Multi-ship anchoring scheduling path planning method based on improved RRT algorithm

Abstract

The present invention relates to a multi-ship anchoring scheduling path planning method based on an improved RRT algorithm, comprising the following steps: S1, constructing a multi-ship anchoring scheduling destination selection mechanism; S2, constructing a multi-ship anchoring scheduling evaluation index selection mechanism; S3, calculating the ship scheduling priority according to the multi-ship anchoring scheduling evaluation index; S4, constructing a multi-ship anchoring scheduling collision avoidance decision-making mechanism; S5, calculating the multi-ship anchoring scheduling path planning result. The present invention can be used to deal with the situation where multiple ships need anchoring scheduling at the same time, optimize their berthing sequence and navigation path, thereby minimizing the ship anchoring scheduling waiting time, reducing the navigation distance of all scheduled ships as much as possible, and improving the berth utilization rate; it has important significance in multi-ship scheduling in complex waters, emergency evacuation of ships, and comprehensive management of water transportation hubs.

Description

Technical Field

[0001] The present invention relates to the technical field of ship scheduling, and more specifically to a multi-ship anchoring scheduling path planning method based on an improved RRT algorithm. Background Art

[0002] As a core public facility for water transport production and shipping economic development, anchorage plays a vital role in the safe operation of ships, smooth waterways and navigation safety. However, the existing anchorage scheduling system mainly relies on manual experience and simple algorithms, and its intelligence and adaptive capabilities are obviously insufficient. This limitation makes it difficult for the scheduling system to make optimal decisions in a complex and changing environment, which not only makes the scheduling efficiency low, but also makes it prone to errors.

[0003] Traditional path planning methods show poor collision avoidance capabilities when multiple ships are anchored at the same time, especially in complex waterway conditions, where the risk of collision between ships increases significantly. The current collision avoidance mechanism lacks the flexibility to deal with dynamically changing environmental factors, and it is difficult to effectively ensure the safety of ships in uncertain environments. In addition, existing scheduling systems generally lack an effective priority allocation mechanism, causing some ships to wait too long during the anchoring process, while other ships can complete the operation quickly. This unbalanced scheduling reduces the overall efficiency of the system and may cause waterway congestion.

[0004] The unreasonable allocation of berth resources is also a major shortcoming of the existing scheduling strategy. Some berths are often vacant while others are overcrowded, which seriously affects the overall efficiency of anchor scheduling. Traditional anchor scheduling methods usually only focus on a single evaluation indicator (such as the shortest path or minimum sailing time), ignoring the comprehensive consideration of multi-dimensional factors such as ship type, weather conditions, anchorage characteristics, etc., resulting in unscientific and unreasonable decision-making.

[0005] In complex water environments, the existing ship scheduling efficiency is low, and it is difficult to effectively manage the scheduling operations of multiple ships at the same time, resulting in a lengthy scheduling process and low berth utilization. The main challenge facing multi-ship anchoring scheduling path planning is how to reasonably arrange the navigation paths of multiple ships in a complex environment to minimize costs, improve efficiency, and maximize safety. In summary, the existing anchoring scheduling methods have significant shortcomings in the problem of multiple ships anchoring at the same time. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide a multi-ship anchoring scheduling path planning method based on an improved RRT algorithm, which can solve the situation where multiple ships need anchoring scheduling at the same time, and reduce the sailing distance of all scheduled ships as much as possible, minimize the ship scheduling waiting time and improve the berth utilization rate.

[0007] The technical solution adopted by the present invention to solve the technical problem is: constructing a multi-ship anchoring scheduling path planning method based on an improved RRT algorithm, comprising the following steps:

[0008] S1. Construct a multi-vessel anchorage dispatch destination selection mechanism;

[0009] S2. Construct a mechanism for selecting evaluation indicators for multi-ship anchoring scheduling;

[0010] S3. Calculate the ship dispatch priority according to the multi-ship anchoring dispatch evaluation index;

[0011] S4. Construct a multi-ship anchoring scheduling and collision avoidance decision-making mechanism;

[0012] S5, calculating the multi-ship anchoring scheduling path planning results;

[0013] In the step S2, the ship conditions, anchorage channel conditions, and anchorage berth conditions are considered when constructing a multi-ship anchorage scheduling evaluation index selection mechanism;

