarding connectivity and directionality according to the road section area data associated with the influencing factors;

[0119] S320: constructing an ocean terrain weight matrix of terrain impact on ship trafficability according to the ocean grid area associated with the influencing factor.

[0120] It should be understood that according to the geographical scope of the above-mentioned necessary path points and the user's pre-input conditions (avoidance zone, no-entry zone, etc.), the effective input road network vector or ocean raster data is screened to construct the road node line topological relationship (connectivity, directionality, etc.) and the ocean terrain weight matrix (terrain affects ship traffic capacity).

[0121] In an embodiment of the present invention, a pre-path planning weight model is constructed according to road attribute information, weather forecast information, path topology relationship and ocean weight matrix, and the pre-path planning weight model is optimized according to the first objective cost function to obtain a pre-path planning result, such as Figure 6 As shown, including:

[0122] S410, constructing a vector data directed graph according to the road attribute information, the weather forecast information and the path topology relationship, and performing a tracking weight calculation on the vector data directed graph according to the Dijkstra algorithm in the first objective cost function to obtain the pre-shortest path and the land route consumption cost of the land in each forecast time period;

[0123] S420. Construct a grid weight model based on weather forecast information and an ocean weight matrix, and optimize the grid weight model based on the A* algorithm in the first objective cost function to obtain the ocean's pre-shortest path and ocean route consumption cost in each forecast time period.

[0124] It should be understood that the vector data directed graph is established by the path topology, road attribute information and weather forecast information (wind speed, rainfall, visibility, temperature, etc.), and the grid weight is established according to the weather forecast information (wind speed, rainfall, visibility, temperature, etc.) and the ocean weight matrix. The matrix data model (meteorological element weight calculation, c is the meteorological element value, w is its influence rate) is used to calculate the shortest path and the overall consumption cost of the route for land and ocean in each forecast time period through the Dijkstra algorithm and the optimized A* algorithm with customized improved cost function.

[0125] In an embodiment of the present invention, the first objective cost function is obtained by optimizing the cost function of the Dijkstra algorithm and the cost function of the A* algorithm, and includes:

[0126] Construct a pre-land path factor analysis matrix based on path topological relationships, road attribute information and weather forecast information;

[0127] Determine the first objective cost function influencing factor of the Dijkstra algorithm according to the pre-land path factor analysis matrix;

[0128] Constructing the first objective cost function of the Dijkstra algorithm according to the first objective cost function influencing factor of the Dijkstra algorithm;

[0129] Construct a pre-ocean grid factor analysis matrix based on the ocean weight matrix and weather forecast information;

[0130] Determine the first objective cost function influencing factor of the A* algorithm according to the pre-ocean grid factor analysis matrix;

[0131] The first objective cost function of the A* algorithm is constructed according to the first objective cost function influence factor of the A* algorithm.

[0132] Specifically, Figure 7 The figure below is a flowchart of cost factor analysis. Taking the first objective cost function of the Dijkstra algorithm as an example, the main calculation steps for improving and optimizing the cost function influencing factors are as follows:

[0133] (1) Calculate the covariance matrix ∑ of the original data X;

[0134] (2) Calculate the characteristic root of the covariance matrix ∑, expressed as λ1≥λ2≥…≥λ p , the corresponding unit eigenvectors are denoted as e1, e2, …, e p , the eigenvector matrix is ​​represented by U. At this time, the covariance matrix ∑ has the following expression formula (1):

[0135]

[0136] Based on formula (1) and model assumptions, the covariance matrix ∑ can be expressed as follows:

[0137] ∑=Cov(X)=Cov(AF+ε)=ACov(F)A T + Cov(ε)=AA T +Dσ,(2)

[0138] in:

[0139] Combining formulas (1) and (2), we can get the estimate of the factor loading matrix:

[0140]

[0141] in, where λ i represents the i-th eigenvalue, e ij Represents λ i The corresponding j-th component of the i-th eigenvector, after obtaining the loading matrix, the factor model can be expressed as:

[0142]

[0143] Its main characteristics can be expressed by influencing factors, which are road grade, road width, road direction, road length and meteorological environmental factors (wind speed, rainfall), etc. Its correlation matrix is:

[0144]

[0145] Generate the corresponding influencing factor characteristic parameter values ​​and substitute them into the algorithm cost function to measure the influence of each factor on path planning. You can also only take the parameters with greater influence on the characteristic value parameters to participate in the calculation to improve the efficiency of the Dijkstra algorithm. The cost function of the Dijkstra algorithm is as follows:

[0146] cost=λ1Width+λ2Lehgth+…+λ n Other.

