Coastal channel water area ship path planning method

Abstract

The invention relates to the technical field of ship path planning methods, in particular to a coastal channel water area ship path planning method, which comprises the following steps of: preprocessing an electronic chart; generating an adjacent graph; channel division; initial path generation: performing path search based on an A star algorithm according to the adjacent graph, and introducing a channel preference factor into a cost function of the A star algorithm to generate an initial path; and finally generating an optimized path. When the A star algorithm is used for path search, a channel preference factor is introduced into a cost function of the A star algorithm, channel priority is considered in the planning process of an initial path, namely, the initial path is generated in a channel area in a channel place, and the initial path is optimized. Meanwhile, the initial path is generated as close to the outer edge of the water channel or the channel of the ship starboard as possible, and the problem that the generated path deviates from the channel and crosses the channel in practical application is avoided.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ship path planning methods, and more particularly to a coastal waterway ship path planning method. BACKGROUND

[0002] The "Intelligent Ship Specification 2025" released by China Classification Society points out that an intelligent ship is composed of six functional modules, and the intelligent navigation module is one of the main research directions. Path planning, as one of the core technologies of the intelligent navigation module, plays an important supporting role in realizing the autonomous, safe and efficient navigation of ships. Unlike general shortest path search or pure obstacle avoidance problems, the path planning of intelligent ships faces higher compliance and safety requirements, and needs to design the route under the premise of following the "International Regulations for Preventing Collisions at Sea". Especially in coastal waters, maritime agencies generally demarcate channels, requiring ships to navigate within the channel to maintain navigation order and safety. At the same time, the ninth article (narrow waterway) of the "International Regulations for Preventing Collisions at Sea" clearly requires ships to "travel as close as practicable to the outer edge of the waterway or channel on their right side", and not to obstruct the safe passage of other ships.

[0003] However, the existing technology mainly focuses on environmental modeling, search algorithm improvement and path smoothing optimization, although it has achieved rich results, but most of the researches have not fully considered the mandatory constraints of coastal channels, and lack of clear compliance with the narrow waterway navigation rules, resulting in the generated path deviating from the channel and crossing the channel in actual application, affecting the compliance and safety of the path. SUMMARY

[0004] The purpose of the present application is to overcome the deficiency that the path generated by the intelligent ship in the prior art deviates from the channel and crosses the channel in actual application, and to provide a coastal waterway ship path planning method that considers channel priority in the planning process to avoid the problem of the generated path deviating from the channel and crossing the channel in actual application.

[0005] To solve the above technical problems, the technical solution adopted by the present application is: A coastal waterway ship path planning method is provided, comprising the following steps: Electronic chart preprocessing: analyzing electronic chart data to extract geographic features, and converting the geographic features into a vectorized geographic feature file; Adjacency graph generation: constructing an environment model based on a quadtree multi-strategy modeling method according to the vectorized geographic feature file, and generating an adjacency graph that can be used for path search based on the environment model; Channel division: generating a center line of a channel according to the vectorized geographic feature file, and dividing the channel into an upbound channel and a downbound channel along the center line, the channel directions of the upbound channel and the downbound channel being parallel to the center line; Initial path generation: performing path search based on an A-star algorithm according to the adjacency graph, and introducing a channel preference factor into a cost function of the A-star algorithm to generate an initial path; the channel preference factor is specifically: judging whether a currently explored path point is located in a channel region; if the path point is in a non-channel region, the cost is not changed; if the path point is in the channel region, the cost is adjusted according to a heading determination function in combination with a channel zoning type in which the path point is located; the heading determination function is determined according to a current heading of a ship and a channel direction defined by a current channel; Final optimized path generation: reducing redundant nodes and unnecessary turns of the initial path to generate a final optimized path.