[0014] The vessel conditions include the type of vessel, the loading conditions of the vessel and the deadweight tonnage of the vessel;

[0015] The anchorage berth conditions include the ratio of the length of the ship to the length of the berth, the ratio of the draft of the ship to the water depth at the front of the berth, the ratio of the length of the ship to the diameter of the turning water area, the unberthing method and the dispatching distance;

[0016] The anchorage channel conditions include the ratio of the ship's draft to the channel depth, the ratio of the ship's width to the channel width, and the degree of channel curvature;

[0017] In the step S3, when calculating the ship scheduling priority according to the multi-ship anchoring scheduling evaluation index, the entropy weight method in the objective weighting method is used to calculate the weight coefficient of each evaluation index of the multi-ship anchoring scheduling, and then the priority of each ship is calculated by the weighted summation method, and the importance of each ship is ranked;

[0018] The specific steps for calculating the weight coefficients of various evaluation indicators for multi-ship anchoring scheduling using the entropy weight method in the objective weighting method are as follows:

[0019] S301, positive processing of the evaluation index of multi-vessel anchoring scheduling priority;

[0020] Normalize the original data of each evaluation index. Assuming that evaluation indicators, , assuming the processed value is:

[0021]

[0022] S302, calculate the Under the evaluation index The proportion of indicators per ship :

[0023]

[0024] S303, calculate the The entropy value of the evaluation index;

[0025] According to the definition of information entropy in information theory, the information entropy of a set of data is calculated as:

[0026]

[0027] Among them, if , then define ;

[0028] S304, calculating the weight of each evaluation index;

[0029] According to the calculation formula of information entropy, the information entropy of each evaluation index is calculated as , then the weight of each evaluation index is calculated by information entropy:

[0030]

[0031] in, is the information entropy complexity.

[0032] According to the above scheme, in step S1, the factors considered when constructing the multi-ship anchorage scheduling destination selection mechanism include ship type, ship characteristics, weather conditions, anchorage type, distance from the ship to the anchorage, and route direction;

[0033] The formula for selecting a suitable anchorage for a ship is as follows:

[0034]

[0035] in, For ships and anchorage The matching degree, , ; The draft of the ship, The water depth at low tide for the anchorage; is the distance between the ship and the anchorage, The larger the value, the higher the matching degree.

[0036] According to the above scheme, in step S4, when constructing a multi-ship anchoring scheduling collision avoidance decision-making mechanism, a ship safety zone is set around the ship:

[0037]

[0038]

[0039] in, is the length of the ship safety zone, is the width of the ship safety zone, is the length of the ship, and the elliptical area around the ship is the safety zone of the ship.

[0040] According to the above scheme, in step S5, the RRT-Connect algorithm is used to quickly plan the initial path for multi-ship anchoring scheduling, and in view of the shortcomings of the RRT-Connect algorithm, three strategies of redundant point deletion, gradient descent path optimization, and kinematic constraints are proposed to perform secondary planning on the initial path, and the initial path generated by the RRT-Connect algorithm is optimized and improved to generate a short, safe, and smooth path.

[0041] According to the above scheme, the multi-ship anchoring scheduling path planning framework based on the improved RRT algorithm includes four parts: known environmental information, initial path planning, secondary path planning and anchoring scheduling path.

[0042] According to the above scheme, the known environmental information includes obstacles, waterways and anchorages. The obstacles are divided into dynamic obstacles and static obstacles. The dynamic obstacles mainly include dynamic ships and dynamic navigation marks. The static obstacles mainly include coastlines and reefs. The waterways and anchorages are the location information of the waterways and the location information of the anchorages.

[0043] The initial path planning uses the RRT-Connect algorithm to quickly plan a collision-free initial path between any two navigation points based on the spatial distribution information of the ship's navigation nodes;

[0044] The secondary path planning includes three strategies: redundant point deletion, gradient descent path optimization, and kinematic constraints.