[0147] It should be understood that the process of determining the first objective cost function influencing factor of the A* algorithm based on the pre-ocean grid factor analysis matrix can refer to the process of the above-mentioned Dijkstra algorithm, except that the cost function factor is different. The cost function factor for the A* algorithm can be relevant parameter information of the ocean weight matrix and weather forecast information, etc.

[0148] In an embodiment of the present invention, a real-time path planning weight model is constructed according to the pre-path planning result and the real-time traffic condition influencing factor, and the real-time path planning weight model is optimized according to the second objective cost function to obtain the real-time path planning result, wherein the real-time traffic condition influencing factor includes at least real-time traffic load information and real-time meteorological information, such as Figure 8 As shown, including:

[0149] S510, selecting the departure time corresponding to the minimum value of the total route consumption in each forecast time period in the pre-route planning result as the departure time for real-time path planning, wherein the total route consumption in each forecast time period includes the sum of the land route consumption cost and the sea route consumption cost in each forecast time period;

[0150] S520, calculating the shortest path from the land segment starting point to the land-sea transfer point in real time according to the real-time meteorological information, the real-time road load information and the Dijkstra algorithm in the second objective cost function;

[0151] S530, calculating in real time the optimal path from the land-sea transfer point in the ocean section to the destination point according to the real-time meteorological information and the A* algorithm based on motion constraints in the second objective cost function;

[0152] S540, obtaining a real-time path planning result according to the shortest path from the starting point of the land segment to the land-sea transfer point and the optimal path from the land-sea transfer point of the ocean segment to the destination point;

[0153] S550: Update the real-time path planning result according to the real-time weather information and the real-time traffic load information.

[0154] It should be understood that by selecting the minimum overall consumption in each forecast time period under the pre-path as the departure time, real-time meteorological data (wind speed, rainfall, visibility, temperature, etc.), real-time road condition data (congestion status), vehicle positioning information and other user-defined input factors are connected, and the Dijkstra algorithm with an improved optimization cost function is used on land to calculate the shortest path from the land section to the land-sea transfer point in real time. After arriving at the land-sea relay point, the A* algorithm with optimized ship motion constraints is used at sea to calculate the optimal path for the ocean section in real time, and the real-time path planning route is updated according to the update frequency of real-time meteorological data and real-time road condition data.

[0155] In an embodiment of the present invention, the second objective cost function is obtained by optimizing the cost function of the Dijkstra algorithm and the cost function of the A* algorithm based on motion constraints, and includes:

[0156] Constructing a real-time land path factor analysis matrix according to the pre-path planning results, real-time road load information and real-time meteorological information;

[0157] Determine the second objective cost function influencing factor of the Dijkstra algorithm according to the real-time land path factor analysis matrix;

[0158] Constructing a second objective cost function of the Dijkstra algorithm according to the second objective cost function influencing factor of the Dijkstra algorithm;

[0159] Constructing a real-time ocean grid factor analysis matrix according to the pre-path planning results and real-time meteorological information;

[0160] Determining the ocean grid cost function influencing factor of the motion-constrained A* algorithm according to the real-time ocean grid factor analysis matrix;

[0161] Constructing an ocean grid cost function of the A* algorithm based on motion constraints according to the ocean grid cost function influencing factor of the A* algorithm;

[0162] Constructing a ship horizontal motion constraint cost function according to the ship motion model, wherein the ship horizontal motion constraint includes constraining the ship's forward direction;

[0163] A second objective cost function of the motion-constrained A* algorithm is obtained according to the ocean grid cost function and the ship horizontal motion constraint cost function.

[0164] It should be noted that when performing real-time path planning, the construction idea of ​​the second objective cost function of the Dijkstra algorithm can refer to the construction idea of ​​the first objective cost function of the Dijkstra algorithm mentioned above, except that the cost function factor of the second objective cost function of the Dijkstra algorithm also includes real-time traffic load information and real-time weather information and other characteristics based on the pre-planned path.