[0006] The coastal waterway ship path planning method of the present application analyzes electronic chart data, extracts geographic elements, converts the geographic elements into a vectorized geographic element file conducive to rapid calculation and spatial analysis, constructs an environment model based on a quadtree multi-strategy modeling method according to the vectorized geographic element file, generates an adjacency graph that can be used for path searching, simultaneously performs waterway division, generates a center line of the waterway according to the vectorized geographic element file, divides the waterway along the center line into an upstream waterway and a downstream waterway, the waterway directions of the upstream waterway and the downstream waterway are both parallel to the center line, performs path searching based on the A-star algorithm according to the adjacency graph, and introduces a waterway preference factor into the cost function of the A-star algorithm to generate an initial path. The waterway preference factor is specifically: judging whether the currently explored path point is located in a waterway region: if the path point is in a non-waterway region, the cost is not changed; if the path point is in a waterway region, the cost is adjusted according to a heading determination function in combination with the waterway zoning type in which the path point is located; the heading determination function determines according to the current heading of the ship and the specified waterway direction of the waterway in which the ship is currently located; the waterway priority is considered in the planning process to avoid the problem that the generated path deviates from the waterway or crosses the waterway in actual application. After the initial path is generated, the redundant nodes and unnecessary turns of the initial path are reduced to generate a final optimized path. By introducing the waterway preference factor into the cost function of the A-star algorithm when the A-star algorithm is used for path searching, the waterway priority is considered in the planning process of the initial path, that is, the initial path is generated in the waterway region where there is a waterway, avoiding the problem that the generated path deviates from the waterway or crosses the waterway in actual application, and by setting the upstream waterway and the downstream waterway, the initial path is generated as close as possible to the outer edge of the waterway on the right side of the ship. Thus, not only can a feasible path be generated in a complex coastal waterway, but also the waterway priority and the constraint of Rule 9 of the International Regulations for Preventing Collisions at Sea can be considered in the planning process, thereby improving the standardization, reliability and actual executability of the path.

[0007] Preferably, in the waterway division, a segmentation line is set at a turning point of the waterway according to the vectorized geographic element file, the waterway is segmented and regularized to obtain waterway segments, and the center line of the waterway is obtained by connecting the midpoints of adjacent two segmentation lines. The main traffic lane and two transition lanes are divided in the upstream waterway and the downstream waterway, the main traffic lane is located between the two transition lanes, and the division lines for dividing the main traffic lane and the two transition lanes are parallel to the center line of the waterway. The waterway is segmented and regularized to simplify irregular waterways and facilitate subsequent processing. The main traffic lane and the two transition lanes are divided in the upstream waterway and the downstream waterway to facilitate subsequent driving of the ship in the middle position of the upstream waterway and the downstream waterway.

[0008] Preferably, the cost function is specifically:

[0009] wherein, represents the current leaf node; represents the cost of node . represents the neighbor leaf node of the current leaf node; represents the cost of node . represents the channel preference factor; represents the Euclidean distance from the current leaf node to the neighbor leaf node; represents the Euclidean distance from node to the end point of the planned path.

[0010] Preferably, the heading decision function is specifically:

[0011]

[0012] wherein, represents the angle difference between the current heading of the ship in path generation and the channel heading specified by the channel in which the ship is currently located; represents the minimum value; represents the heading of the current ship in path generation; represents the channel heading specified by the channel in which the ship is currently located; represents the coordinates of the current position of the ship; represents the coordinates of the previous position of the ship; represents the arctangent function.

[0013] Preferably, the channel preference factor is specifically:

[0014]

[0015] wherein, represents the channel preference factor; represents the basic cost; represents the weight; represents the weight when the channel zoning type is the main traffic zone; represents the weight when the channel zoning type is the transition zone.

[0016] Preferably, the method for segmenting and regularizing the channel according to the vectorized geographic feature file, to obtain channel segments, specifically includes the following steps: extracting channel boundaries from the vectorized geographic feature file, the channel boundaries being composed of multiple channel line segment boundaries; Calculate the included angle of adjacent fairway line segment boundaries along the fairway boundary, and determine the intersection point of the adjacent fairway line segment boundaries as a turning point of the fairway when the included angle is greater than a threshold value; At the turning point, set the direction of the segmentation line according to the azimuth angle of the fairway line segment boundary: when the azimuth angle falls within the range of 0°±45° or 180°±45°, set the direction of the segmentation line as a horizontal direction; when the azimuth angle does not fall within the range of 0°±45° and 180°±45°, set the direction of the segmentation line as a vertical direction; the segmentation line extends to the fairway line segment boundary on both sides to obtain a plurality of fairway initial segments. When the length of the fairway initial segment is not greater than a preset length, set the fairway initial segment as a final fairway segment; when the length of the fairway initial segment is greater than the preset length, supplementally set the segmentation line at equal intervals to the fairway initial segment so that the length of the fairway initial segment is not greater than the preset length, and obtain the final fairway segment. Collect the plurality of final fairway segments to obtain a fairway segment.

[0017] Preferably, the final optimized path generation specifically comprises the following steps: Divide the initial paths located in the fairway into fairway initial paths respectively, and divide the initial paths not located in the fairway into free space segment paths respectively; Segment each fairway initial path according to the fairway segment to obtain a fairway segment path; Simplify the fairway segment path and the free space segment path respectively by using the RDP algorithm to reduce the redundant nodes and unnecessary turns of the initial path, and generate a final optimized path. This facilitates the final optimized path to be more in line with the shape of the fairway boundary and avoids directly crossing the fairway.