[0045] According to the above scheme, the redundant point deletion strategy includes the following steps:

[0046] (1) Assuming the initial path It consists of N path nodes: ;

[0047] (2) Utilization Traversal , calculate the path Whether a collision occurs, if so, Add to the non-deletable point set middle;

[0048] (3) Traverse the collection again Find the shortest path between two adjacent points and finally integrate them to get the shortest path;

[0049] The gradient descent path optimization strategy is as follows: If the loss function containing the target parameters is reasonably designed , then by optimizing the loss function To obtain the optimal target parameters, in the path planning task, the coordinate points of the path As the target parameter to be optimized, and design a suitable loss function , and finally use the gradient descent method to find the optimal target parameters , thereby optimizing the initial path;

[0050] The loss function is composed of three indicators: the distance between the path point and the obstacle, the path length, and the tortuosity of the path. :

[0051]

[0052] To penalize points on the path that are close to obstacles, define for:

[0053]

[0054] in for The weight coefficient of represents any point on the path, Indicates departure The nearest obstacle point, when the path The distance between any point on the hour ,otherwise , when the gradient descent method is used to reduce value, making the waypoint away from obstacles;

[0055] Defining Cost Items and Constrain the direction of parameter optimization;

[0056]

[0057]

[0058] in , They are and The weight parameter, Path Node The slope of the tangent line at Representative path The total length of The bigger; Representative path The more tortuous the path, The larger the loss function is, the Add to and Constrain the direction of gradient descent to ensure that the path optimization result does not lead to excessive path length or tortuosity;

[0059] The kinematic constraint strategy is specifically: using the Durbin curve to further optimize the global path, the composition is one of the six combinations of RSR, RSL, LSR, LSL, RLR, and LRL:

[0060] The calculation steps for obtaining the optimal Durbin curve are as follows:

[0061] (1) Traverse the six combinations of RSR, RSL, LSR, LSL, RLR, and LRL;

[0062] (2) Draw the maximum curvature circle passing through the starting point and the end point under this combination;

[0063] (3) Find the tangent lines between two circles;

[0064] (4) Connect the three paths to obtain the Durbin curve;

[0065] (5) Select the shortest path combination from the six calculation results as the optimal Durbin curve.

[0066] The multi-ship anchoring scheduling path planning method based on the improved RRT algorithm of the present invention has the following beneficial effects:

[0067] 1. The present invention has significant advantages in solving the problem of multiple ships needing anchoring scheduling at the same time. It realizes a new method of efficient, safe and reasonable path planning in a complex environment, which significantly improves the intelligence level and decision-making efficiency of the system, provides an innovative solution for the safety and efficiency of ship scheduling, and has stronger adaptive ability and intelligent collision avoidance ability in a multi-ship anchoring environment, significantly improving the overall efficiency and safety of anchorage management;

[0068] 2. The present invention can optimize the selection of anchoring destinations. By constructing a multi-ship anchoring scheduling destination selection mechanism, it comprehensively considers various factors such as ship type, characteristics, weather conditions, anchorage type, distance to anchorage, and route direction. It not only improves the scientificity and rationality of anchoring destination selection, but also ensures that each ship can find the anchorage that best suits its characteristics, thereby reducing unnecessary sailing distance and time and improving overall scheduling efficiency;

[0069] 3. The comprehensive evaluation index system of the present invention introduces a comprehensive multi-ship anchoring scheduling evaluation index selection mechanism, covering ship conditions, such as type, loading condition, deadweight tonnage, anchorage channel conditions, such as draft and channel depth ratio, width ratio, and berth conditions, such as length ratio, draft ratio, etc. The entropy weight method is used to calculate the weight coefficient of each evaluation index, and then determine the priority of each ship, which can more objectively reflect the differences between different ships, making scheduling decisions more fair and reasonable, and further improving the safety and efficiency of anchoring operations;

[0070] 4. In the collision avoidance decision mechanism of the present invention, a safety zone is set for each ship, and a collision avoidance decision mechanism is constructed on this basis, which can effectively avoid the collision risk that may occur when multiple ships are simultaneously conducting anchor scheduling, and ensure the safety of the entire scheduling process. By applying secondary programming strategies such as redundant point deletion, gradient descent path optimization and kinematic constraints on the initial path, the safety and smoothness of the path are further enhanced, and potential risks are reduced;

[0071] 5. The path planning efficiency of the present invention is improved. The RRT-Connect algorithm is used to quickly generate a collision-free initial path, and the path is improved in combination with a variety of optimization strategies; these strategies include but are not limited to deleting redundant points to simplify the path structure, optimizing the path shape by gradient descent method to improve smoothness, and optimizing the Durbin curve to ensure that the path meets the actual navigation requirements. The resulting path is not only short and safe, but also meets complex kinematic constraints, greatly improving the overall efficiency of anchor scheduling;