[0165] The second objective cost function of the A* algorithm based on motion constraints in the embodiment of the present invention is specifically the sum of the ocean grid cost function and the ship horizontal motion constraint cost function, wherein the ocean grid cost function can specifically refer to the first objective cost function of the A* algorithm mentioned above, the only difference is that the cost function factor of the ocean grid cost function also includes real-time meteorological information.

[0166] The specific construction process of the ship's horizontal motion constraint cost function is combined with Fig. 9 As shown, including the following:

[0167] The evaluation function of the A* algorithm based on motion constraints is f(n)=g(n)+h(n), where g(n) represents the ocean grid cost function and h(n) represents the ship's horizontal motion constraint cost function. In the embodiment of the present invention, the prediction cost function is more complex, and a cost heuristic function is designed to reflect the actual motion of the ship and the terrain environment characteristics. Different motion conditions and adjacent matrix weights are used as inputs, and the next node search must meet the ship's marine motion constraints. Considering the connectivity of the motion space and the direction of the ship's motion, the two-dimensional discrete grid space and direction are continuously interpolated to the ship's motion trajectory to ensure the smoothness of the route. The processing flow is as follows: Fig. 9 shown.

[0168] Among them, the ship motion model implementation needs to conform to the ship's turning trajectory at sea, relying on the turning speed and its own weight, and calculating the h(n) prediction cost function of the ship's horizontal motion constraint at the current point. The method is as follows:

[0169] In the embodiment of the present invention, the horizontal motion direction of the vessel dv x 、dv y Respectively represent the horizontal component of the ship; inertia turning radius Among them, V s represents the ship speed, g represents the acceleration due to gravity, Indicates the rudder angle.

[0170] (1) Calculate the ship motion model to realize the turning angle with speed and weight as input The responsive ship motion model is the output.

[0171] (2) Calculate the change in the ship angle by knowing the grid accuracy and turning radius dy are the accuracy values ​​in the x and y directions respectively, and R represents the turning radius of the ship.

[0172] (3) The grid accuracy is used as the step size, and the initial angle With turning angle Determine that the next grid to be tracked is within the 8-neighborhood node range, such as Select three grids in the direction of the ship's advance to participate in the calculation of weights and add them to the weight function h(n) calculation.

[0173] (4) Consider the heuristic function for calculating the ship motion model as:

[0174]

[0175] Among them, n goal represents the end point of the path, n represents the current node, Indicates the grid angle involved in step (3).

[0176] (5) Considering the characteristics of ship movement at sea, taking into account the shortest path and the law of ship movement at sea, the h(n) function in step (4) is substituted into the A* algorithm to calculate the optimal path, as follows: Fig.10 The direction of the ship's path lines before and after optimization is more consistent with the ship's movement status at sea.

[0177] Based on the above method, the real-time path planning result is finally obtained, and the shortest or optimal path planned is output, including the traffic status, time consumption, weather environment, etc. of each path, and the route buffer point of interest information (gas stations, restaurants, etc.) is calculated.

[0178] The land-sea relay hybrid route planning method of the embodiment of the present invention is used to implement the algorithms for the pre-route planning and real-time route planning schemes respectively. The pre-route planning is to first have a road and then assign weather and road condition information at different times, and recommend a travel time with better road conditions. Fig.11 As shown in the figure, real-time path planning takes into account the real-time weather cost in the algorithm calculation. Fig.12 The path shown.

[0179] In summary, the land-sea relay hybrid path planning method provided by the present invention mainly considers the impact of weather on the path and other influencing factors, and is applicable to different application scenarios of land, sea, and land-sea relay. It is more accurate, efficient, and maximizes transportation safety compared to conventional path planning algorithm results. This method provides a solid guarantee for improving the value of marine economy, reducing transportation costs, and opening up land-sea transportation and marine aquaculture.

[0180] For example, in the marine industry system, the foundation is land-based seedling cultivation, and the core is deep-sea aquaculture, the focus of which is effective connection and accurate feeding. Therefore, having the ability to transport feed and transfer adult fish at sea is the key to deep-sea aquaculture. Under the digital and intelligent requirements of the marine industry system, the land-sea relay hybrid path planning method provided by the present invention can provide path predictions for feed transportation from land to marine farms and for transfer of adult fish from marine farms to land. Based on the analysis of basic data such as meteorological environment, marine topography, and land transportation, the traffic status of feed transportation routes in various time periods in the future or real-time dynamic route planning is predicted, providing reliable land and sea combined transportation routes and predicting the optimal transportation time.