[0018] Preferably, in the electronic chart preprocessing, the electronic chart data comprises S-57 or S-101 standard data, and the extracted geographic elements comprise fairways, lands, and depth areas.

[0019] Preferably, in the adjacency graph generation, divide the area in the depth area where the water depth is less than the safe water depth of the ship into a shallow water area, divide the area in the depth area where the water depth is not less than the safe water depth of the ship into a deep water area, divide the shallow water area and the land into an unnavigable area, and divide the deep water area into a navigable area; differentially subdivide and mark the fairways, the unnavigable area, and the navigable area based on a quadtree multi-strategy modeling method to construct an environment model.

[0020] Preferably, the construction of the environment model specifically comprises the following strategies: Generate a leaf node: Generate an initial quadtree node; A uniform subdivision strategy is adopted for the waterway, and initial quadtree nodes located in the waterway are subdivided to a preset depth to generate leaf nodes of equal size; A boundary trigger subdivision strategy is adopted for the unnavigable area, and subdivision is triggered only when the boundary of the unnavigable area intersects with an initial quadtree node, and leaf nodes are generated after subdivision to a specified depth; A hierarchical marking strategy is adopted for the navigable area, and initial quadtree nodes located in the deep water area are subdivided to a specified depth to generate leaf nodes; Mark the leaf nodes: Add a waterway mark on the leaf nodes located in the waterway, add an unnavigable mark on the leaf nodes located in the unnavigable area, and add a navigable mark on the leaf nodes located in the navigable area. The environment model is constructed based on the quadtree multi-strategy modeling method, different regions can obtain corresponding subdivision and marking effects according to their characteristics, which not only ensures the fine expression of the model in key regions, but also avoids generating redundant nodes in non-key regions, so as to balance the modeling accuracy and efficiency.

[0021] Compared with the prior art, the beneficial effects of the present application are: 1、The coastal waterway ship path planning method of the present application introduces a channel preference factor in the cost function of the A* algorithm when the A* algorithm is used for path search, considers channel priority in the planning process of the initial path, that is, generates the initial path in the channel area when there is a channel, avoids the problems of deviating from the channel and crossing the channel in the actual application of the generated path, and generates the initial path close to the outer edge of the waterway or channel on the right side of the ship by setting the uplink channel and downlink channel; so as to not only generate a feasible path in a complex coastal waterway, but also consider channel priority and the constraint of Rule 9 of the International Regulations for Preventing Collisions at Sea in the planning process, thereby improving the standardization, reliability and actual executability of the path.

[0022] 2、The coastal waterway ship path planning method of the present application processes the channel by segmentation and normalization, not only obtains the channel direction of the channel segmented path, but also determines the spatial range of the main traffic lane and the transition lane, so as to compare the angle difference between the current heading of the ship in the path generation and the channel direction specified by the channel in the cost function, and give different weights combined with the channel zoning type; the A* algorithm can consider distance, heading consistency and channel zoning priority when accumulating the cost, so as to generate a path located in the main traffic lane and with reasonable heading, effectively avoiding reverse and deviating from the channel.

[0023] 3. The coastal waterway ship path planning method of the present application, by segmenting and regularizing the waterway, obtains a waterway segment path, and based on the initial path, adopts a segment optimization strategy, simplifies the waterway segment path and the free space segment path by using the RDP algorithm, sets different thresholds according to different environmental characteristics, can effectively remove redundant nodes and unnecessary turns, and obtains a more smooth, simple and practical path.

[0024] 4. The coastal waterway ship path planning method of the present application, by constructing an environment model based on a quadtree multi-strategy modeling method, different regions can obtain corresponding subdivision and marking effects according to their characteristics, which not only ensures the fine expression of the model in key areas, but also avoids the generation of redundant nodes in non-key areas, so as to balance the modeling accuracy and efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 is a flowchart of the coastal waterway ship path planning method of the present application; Figure 2 is a flowchart of the waterway division in the present application; Figure 3 is a schematic diagram of the initial path generated by the coastal waterway ship path planning method of the present application; Figure 4 is a flowchart of the calculation of the waterway preference factor in the present application; Figure 5 is a flowchart of the generation of the final optimized path in the present application; Figure 6 is an environment map used in the present application; Figure 7 is a quadtree grid map generated by the coastal waterway ship path planning method of the present application; Figure 8 is a schematic diagram of the adjacency graph generated by the coastal waterway ship path planning method of the present application; Figure 9 is a path comparison diagram of the traditional A-star algorithm result and the final optimized path of the present application; Figure 10 is a comparison diagram of the traditional A-star algorithm result and the result obtained by the present application. DETAILED DESCRIPTION

[0026] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all. The present application will be further described below with reference to specific embodiments.