[0072] 6. The present invention improves the utilization rate of berths: Through the effective implementation of the above series of measures, the sailing distance and waiting time of all dispatched ships can be minimized while ensuring safety, thereby effectively improving the utilization rate of berths, which is of great significance for alleviating anchorage congestion and improving anchorage dispatching efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:

[0074] Figure 1 It is a flow chart of a multi-ship anchoring scheduling path planning method based on an improved RRT algorithm of the present invention;

[0075] Figure 2 It is a framework diagram of multi-ship anchoring scheduling path planning based on the improved RRT algorithm of the present invention;

[0076] Figure 3 It is a schematic diagram of the RRT-Connect algorithm expansion of the present invention. DETAILED DESCRIPTION

[0077] In order to have a clearer understanding of the technical features, purposes and effects of the present invention, specific embodiments of the present invention are now described in detail with reference to the accompanying drawings.

[0078] like Figure 1 As shown, the multi-ship anchoring scheduling path planning method based on the improved RRT algorithm of the present invention comprises the following steps:

[0079] S1. Construct a multi-ship anchorage scheduling destination selection mechanism. When constructing a multi-ship anchorage scheduling destination selection mechanism, the first factors to be considered are the ship type and anchorage type. The multi-ship anchorage scheduling destination is selected according to the anchorage type corresponding to the ship, and the anchorage with suitable water depth is selected according to the ship's draft combined with the ship's own characteristics, ship maneuverability, etc.; secondly, the anchorage with suitable geographical location is selected according to weather conditions; finally, the anchorage with a short distance for multi-ship anchorage scheduling is selected according to the distance between the ship and the anchorage and the route direction.

[0080] The formula for selecting a suitable anchorage for a ship is as follows:

[0081]

[0082] in, For ships and anchorage The matching degree, , ; The draft of the ship, The water depth at low tide is for the anchorage; is the distance between the ship and the anchorage. The larger the value, the higher the matching degree.

[0083] S2. Construct a multi-ship anchoring scheduling evaluation index selection mechanism. When constructing a multi-ship anchoring scheduling evaluation index selection mechanism, factors such as ship conditions, anchorage channel conditions, and anchorage berth conditions are considered.

[0084] Ship conditions include ship type, ship loading condition and ship deadweight tonnage; anchorage berth conditions include the ratio of ship length to berth length, the ratio of ship draft to water depth at the front of berth, the ratio of ship length to the diameter of turning water area, unberthing method and dispatching distance; anchorage channel conditions include the ratio of ship draft to channel depth, the ratio of ship width to channel width and the degree of channel curvature.

[0085] Ship type: According to different uses, ships are divided into different types, and different ships have different degrees of impact on ship safety. Ship types are divided into three types: tankers, container ships, and cargo ships. The three types of ships are assigned values ​​from large to small. The larger the value, the higher the priority of multi-ship anchor scheduling. Ship loading condition: The loading of the ship directly affects the draft stability of the ship. When fully loaded, the inertia of the ship is large, so the degree of danger is high, and the relative risk during ship scheduling is also large. The ship loading condition is divided into three conditions: full load, half load, and empty load. The ship loading condition is assigned values ​​from large to small. The larger the value, the higher the priority of multi-ship anchor scheduling. Ship deadweight tonnage: The deadweight tonnage of the ship represents the scale of the ship. The larger the index, the higher the priority of multi-ship anchor scheduling. The ratio of the length of the ship to the length of the berth: The longer the length of the ship, the more difficult it is to operate the ship, and the more difficult the ship scheduling process is. Therefore, the ratio of the length of the ship to the berth is selected as the evaluation index of multi-ship anchor scheduling. The larger the value, the higher the priority of multi-ship anchor scheduling. The ratio of the ship's draft to the water depth at the front of the berth: The navigation of the ship is affected by the water depth. When the water depth is not enough, the maneuverability of the ship will be reduced. The ratio of the ship's draft to the water depth at the front of the berth is selected as the evaluation index for multi-ship anchor scheduling. The larger the ratio, the higher the priority of multi-ship anchor scheduling. The ratio of the ship's length to the diameter of the turning water area: The larger the ratio, the more difficult it is for the ship to operate in the water area, and the higher the priority of multi-ship anchor scheduling. Unberthing method: Unberthing methods include forward departure, U-turn departure and vertical departure. Different departure methods have different risk values. The three departure methods are assigned from small to large. The larger the value, the higher the priority of multi-ship anchor scheduling.