[0181] Therefore, the land-sea relay hybrid path planning method provided by the present invention has limited input conditions for conventional path planning and is relatively friendly to general transportation. However, it is not practical enough for planning requirements such as meteorological elements, user-defined attribute elements, and land-sea relay global paths, especially for special vehicle requirements such as hazardous chemicals transportation and fire emergency, which reduces the matching degree of user route requirements and even brings unpredictable security threats. It is proposed to use user-defined factors affecting route planning as the input of the principal component factor analysis method, generate parameters affecting path weights, and obtain the shortest and optimal path algorithm cost function.

[0182] In addition, the current road condition timeliness and navigation location information update of big data, obtaining its data to participate in path planning can effectively integrate external data to participate in the algorithm weight calculation, which can not only improve the location accuracy and analysis timeliness of the algorithm of the present invention, but also make full use of the convenience of big data.

[0183] In the embodiment of the present invention, users can choose different strategies, automatically match algorithms for route planning, and use principal component factor analysis to calculate the weight parameters affecting path costs. It has strong scalability and users can add consideration factors according to their own special path needs, such as custom input rules, influencing weights, etc., to improve user recognition of planned routes and reduce unnecessary financial and human resources consumption.

[0184] As another embodiment of the present invention, a land-sea relay hybrid path planning device 100 is provided, which is used to implement the land-sea relay hybrid path planning method described above, wherein: Fig.13 As shown, including:

[0185] An acquisition module 110 is used to acquire an influencing factor input by a user, wherein the influencing factor includes at least: a must-go path point, an avoidance area, and an area not considered in terms of direction;

[0186] A preprocessing module 120, for preprocessing the road network data, the route data and the ocean terrain data according to the influencing factors, and obtaining the road section area data and the ocean grid area data associated with the influencing factors;

[0187] A topological relationship building module 130, for building a path topological relationship according to the road section area data associated with the influencing factor and building an ocean weight matrix according to the ocean grid area data associated with the influencing factor;

[0188] A pre-path planning module 140 is used to construct a pre-path planning weight model according to road attribute information, weather forecast information, path topology relationship and ocean weight matrix, and optimize the pre-path planning weight model according to a first target cost function to obtain a pre-path planning result, wherein the first target cost function is obtained by optimizing the cost function of the Dijkstra algorithm and the cost function of the A* algorithm;

[0189] A real-time path planning module 150 is used to construct a real-time path planning weight model according to the pre-path planning result and the real-time traffic condition influencing factor, and optimize the real-time path planning weight model according to a second target cost function to obtain a real-time path planning result, wherein the real-time traffic condition influencing factor includes at least real-time traffic load information and real-time weather information, and the second target cost function is obtained by optimizing the cost function of the Dijkstra algorithm and the cost function of the A* algorithm based on motion constraints;

[0190] The output module 160 is used to output the real-time path planning result.

[0191] The land-sea relay hybrid path planning device provided by the present invention obtains the influence factors input by the user, and pre-processes the road network data, route data and ocean terrain data according to the influence factors, constructs the path topology relationship and the ocean weight matrix based on the pre-processed data, constructs the pre-path planning weight model according to the path topology relationship and the ocean weight matrix, and obtains the pre-path planning result after optimization processing based on the first target cost function, and constructs the real-time path planning weight model based on the pre-path planning result in combination with the real-time road condition influence factor, and obtains the real-time path planning result after optimization processing based on the second target cost function. This land-sea relay hybrid path planning device, according to different user needs, considers factors such as the location of the transport vehicle, real-time road conditions, road direction, transport conditions and meteorological environment, adopts two planning strategies of pre-path planning and real-time path planning to determine the target route and transport travel time, so that the safety of vehicle operation is maximized, and the combination of the two, whether it is land, sea or land-sea transshipment transportation, the land-sea relay path planning method combined with meteorological factors can effectively reduce transportation time and cost.

[0192] The specific working principle of the land-sea relay hybrid path planning device provided by the present invention can be referred to the description of the land-sea relay hybrid path planning method above, and will not be repeated here.