[0027] In the description of the present application, it should be understood that the orientation or positional relationship indicated by the terms "upper", "lower", "left", "right", etc. is based on the orientation or positional relationship shown in the drawings, which is only for the convenience of describing the present application and simplifying the description, and does not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, therefore the positional relationship described in the drawings is only for exemplary illustration, and cannot be understood as a limitation of the present patent, and for those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0028] Embodiment one The present embodiment is a first embodiment of a coastal waterway ship path planning method, as shown in Figure 1 , comprising the following steps: Electronic chart preprocessing: analyzing electronic chart data to extract geographic features, and converting the geographic features into a vectorized geographic feature file; Adjacency graph generation: constructing an environment model based on the vectorized geographic feature file using a quadtree multi-strategy modeling method, and generating an adjacency graph that can be used for path search based on the environment model; Channel division: extracting channel boundaries from the vectorized geographic feature file, the channel boundaries being composed of multiple channel segment boundaries, as shown in Figure 2 ; calculating the included angle of adjacent channel segment boundaries along the channel boundaries, and determining the intersection point of adjacent channel segment boundaries as a turning point of the channel when the included angle is greater than a threshold value, for example, the threshold value can be set to 5°; at the turning point, setting the segment line direction according to the azimuth angle of the channel segment boundary, the azimuth angle being defined as 0° in the north direction of the map, increasing in the clockwise direction, 90° in the east direction, 180° in the south direction, and 270° in the west direction; when the azimuth angle falls within the range of 0°±45° or 180°±45°, it is determined to be approximately vertical, that is, north-south, and the segment line direction is set to the horizontal direction; when the azimuth angle does not fall within the range of 0°±45° and 180°±45°, it is determined to be approximately horizontal, that is, east-west, and the segment line direction is set to the vertical direction; the segment line is extended to the channel segment boundary on both sides until it intersects, to ensure complete cutting of the channel and formation of regular sub-regions, obtaining multiple channel initial segments; when the length of a channel initial segment is not greater than a preset length, the channel initial segment is set as a channel final segment; when the length of a channel initial segment is greater than the preset length, supplementing the channel initial segment with segment lines at equal intervals to ensure that the length of the channel initial segment is not greater than the preset length, obtaining the channel final segment; to avoid generating excessively long sub-regions; and collecting the multiple channel final segments to obtain channel segments, as shown in Figure 2As shown in (b). After the channel is segmented, each segment line intersects with the left and right boundaries of the channel. The geometric midpoint of the intersection of each segment line and the channel boundary is calculated, and the midpoints of two adjacent segment lines are connected to obtain the centerline of the channel. The channel is divided into an upstream channel and a downstream channel along the centerline according to the right-hand driving rule. The course of the upstream and downstream channels is parallel to the centerline. The course direction is determined according to the right-hand driving rule, that is, when the centerline direction is θ, the prescribed course of the left and right halves of the channel is θ+180° and θ, respectively. A main passage zone and two transition zones are divided in both the upstream and downstream channels. The main passage zone is located between the two transition zones. The dividing lines of the main passage zone and the two transition zones are set parallel to the centerline of the channel, as shown in (b). Figure 2 As shown in (c), the channel is segmented and regularized to simplify irregular channels and facilitate subsequent processing. Both the upstream and downstream channels are divided into a main passage zone and two transition zones. That is, the channel zone types in the upstream and downstream channels include a main passage zone and a transition zone. The main passage zone is the area for priority path generation, and the transition zone is the area for auxiliary path generation, so that ships can navigate in the middle of the upstream and downstream channels.

[0029] Initial path generation: Path search is performed based on the adjacency graph using the A* algorithm. A channel preference factor is introduced into the cost function of the A* algorithm to generate an initial path, controlling the priority selection of the main channel passage during the path search process. Figure 3 As shown. The specific calculation steps for the channel preference factor are as follows. Figure 4 As shown: Determine whether the currently explored path point is located in the waterway area: If the path point is outside the waterway area, the cost is not changed; if the path point is within the waterway area, the cost is adjusted according to the heading determination function and the waterway zone type where the path point is located; the heading determination function makes a determination based on the ship's current heading and the waterway heading prescribed by the current waterway. The cost function is as follows:

[0030] in, Indicates the current leaf node; Represents a node The cost; Indicates the neighboring leaf nodes of the current leaf node; Represents a node The cost; Indicates the channel preference factor; This represents the Euclidean distance from the current leaf node to its neighboring leaf nodes. , and These are nodes and nodes the coordinates of the current position of the ship; the node the Euclidean distance to the end point of the planned path.