[0086] Evacuation distance: In step S1, a multi-ship anchoring scheduling destination selection mechanism is established. The starting point of the ship is the same, but the end point of the anchoring scheduling is different. Therefore, several alternative multi-ship anchoring scheduling destinations are assigned values ​​as decision indicators. The closer the scheduling distance, the smaller the assigned value, and the higher the priority of the ship, which will improve the efficiency of multi-ship anchoring scheduling to a certain extent. Ratio of ship draft to channel depth: The ratio of ship draft to channel depth is selected as the evaluation index. The larger the ratio, the higher the risk factor of the ship in the scheduling process, and the higher the priority of multi-ship anchoring scheduling. Ratio of ship width to channel width: The ratio of ship width to channel width is selected as the indicator. The larger the ratio, the higher the risk of the ship in the scheduling process, and the higher the priority of multi-ship anchoring scheduling. Channel curvature: The more curved the channel, the more unfavorable it is for ship maneuvering. Therefore, the maximum turning angle of the navigation channel between the ship berth and the anchorage is taken as the evaluation index. The larger the turning angle, the higher the priority of multi-ship anchoring scheduling.

[0087] S3. Calculate the ship scheduling priority according to the multi-ship anchoring scheduling evaluation index. When calculating the ship scheduling priority according to the multi-ship anchoring scheduling evaluation index, the entropy weight method in the objective weighting method is used to calculate the weight coefficient of each evaluation index of the multi-ship anchoring scheduling, and then the weighted summation method is used to calculate the priority of each ship, and the importance of each ship is ranked.

[0088] The specific steps of using the entropy weight method in the objective weighting method to calculate the weight coefficients of each evaluation index of multi-ship anchoring scheduling are as follows:

[0089] (1) Positive processing of the priority evaluation indicators for multi-ship anchoring scheduling.

[0090] Normalize the raw data of each evaluation index. Assuming that evaluation indicators, . Assume that the processed value is:

[0091]

[0092] (2) Calculate the Under the evaluation index The proportion of indicators per ship :

[0093]

[0094] (3) Calculate the The entropy value of the evaluation index.

[0095] According to the definition of information entropy in information theory, the information entropy of a set of data is calculated as:

[0096]

[0097] Among them, if , then define .

[0098] (4) Calculate the weight of each evaluation indicator.

[0099] According to the calculation formula of information entropy, the information entropy of each evaluation index is calculated as , then the weight of each evaluation index is calculated by information entropy.

[0100]

[0101] in, is the information entropy complexity.

[0102] According to the above steps, the weight coefficient of each evaluation index is obtained. The weight coefficient represents the importance of the evaluation index in the entire multi-ship anchor scheduling priority evaluation system. Next, the weighted summation method is used to calculate the priority of each ship and rank the importance of ship anchor scheduling. In the process of multi-ship anchor scheduling, the ship priorities are compared to clarify which ships need to avoid other ships.

[0103] S4. Construct a multi-ship anchoring scheduling collision avoidance decision-making mechanism. When constructing a multi-ship anchoring scheduling collision avoidance decision-making mechanism, it is necessary to set up a safety zone around the ship. When another ship enters the safety zone of the ship, or the safety zones of two ships intersect, it is determined that the two ships have collided.

[0104] Establish ship safety domain in accordance with relevant standards for inland waterways:

[0105]

[0106]

[0107] in, is the length of the ship safety zone, is the width of the ship safety zone, is the length of the ship, and the elliptical area around the ship is the safety zone of the ship.

[0108] After setting the destination of the ship's anchorage berth scheduling, the environment around the ship is detected. At a certain moment, when the ship's safety domain is detected and there are no other ships in the ship's safety domain, the ship performs path planning in a static environment; when there are other ships in the ship's safety domain, the priorities of the ships are compared. If the priority of the other ship is higher than that of the ship, the ship regards the other ship as part of the dynamic obstacle and performs path planning in the dynamic environment. Otherwise, the other ship is not considered and the path planning in the static environment continues.