[0193] As another embodiment of the present invention, a land-sea relay hybrid path planning system 10 is provided, wherein: Fig.14 As shown, it includes: a human-computer interaction device 200 and the land-sea relay hybrid path planning device 100 mentioned above, and the human-computer interaction device 200 is communicatively connected with the land-sea relay hybrid path planning device 100;

[0194] The human-computer interaction device 200 is used to receive the influencing factors input by the user, and to display the pre-path planning results and the real-time path planning results to the user;

[0195] The land-sea relay hybrid path planning device 100 is used to process data information input by a user, and perform pre-path planning and real-time path planning in combination with a target cost function.

[0196] The land-sea relay hybrid path planning system provided by the embodiment of the present invention adopts the land-sea relay hybrid path planning device mentioned above, and comprehensively considers various environmental factors on land and sea, including road networks, traffic rules, and the most important meteorological conditions (such as wind speed, rainfall, temperature changes, visibility, etc.), to provide the best path relay mode for land and water transportation. The land-sea relay hybrid path planning system makes up for the fact that conventional path planning methods cannot predict the impact of the meteorological environment on path planning, breaks the bottleneck of land and sea transfer path planning, and can improve transportation efficiency, safety and cost-effectiveness. The land-sea relay hybrid path planning system can be applied to a variety of application fields such as logistics transportation, emergency rescue operations, aquatic aquaculture, etc. on land, water and across three scenarios of land and water.

[0197] According to different user needs, considering factors such as the location of the transport vehicle, real-time road conditions, road direction, transport conditions and meteorological environment, two planning strategies, pre-path planning and real-time path planning, are used to determine the target route and transport travel time, so as to maximize the safety of vehicle operation. The pre-path planning strategy aims to calculate the shortest path of a mixture of static roads and sea routes without considering the weather conditions, obtain the 24 / 48-hour forecast of the cloud weather environment (rainfall, wind speed, temperature, visibility and other factors) according to the planned path range, construct the meteorological multi-factor weight matrix of the region with the accuracy of meteorological data and map it to the planned route segment, and calculate the overall cost consumption value of the route in each future time period in combination with the path travel time consumption, and finally give the departure time and estimated consumption time with the smallest cost value on the route. The real-time path planning strategy aims to obtain vehicle location information in real time. When there are changes in meteorological conditions within a certain range from the target that may affect the efficiency and safety of driving, users can choose to re-plan the optimal route from the current location to the destination taking into account real-time meteorological factors, providing meteorological real-time situation awareness and path analysis methods, closely integrating meteorology with path planning, and both solutions support functions such as adding must-pass points, avoidance areas, and areas that do not consider directions on the route. Compared with the path planning methods of the prior art that do not consider meteorological environmental factors, cannot realize mixed calculation of land and sea transportation, and cannot predict transportation time, which brings unpredictable risks and poor timeliness, the embodiments of the present invention have good adaptability and reliability, and can be applied to special vehicle requirements such as industrial transportation, fire emergency, and land and sea transportation. Users only need to combine strategies to meet various user needs to the greatest extent.

[0198] In summary, the land-sea relay hybrid path planning system of the present invention comprehensively considers multiple application scenarios and multiple factors. On land, it mainly considers road data, including static road information, such as road network topology information, road length, road grade, etc. and dynamic road information, such as real-time road conditions information, traffic forecast information, etc. On the ocean, the route or grid data is comprehensively considered. The route data between ports is relatively weak. On this basis, weighted ocean grid data is generated through meteorological and ocean data, and the ocean planning path is calculated based on the grid to solve the pain points of ocean transportation. The organic combination of the two, whether in land, ocean or land-sea transshipment transportation, the land-sea relay path planning method combined with meteorological factors can effectively reduce transportation time and cost, improve transportation efficiency and safety, and reduce energy consumption and environmental pollution.