[0031] The heading decision function is specifically:

[0032]

[0033] wherein, represents the angle difference between the current heading of the ship and the channel heading specified by the channel in which the ship is currently located in the path generation; represents the minimum value; represents the heading of the current ship in the path generation; represents the channel heading specified by the channel in which the ship is currently located; represents the coordinates of the current position of the ship; represents the coordinates of the previous position of the ship; represents the arctangent function.

[0034] The channel preference factor is specifically:

[0035]

[0036] wherein, represents the channel preference factor; represents the basic cost; represents the weight; represents the weight when the channel zoning type is the main traffic zone; represents the weight when the channel zoning type is the transition zone. The weight in this expression is iteratively adjusted according to the specific generation scheme until the initial path generated is preferentially on the channel when there is a channel, and the initial path is within the channel area.

[0037] Specifically, the above channel preference factor expression is explained as follows: If the angle difference is within 60°, it is considered that the heading is consistent. At this time, the cost is further adjusted according to the channel zoning type: the cost is significantly reduced in the main traffic zone as the preferential navigation area; in the transition zone, the cost is moderately increased, higher than the main traffic zone but still lower than the non-channel area; If the angle difference is greater than 60° and less than 90°, it is considered that the heading is acceptable, but the cost of this path point is not changed; If the angle difference is more than 90°, it is considered that the heading is inconsistent, and the cost is significantly increased, so that this path point cannot be selected, thereby avoiding the ship from reversing or deviating from the channel direction.

[0038] Final optimized path generation: reducing the redundant nodes and unnecessary turns of the initial path, generating the final optimized path.

[0039] The ship autonomously sails in the coastal waterway area according to the final optimized path.

[0040] The working principle of the coastal waterway ship path planning method of the embodiment is as follows: The electronic chart data is parsed, geographic elements are extracted, the geographic elements are converted into a vectorized geographic element file conducive to rapid calculation and spatial analysis, an environment model is constructed based on a quadtree multi-strategy modeling method based on the vectorized geographic element file, an adjacency graph that can be used for path search is generated, channel division is performed, a center line of the channel is generated based on the vectorized geographic element file, the channel is divided into an uplink channel and a downlink channel along the center line, the channel directions of the uplink channel and the downlink channel are both parallel to the center line, path search is performed based on the A-star algorithm based on the adjacency graph, and a channel preference factor is introduced into the cost function of the A-star algorithm to generate an initial path. The channel preference factor is specifically: judging whether the currently explored path point is located in the channel area: if the path point is in the non-channel area, the cost is not changed; if the path point is in the channel area, the cost is adjusted according to a channel direction determination function in combination with the channel zoning type in which the path point is located; the channel direction determination function determines according to the current heading of the ship and the specified channel direction of the channel in which the ship is currently located; the channel priority is considered in the planning process to avoid the problem that the generated path deviates from the channel and crosses the channel in actual application. After the initial path is generated, the redundant nodes and unnecessary turns of the initial path are reduced to generate a final optimized path, as shown in FIG. 4. Figure 9

[0041] By introducing the channel preference factor into the cost function of the A-star algorithm when the A-star algorithm is used for path search, the channel priority is considered in the planning process of the initial path, that is, the initial path is generated in the channel area where there is a channel, and the problem that the generated path deviates from the channel and crosses the channel in actual application is avoided.

[0042] ​By dividing the waterway into segments, not only the centerline direction of each segment is obtained, which is used as the specified heading, but also the spatial range of the main traffic lane and the transition lane is determined. These information is directly called in the improved A-star algorithm path search: when the algorithm extends from the current node n to the neighbor node m, first, the specified heading and the traffic lane type of the segment that node m belongs to is determined according to the position of node m. Then, in the cost function, the specified heading is compared with the actual heading (calculated by the direction from n to m), the angle difference is obtained, and the differentiated weight is given in combination with the traffic lane type. In this way, the A-star algorithm can consider the distance, heading consistency and traffic lane priority when accumulating the cost, so as to generate a path that is located in the main traffic lane and has a reasonable heading, effectively avoiding reverse and deviation from the waterway.

[0043] At the same time, by setting the uplink waterway and the downlink waterway, the initial path is generated as close as possible to the outer edge of the waterway or the waterway on the right side of the ship; thereby not only a feasible path can be generated in the complex coastal waterway area, but also the waterway priority and the constraint of Rule 9 of the International Regulations for Preventing Collisions on the Sea can be considered in the planning process, so as to improve the standardization, reliability and actual executability of the path.