[0109] S5. Calculate the path planning results for multi-ship anchoring scheduling. The RRT-Connect algorithm is used to quickly plan the initial path for multi-ship anchoring scheduling. In view of the shortcomings of the RRT-Connect algorithm, three strategies, namely redundant point deletion, gradient descent path optimization, and kinematic constraints, are proposed to perform secondary planning on the initial path. The initial path generated by the RRT-Connect algorithm is optimized and improved to generate a short, safe, and smooth path.

[0110] like Figure 2 As shown in the figure, the multi-ship anchoring scheduling path planning framework based on the improved RRT algorithm. The multi-ship anchoring scheduling path planning framework based on the improved RRT algorithm includes four parts: known environment information, initial path planning, secondary path planning and anchoring scheduling path.

[0111] Known environmental information includes obstacles, waterways, anchorages and other related information. Obstacles are divided into dynamic obstacles and static obstacles. Dynamic obstacles mainly include dynamic ships, dynamic navigation marks, etc., and static obstacles mainly include coastlines, reefs, etc. Waterways and anchorages mainly include the location information of waterways and anchorages.

[0112] Initial path planning uses the RRT-Connect algorithm to quickly plan a collision-free initial path between any two navigation points based on the spatial distribution information of the ship's navigation nodes. The RRT-Connect algorithm is a method that directly searches by sampling in continuous space. It does not require discrete division and mapping of the environment, avoids additional overhead in environmental modeling, and uses a heuristic tree for heuristic search, so it has high search capability and efficiency.

[0113] like Figure 3 As shown in the figure, the RRT-Connect algorithm expansion diagram shows the expansion process of the RRT-Connect algorithm. and end point Two search trees are established for path search, and the search capability and computational efficiency are taken into account at the same time through the two strategies of random expansion and heuristic expansion. Adopt random expansion strategy to expand, randomly sample 1 point in space each time And expand in this direction to get the path segment , to explore the space; and the search tree Adopt heuristic expansion strategy to expand, each time only to the point Direction expansion to obtain path segments , achieving fast convergence of the algorithm.

[0114] Secondary path planning includes three strategies: redundant point deletion, gradient descent path optimization, and kinematic constraints.

[0115] 1. The specific strategy for deleting redundant points is as follows:

[0116] In the RRT-Connect algorithm, the growth of the tree is random, and its random expansion strategy leads to many redundant points in the planned path, which further causes the path tortuous, non-smooth, and too long path length. Therefore, the redundant points in the path can be deleted to achieve the purpose of reducing the path length.

[0117] The redundant point removal strategy is a method based on traversal and detection, which specifically includes the following steps:

[0118] (1) Assuming the initial path It consists of N path nodes: .

[0119] (2) Utilization Traversal , calculate the path Whether a collision occurs, if so, Add to the non-deletable point set middle.

[0120] (3) Traverse the collection again Find two adjacent points and find the shortest path between them, and finally integrate to get the shortest path.

[0121] 2. The specific gradient descent path optimization strategy is:

[0122] The path search of the RRT-Connect algorithm is based on a random sampling strategy, and the searched path may be close to the edge of the obstacle. Therefore, a method needs to be proposed to adjust the path points close to the obstacles in the path so that these path points maintain a safe distance from the obstacles.

[0123] Gradient descent is a classic parameter optimization method. If the loss function containing the target parameters is properly designed , then we can optimize the loss function To find the optimal target parameters. Drawing on the idea of ​​gradient descent, in path planning tasks, the coordinate points of the path can be As the target parameter to be optimized, and design a suitable loss function , and finally use the gradient descent method to find the optimal target parameters , thereby optimizing the initial path.

[0124] Since the planned path should not be close to obstacles, the length should be as short as possible, and the curvature should be as smooth as possible, the distance between the path point and the obstacle, the path length, and the tortuosity of the path can be used together to form a loss function. :

[0125]

[0126] To penalize points in the path that are close to obstacles, define for:

[0127]

[0128] in, for The weight coefficient of represents any point on the path, Indicates departure The nearest obstacle point. It can be found that when the path The distance between any point on the hour ,otherwise Therefore, the gradient descent method is used to reduce value, you can keep the waypoint away from obstacles.

[0129] In order to avoid increasing the length of the path and the tortuosity of the path in the reverse direction during the parameter optimization process, the cost term is defined and Constrain the direction of parameter optimization.