[0199] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present invention, but the present invention is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and essence of the present invention, and these modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. A land-sea relay hybrid path planning method, characterized in that: include: Obtaining the influencing factors input by the user, wherein the influencing factors include at least: a must-go path point, an avoidance area, and an area not considered in terms of direction; Preprocessing the road network data, route data and ocean terrain data according to the influencing factors to obtain the road section area data and ocean grid area data associated with the influencing factors; Constructing a path topology relationship based on the road section area data associated with the influencing factor and constructing an ocean weight matrix based on the ocean grid area data associated with the influencing factor; Constructing a pre-path planning weight model according to road attribute information, weather forecast information, path topology relationship and ocean weight matrix, and optimizing the pre-path planning weight model according to a first target cost function to obtain a pre-path planning result, wherein the first target cost function is obtained by optimizing the cost function of the Dijkstra algorithm and the cost function of the A* algorithm; A real-time path planning weight model is constructed according to the pre-path planning result and the real-time traffic condition influencing factor, and the real-time path planning weight model is optimized according to a second target cost function to obtain a real-time path planning result, wherein the real-time traffic condition influencing factor includes at least real-time traffic load information and real-time weather information, and the second target cost function is obtained by optimizing the cost function of the Dijkstra algorithm and the cost function of the A* algorithm based on motion constraints; Output the real-time path planning result.

2. The land-sea relay hybrid path planning method according to claim 1 is characterized in that: Preprocessing the road network data, route data and ocean terrain data according to the influencing factors to obtain the road section area data and ocean grid area data associated with the influencing factors includes: Classifying the road network data, route data and ocean terrain data into path planning types according to the influencing factors, wherein the path planning types include three types: road, ocean and road-ocean mixed paths; Sorting the necessary path points according to the distance priority principle, and calculating the geographical range of all the sorted necessary path points; Calculate the shortest distance from the necessary path point to the road area vector line within the geographical range of the necessary path point to determine a new path point located on the road area vector line; Obtaining road section area data associated with the influencing factor according to the new path point; The ocean grid area data associated with the influencing factor is determined according to the path planning type.

3. The land-sea relay hybrid path planning method according to claim 2 is characterized in that: Calculating the shortest distance from the necessary path point to the road area vector line within the geographical range of the necessary path point to determine a new path point located on the road area vector line, including: Determine the projection point of the current must-pass path point on the road area vector line within the geographical range of the must-pass path point; Calculate the Euclidean distance from the current path point to the projection point; Traverse all vector lines in the road area and select the shortest distance point as the new path point corresponding to the current must-go path point.

4. The land-sea relay hybrid path planning method according to claim 1 is characterized in that: Constructing a path topology relationship according to the road section area data associated with the influencing factor and constructing an ocean weight matrix according to the ocean grid area data associated with the influencing factor, including: Constructing a path topological relationship of road node lines regarding connectivity and directionality according to the road section area data associated with the influencing factors; An ocean terrain weight matrix of terrain impact on ship navigability is constructed according to the ocean grid areas associated with the influencing factors.

5. The land-sea relay hybrid path planning method according to claim 1, characterized in that: A pre-path planning weight model is constructed according to road attribute information, weather forecast information, path topology relationship and ocean weight matrix, and the pre-path planning weight model is optimized according to the first objective cost function to obtain a pre-path planning result, including: Constructing a directed graph of vector data according to the road attribute information, the weather forecast information and the path topological relationship, and performing a tracking weight calculation on the directed graph of vector data according to the Dijkstra algorithm in the first objective cost function to obtain the shortest path in advance and the land route consumption cost of the land in each forecast time period; A grid weight model is constructed based on weather forecast information and an ocean weight matrix, and the grid weight model is optimized according to the A* algorithm in the first objective cost function to obtain the ocean's pre-shortest path and ocean route consumption cost in each forecast time period.

6. The land-sea relay hybrid path planning method according to claim 1, characterized in that: A real-time path planning weight model is constructed according to the pre-path planning result and the real-time traffic condition influencing factor, and the real-time path planning weight model is optimized according to the second objective cost function to obtain a real-time path planning result, wherein the real-time traffic condition influencing factor includes at least real-time traffic load information and real-time meteorological information, including: Selecting the departure time corresponding to the minimum value of the total route consumption in each forecast time period in the pre-route planning result as the departure time for real-time path planning, wherein the total route consumption in each forecast time period includes the sum of the land route consumption cost and the sea route consumption cost in each forecast time period; Calculate the shortest path from the land segment starting point to the land-sea transfer point in real time according to the real-time weather information, the real-time traffic load information and the Dijkstra algorithm in the second objective cost function; Calculate the optimal path from the land-sea transfer point in the ocean section to the destination point in real time according to the real-time meteorological information and the A* algorithm based on motion constraints in the second objective cost function; Obtaining a real-time path planning result based on the shortest path from the starting point of the land segment to the land-sea transfer point and the optimal path from the land-sea transfer point of the ocean segment to the destination point; The real-time path planning result is updated according to the real-time weather information and the real-time traffic load information.