[0044] Embodiment Two This embodiment is the second embodiment of the method for ship path planning in the coastal waterway area, and based on the first embodiment, the specific process of generating the final optimized path in this embodiment is as shown in Figure 5 , and specifically includes the following steps: As shown in (a) of Figure 5 , the initial path is located in the waterway. The initial path located in the waterway is divided into waterway initial paths, and the initial path not located in the waterway is divided into free space segment paths. Each waterway initial path is segmented according to the waterway segment path to obtain the waterway segment path, as shown in (b) of Figure 5 . The RDP algorithm, which is the full name of Ramer-Douglas-Peucker algorithm, is used to simplify the waterway segment path and the free space segment path respectively: specifically, in the shape regular waterway segment path, the threshold is set to be large, so that the optimized route is more consistent with the waterway direction; in the shape irregular waterway segment path, the threshold should be small, so as to avoid the optimized route deviating from the waterway due to excessive simplification. If the free space path segment is in a complex environment, a small threshold value should be set to avoid the optimized path being excessively simplified and causing the path to be unsafe; on the contrary, the free space segment path is in a relatively safe environment, and the threshold should be set to be large, so as to more effectively reduce the number of path points, thereby improving the efficiency of path simplification.

[0045] The initial path is optimized to reduce the redundant nodes and unnecessary turns of the initial path, and different thresholds are set according to the environmental conditions of the area where the initial path is located for segmented optimization, so that the final optimized path is more in line with the shape of the channel boundary, and direct crossing of the channel is avoided, thereby generating the final optimized path, such as Figure 5 (c) shown in the middle.

[0046] Embodiment three This embodiment is a third embodiment of a coastal channel water area ship path planning method. Based on embodiment one, in the electronic chart preprocessing of this embodiment, the electronic chart data includes S-57 or S-101 standard data, the channel, land and water depth area geographic features are extracted, and the geographic features are converted into a vectorized geographic feature file. In this embodiment, the vectorized geographic feature file is an ESRI Shapefile format file, which is used for environmental modeling, as shown in Figure 6 . Specifically, the electronic chart used in this embodiment follows the S-57 standard. In the electronic chart conversion process, GDAL (Geospatial Data Abstraction Library) is used. This library is an open source cross-platform tool that supports multiple formats including S-57 and ESRI Shapefile. The conversion process is stable and reliable. The ogr2ogr tool of GDAL is used to read the S-57 file in ISO / IEC 8211 format, extract the geometry and attribute information of the object, and output it as an independent Shapefile layer according to the geographic feature type, which is used for subsequent environmental modeling and path search.

[0047] In the adjacency graph generation, the geographic coordinate system of the vectorized geographic feature file is converted to the projected coordinate system, the area with water depth less than the ship's safe water depth in the water depth area is divided into shallow water area, and the area with water depth not less than the ship's safe water depth in the water depth area is divided into deep water area. The ship's safe water depth is the sum of the ship's full load draft and the freeboard. The shallow water area and the land are divided into non-navigable areas, and the deep water area is divided into navigable areas. Based on the four-tree multi-strategy modeling method, the channel, non-navigable areas and navigable areas are differentiated and marked, an environmental model is constructed, and an adjacency graph that can be used for path search is generated.

[0048] The constructed environmental model is shown in Figure 7 . Specifically, the following strategies are included: Generate leaf nodes: Generate initial quadtree nodes; Uniform subdivision strategy is adopted for the channel, and the initial quadtree nodes located in the channel are subdivided to a preset depth to generate leaf nodes of equal size; The boundary trigger subdivision strategy is used for the unnavigable area, and the subdivision is triggered only when the boundary of the unnavigable area intersects with the initial quadtree node, and the leaf node is generated after the subdivision to the specified depth; The hierarchical marking strategy is used for the navigable area, and the initial quadtree node located in the deep water area is subdivided to the specified depth to generate the leaf node; Marking the leaf node: Adding the channel mark on the leaf node located in the channel, adding the unnavigable mark on the leaf node located in the unnavigable area, and adding the navigable mark on the leaf node located in the navigable area.

[0049] The environment model is constructed based on the quadtree multi-strategy modeling method, different areas can obtain corresponding subdivision and marking effects according to their characteristics, which not only ensures the fine expression of the model in the key area, but also avoids the generation of redundant nodes in the non-key area, so as to balance the modeling accuracy and efficiency.

[0050] After completing the environment modeling of the electronic chart, all the quadtree leaf nodes are traversed to filter out the navigable nodes located in the channel and the deep water area, and then the adjacent nodes in the neighborhood range of each navigable node are searched, and if there is no land or shallow water obstacle between the two nodes, a connected edge is established to form the traffic relationship between the nodes. Through the above method, the navigable nodes and the connected edges constitute an undirected graph structure, which provides a reliable topological basis for subsequent path planning, as shown in Figure 8 Fig. 2 is a schematic diagram of the generated adjacency graph.