[0130]

[0131]

[0132] in, , They are and The weight parameter, Path Node The slope of the tangent line at . It can be found that Representative path The total length of The bigger; Representative path The more tortuous the path, The larger the value, the greater the loss function. Add to and The direction of gradient descent can be constrained to ensure that the path optimization result does not lead to excessive path length or tortuosity.

[0133] In summary, the gradient descent method can be used to optimize the loss function , thereby reversely optimizing the target parameters , so that the path reaches the optimal level in terms of smoothness, path length, and distance from obstacles.

[0134] 3. The specific kinematic constraint strategy is:

[0135] The motion model of the ship is simplified into a three-degree-of-freedom motion model, namely, movement in the X direction, movement in the Y direction, and rotation in the plane. Therefore, kinematic constraints such as the ship's posture angle and the ship's minimum turning radius need to be considered when planning the path. The above path planning strategy does not consider kinematic constraints such as the ship's minimum turning radius and the ship's posture, so the resulting path is still not suitable for ship navigation. Therefore, it is proposed to use the Durbin curve to further optimize the global path.

[0136] The Durbin curve is a method to find the shortest bounded curvature smooth path between two directed points. Durbin, the proposer of the algorithm, proved that the shortest bounded curvature smooth path must consist of three segments, and the composition must be one of the six combinations of RSR, RSL, LSR, LSL, RLR, and LRL.

[0137] The calculation steps for obtaining the optimal Durbin curve are as follows:

[0138] (1) Traverse the six combinations of RSR, RSL, LSR, LSL, RLR, and LRL;

[0139] (2) Draw the maximum curvature circle passing through the starting point and the end point under this combination;

[0140] (3) Find the tangent lines between two circles;

[0141] (4) Connect the three paths to obtain the Durbin curve;

[0142] (5) Select the shortest path combination from the six calculation results as the optimal Durbin curve.

[0143] By obtaining known environmental information, the RRT-Connect algorithm is used to quickly plan the initial path of multi-ship anchoring scheduling, and the initial path is secondary planned. The initial path generated by the RRT-Connect algorithm is optimized and improved, and finally the ship anchoring scheduling path is obtained.

[0144] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are within the protection of the present invention.

Claims

1. A multi-ship anchoring scheduling path planning method based on an improved RRT algorithm, characterized in that: The following steps are involved: S1. Construct a multi-vessel anchorage dispatch destination selection mechanism; S2. Construct a mechanism for selecting evaluation indicators for multi-ship anchoring scheduling; S3. Calculate the ship dispatch priority according to the multi-ship anchoring dispatch evaluation index; S4. Construct a multi-ship anchoring scheduling and collision avoidance decision-making mechanism; S5, calculating the multi-ship anchoring scheduling path planning results; In the step S2, the ship conditions, anchorage channel conditions, and anchorage berth conditions are considered when constructing a multi-ship anchorage scheduling evaluation index selection mechanism; The vessel conditions include the type of vessel, the loading conditions of the vessel and the deadweight tonnage of the vessel; The anchorage berth conditions include the ratio of the length of the ship to the length of the berth, the ratio of the draft of the ship to the water depth at the front of the berth, the ratio of the length of the ship to the diameter of the turning water area, the unberthing method and the dispatching distance; The anchorage channel conditions include the ratio of the ship's draft to the channel depth, the ratio of the ship's width to the channel width, and the degree of channel curvature; In the step S3, when calculating the ship scheduling priority according to the multi-ship anchoring scheduling evaluation index, the entropy weight method in the objective weighting method is used to calculate the weight coefficient of each evaluation index of the multi-ship anchoring scheduling, and then the priority of each ship is calculated by the weighted summation method, and the importance of each ship is ranked; The specific steps for calculating the weight coefficients of various evaluation indicators for multi-ship anchoring scheduling using the entropy weight method in the objective weighting method are as follows: S301, positive processing of the evaluation index of multi-vessel anchoring scheduling priority; Normalize the original data of each evaluation index. Assuming that evaluation indicators, , assuming the processed value is: S302, calculate the Under the evaluation index The proportion of indicators per ship : S303, calculate the The entropy value of the evaluation index is used to calculate the information entropy of a set of data: Among them, if , then define ; S304, calculating the weight of each evaluation index; According to the calculation formula of information entropy, the information entropy of each evaluation index is calculated as , then the weight of each evaluation index is calculated by information entropy: in, is the information entropy complexity.