7. The land-sea relay hybrid path planning method according to claim 1, characterized in that: The first objective cost function is obtained by optimizing the cost function of the Dijkstra algorithm and the cost function of the A* algorithm, and includes: Construct a pre-land path factor analysis matrix based on path topological relationships, road attribute information and weather forecast information; Determine the first objective cost function influencing factor of the Dijkstra algorithm according to the pre-land path factor analysis matrix; Constructing the first objective cost function of the Dijkstra algorithm according to the first objective cost function influencing factor of the Dijkstra algorithm; Construct a pre-ocean grid factor analysis matrix based on the ocean weight matrix and weather forecast information; Determine the first objective cost function influencing factor of the A* algorithm according to the pre-ocean grid factor analysis matrix; The first objective cost function of the A* algorithm is constructed according to the first objective cost function influence factor of the A* algorithm.

8. The land-sea relay hybrid path planning method according to claim 1, characterized in that: The second objective cost function is obtained by optimizing the cost function of the Dijkstra algorithm and the cost function of the A* algorithm based on motion constraints, and includes: Constructing a real-time land path factor analysis matrix according to the pre-path planning results, real-time road load information and real-time meteorological information; Determine the second objective cost function influencing factor of the Dijkstra algorithm according to the real-time land path factor analysis matrix; Constructing a second objective cost function of the Dijkstra algorithm according to the second objective cost function influencing factor of the Dijkstra algorithm; Constructing a real-time ocean grid factor analysis matrix according to the pre-path planning results and real-time meteorological information; Determining the ocean grid cost function influencing factor of the motion-constrained A* algorithm according to the real-time ocean grid factor analysis matrix; Constructing an ocean grid cost function of the A* algorithm based on motion constraints according to the ocean grid cost function influencing factor of the A* algorithm; Constructing a ship horizontal motion constraint cost function according to the ship motion model, wherein the ship horizontal motion constraint includes constraining the ship's forward direction; A second objective cost function of the motion-constrained A* algorithm is obtained according to the ocean grid cost function and the ship horizontal motion constraint cost function.

9. A land-sea relay hybrid path planning device, used to implement the land-sea relay hybrid path planning method according to any one of claims 1 to 8, characterized in that: include: An acquisition module is used to acquire an influencing factor input by a user, wherein the influencing factor includes at least: a must-pass path point, an avoidance area, and an area not considered in terms of direction; A preprocessing module, used for preprocessing the road network data, route data and ocean terrain data according to the influencing factors, and obtaining the road section area data and ocean grid area data associated with the influencing factors; A topological relationship building module, used to build a path topological relationship according to the road section area data associated with the influencing factor and to build an ocean weight matrix according to the ocean grid area data associated with the influencing factor; A pre-path planning module, used to construct a pre-path planning weight model according to road attribute information, weather forecast information, path topology relationship and ocean weight matrix, and optimize the pre-path planning weight model according to a first target cost function to obtain a pre-path planning result, wherein the first target cost function is obtained by optimizing the cost function of the Dijkstra algorithm and the cost function of the A* algorithm; A real-time path planning module, configured to construct a real-time path planning weight model according to the pre-path planning result and the real-time traffic condition influencing factor, and optimize the real-time path planning weight model according to a second target cost function to obtain a real-time path planning result, wherein the real-time traffic condition influencing factor includes at least real-time traffic load information and real-time weather information, and the second target cost function is obtained by optimizing the cost function of the Dijkstra algorithm and the cost function of the A* algorithm based on motion constraints; The output module is used to output the real-time path planning result.

10. A land-sea relay hybrid path planning system, characterized in that: include: A human-computer interaction device and the land-sea relay hybrid path planning device according to claim 9, wherein the human-computer interaction device is communicatively connected with the land-sea relay hybrid path planning device; The human-computer interaction device is used to receive the influencing factors input by the user, and to display the pre-path planning results and the real-time path planning results to the user; The land-sea relay hybrid path planning device is used to process data information input by a user, and perform pre-path planning and real-time path planning in combination with a target cost function.

Images