[0051] Example Four In this embodiment, the electronic chart used is the data of Zhanjiang Bay area, which is obtained by fusing three electronic charts CN203315, CN488103 and CN488102. The longitude and latitude range of the research area is 110.398°-110.569° east longitude and 21.063°-21.150° north latitude, covering the main channel of Zhanjiang Bay and the surrounding waters, as shown in Figure 6 In order to facilitate subsequent modeling and path planning calculation, and to ensure the accuracy of distance and angle calculation, the longitude and latitude coordinates (WGS84 geographical coordinates) are converted to UTM projection coordinates in this paper. The projection has smaller distortion in a single projection zone, which can ensure the accuracy of spatial calculation on a regional scale.

[0052] The ship parameters are as follows: the overall length is 147 m, the type width is 22.6 m, the type depth is 11.2 m, the full load draft is 8.2 m, the rich water depth is set as 20% of the ship draft, the safety water depth is the sum of the ship draft and the rich water depth, and therefore the safety water depth of the ship is 9.84 m. In order to reflect the water depth limit condition of the actual ship navigation, the area with a water depth less than 10 m is divided into a shallow water area. The environmental modeling result is shown in Figure 7 .

[0053] As shown in Figure 9 and Figure 10 , the path length generated by the traditional A-star algorithm is the shortest, but the path fails to select a channel and does not comply with the channel navigation rules, and there are obvious defects. Compared with the traditional A-star, the A-star algorithm of the application can generate a path in the main traffic lane of the channel in priority, and meet the rule requirement that the ship should navigate as close as possible to the right side edge in the channel, the path length increases by 6.7%, the node number and the inflection point number are increased, and the path presents more tiny broken lines. After the A-star algorithm result of the application is segmented and optimized as an initial path, the path length is shortened, is close to the traditional A-star algorithm, the node number is reduced to only 18, the inflection point number is only 14, and the path is always kept in the channel, while the key path characteristics are retained, the path simplicity and smoothness are significantly improved.

[0054] In summary, the traditional A-star algorithm is superior in path length, but lacks practical application value due to the failure to select a channel and comply with the navigation rules; the A-star algorithm of the application can generate a path in the main traffic lane of the channel in priority, and meet the rule requirement that the ship should navigate as close as possible to the right side edge in the channel, but the path complexity is increased, affecting the executability; the segmented optimization strategy further improves the path quality on the basis of the A-star algorithm of the application, while the key path characteristics are retained, the node number and the inflection point number are significantly reduced, the path is more simple and smooth, and has practical feasibility, and is more suitable for actual ship navigation.

[0055] In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, the person skilled in the art can combine and combine the different embodiments or examples described in the present specification and the features of the different embodiments or examples without contradiction.

[0056] Obviously, the above embodiments of the present application are merely exemplary but not intended to limit the embodiments of the present application. Based on the above description, any other variations or changes can be made by those skilled in the art without departing from the spirit and principles of the present application. It is not necessary to list all the embodiments here. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present application should be included in the protection scope of the claims of the present application.

Claims

1. A method for planning a ship route in a coastal water area, characterized by, The method comprises the following steps: electronic chart preprocessing: analyzing electronic chart data to extract geographic elements, and converting the geographic elements into a vectorized geographic element file; adjacency graph generation: constructing an environment model based on a quadtree multi-strategy modeling method according to the vectorized geographic element file, and generating an adjacency graph that can be used for path searching based on the environment model; channel division: generating a center line of a channel according to the vectorized geographic element file, and dividing the channel into an uplink channel and a downlink channel along the center line, wherein the channel directions of the uplink channel and the downlink channel are parallel to the center line; initial path generation: performing path searching based on an A-star algorithm according to the adjacency graph, and introducing a channel preference factor into a cost function of the A-star algorithm to generate an initial path; the channel preference factor is specifically: judging whether a currently explored path point is located in a channel region: if the path point is in a non-channel region, the cost is not changed; if the path point is in a channel region, the cost is adjusted according to a heading determination function in combination with a channel zoning type in which the path point is located; the heading determination function is determined according to a current heading of a ship and a specified channel direction of a channel in which the ship is currently located; final optimized path generation: reducing redundant nodes and unnecessary turns of the initial path to generate a final optimized path.

2. The coastal waterway ship route planning method according to claim 1, characterized by, In the channel division, segment lines are set at turning points of the channel according to the vectorized geographic element file, the channel is segmented and regularized to obtain channel segments, the center line of the channel is obtained by connecting midpoints of adjacent two segment lines, and a main traffic zone and two transition zones are divided in the uplink channel and the downlink channel, wherein the main traffic zone is located between the two transition zones, and the division lines for dividing the main traffic zone and the two transition zones are parallel to the center line of the channel.