2. The multi-ship anchoring scheduling path planning method based on the improved RRT algorithm according to claim 1 is characterized in that: In the step S1, factors considered when constructing a multi-ship anchorage scheduling destination selection mechanism include ship type, ship characteristics, weather conditions, anchorage type, distance from the ship to the anchorage, and route direction; The formula for selecting a suitable anchorage for a ship is as follows: in, For ships and anchorage The matching degree, , ; The draft of the ship, The water depth at low tide is for the anchorage; is the distance between the ship and the anchorage.

3. The multi-ship anchoring scheduling path planning method based on the improved RRT algorithm according to claim 1 is characterized in that: In step S4, when constructing a multi-ship anchoring scheduling collision avoidance decision-making mechanism, a ship safety zone is set around the ship: in, is the length of the ship safety zone, is the width of the ship safety zone, is the length of the ship, and the elliptical area around the ship is the safety zone of the ship.

4. The multi-ship anchoring scheduling path planning method based on the improved RRT algorithm according to claim 1 is characterized in that: In step S5, the RRT-Connect algorithm is used to quickly plan the initial path for multi-ship anchoring scheduling, and the initial path is secondary planned through three strategies: redundant point deletion, gradient descent path optimization, and kinematic constraints. The initial path generated by the RRT-Connect algorithm is optimized and improved to generate a short, safe, and smooth path.

5. The multi-ship anchoring scheduling path planning method based on the improved RRT algorithm according to claim 4 is characterized in that: The multi-ship anchoring scheduling path planning framework based on the improved RRT algorithm includes: known environmental information, initial path planning, secondary path planning and anchoring scheduling path.

6. The multi-vessel anchoring scheduling path planning method based on the improved RRT algorithm according to claim 5 is characterized in that: The known environmental information includes obstacles, waterways and anchorages. The obstacles are divided into dynamic obstacles and static obstacles. The dynamic obstacles mainly include dynamic ships and dynamic navigation marks. The static obstacles mainly include coastlines and reefs. The waterways and anchorages are the location information of the waterways and the location information of the anchorages. The initial path planning uses the RRT-Connect algorithm to quickly plan a collision-free initial path between any two navigation points based on the spatial distribution information of the ship's navigation nodes; The secondary path planning includes three strategies: redundant point deletion, gradient descent path optimization, and kinematic constraints.

7. The multi-vessel anchoring scheduling path planning method based on the improved RRT algorithm according to claim 6 is characterized in that: The redundant point deletion strategy includes the following steps: (1) Assuming the initial path It consists of N path nodes: ; (2) Utilization Traversal , calculate the path Whether a collision occurs, if so, Add to the non-deletable point set middle; (3) Traverse the collection again Find the shortest path between two adjacent points and finally integrate them to get the shortest path; The gradient descent path optimization strategy is as follows: If the loss function containing the target parameters is reasonably designed , then by optimizing the loss function To obtain the optimal target parameters, in the path planning task, the coordinate points of the path As the target parameter to be optimized, and design a suitable loss function , and finally use the gradient descent method to find the optimal target parameters , thereby optimizing the initial path; The loss function is composed of three indicators: the distance between the path point and the obstacle, the path length, and the tortuosity of the path. : To penalize points on the path that are close to obstacles, define for: in for The weight coefficient of represents any point on the path, Indicates departure The nearest obstacle point, when the path The distance between any point on the hour ,otherwise , when the gradient descent method is used to reduce value, so that the waypoint is away from obstacles; Defining Cost Items and Constrain the direction of parameter optimization; in , They are and The weight parameter, Path Node The slope of the tangent line at Representative path Total length; Representative path The degree of tortuosity; The kinematic constraint strategy is specifically: using the Durbin curve to further optimize the global path, the composition is one of the six combinations of RSR, RSL, LSR, LSL, RLR, and LRL: The calculation steps for obtaining the optimal Durbin curve are as follows: (1) Traverse the six combinations of RSR, RSL, LSR, LSL, RLR, and LRL; (2) Draw the maximum curvature circle passing through the starting point and the end point under this combination; (3) Find the tangent lines between two circles; (4) Connect the three paths to obtain the Durbin curve; (5) Select the shortest path combination from the six calculation results as the optimal Durbin curve.

Images