3. The coastal waterway ship route planning method according to claim 2, characterized by, The cost function is specifically: wherein, represents the current leaf node; represents the cost of node ; represents the neighbor leaf node of the current leaf node; represents the cost of node ; represents the lane preference factor; represents the Euclidean distance from the current leaf node to the neighbor leaf node; represents the Euclidean distance from node to the end point of the planned path.

4. The coastal waterway ship route planning method according to claim 3, characterized by, The heading determination function is specifically: wherein, represents an angle difference between a current heading of the ship and a channel heading prescribed by a channel in which the ship is currently located in the path generation; represents a minimum value; represents a heading of the ship in the path generation; represents a channel heading prescribed by a channel in which the ship is currently located; represents a coordinate of a current position of the ship; represents a coordinate of a previous position of the ship; represents an arc tangent function.

5. The coastal waterway ship route planning method according to claim 4, characterized by, The channel preference factor is specifically: wherein, represents a channel preference factor; represents a basic cost; represents a weight; represents a weight when the channel zoned type is a main traffic zone; represents a weight when the channel zoned type is a transition zone.

6. The coastal waterway ship route planning method according to claim 2, wherein The setting of the segment lines at the turning points of the channel according to the vectorized geographic element file to segment and regularize the channel to obtain channel segments specifically comprises the following steps: extracting channel boundaries from the vectorized geographic element file, wherein the channel boundaries are composed of a plurality of channel segment boundaries; calculating an included angle of adjacent channel segment boundaries along the channel boundaries, and determining an intersection point of the adjacent channel segment boundaries as a turning point of the channel when the included angle is greater than a threshold value; at the turning point, setting a direction of a segment line according to an azimuth angle of a channel segment boundary: when the azimuth angle falls within a range of 0°±45° or 180°±45°, the direction of the segment line is set as a horizontal direction; when the azimuth angle does not fall within the range of 0°±45° and 180°±45°, the direction of the segment line is set as a vertical direction; the segment line extends to the channel segment boundary on both sides until it intersects the channel segment boundary, and a plurality of channel initial segments are obtained; when a length of a channel initial segment path is not greater than a preset length, the channel initial segment is set as a channel final segment; when the length of the channel initial segment is greater than the preset length, an equal-interval segment line is supplemented to the channel initial segment to make the length of the channel initial segment not greater than the preset length, and a channel final segment is obtained; the plurality of channel final segments are summarized to obtain channel segments.

7. The coastal waterway ship route planning method according to claim 2, characterized by, The final optimized path generation specifically comprises the following steps: The initial paths located in the fairway are divided into fairway initial paths respectively, and the initial paths not located in the fairway are divided into free space segment paths respectively; Each of the fairway initial paths is segmented according to a fairway segment to obtain a fairway segment path; The fairway segment paths and the free space segment paths are respectively simplified by using an RDP algorithm to reduce redundant nodes and unnecessary turns of the initial paths, and a final optimized path is generated.

8. The coastal waterway ship route planning method according to any one of claims 1 to 7, characterized by, In the electronic chart preprocessing, the electronic chart data comprises S-57 or S-101 standard data, and the extracted geographic elements comprise a fairway, land and a water depth area.

9. The coastal waterway ship route planning method according to claim 8, wherein, In the adjacency graph generation, an area with a water depth less than a ship safety water depth in the water depth area is divided into a shallow water area, an area with a water depth not less than the ship safety water depth in the water depth area is divided into a deep water area, the shallow water area and the land are divided into an unnavigable area, and the deep water area is divided into a navigable area; the fairway, the unnavigable area and the navigable area are differentiated and subdivided and marked based on a quadtree multi-strategy modeling method to construct an environment model.

10. The coastal waterway ship route planning method according to claim 9, wherein The construction of the environment model specifically comprises the following strategies: Generating a leaf node: Generating an initial quadtree node; For the fairway, a uniform subdivision strategy is used to subdivide the initial quadtree node located in the fairway to a preset depth to generate leaf nodes of equal size; For the unnavigable area, a boundary trigger subdivision strategy is used to trigger subdivision only when the boundary of the unnavigable area intersects with the initial quadtree node, and the leaf nodes are generated after the initial quadtree node is subdivided to a specified depth; For the navigable area, a hierarchical marking strategy is used to subdivide the initial quadtree node located in the deep water area to a specified depth to generate leaf nodes; Marking the leaf nodes: Adding a fairway mark on the leaf nodes located in the fairway, adding an unnavigable mark on the leaf nodes located in the unnavigable area, and adding a navigable mark on the leaf nodes located in the navigable area.

Images