A ship automatic berthing path planning method considering ship maneuvering performance

Abstract

The application discloses a ship automatic berthing path planning method considering ship maneuvering performance, and relates to the technical field of ship berthing. The method comprises the following steps: establishing a ship berthing maneuvering performance calculation function step, including a maneuvering turning radius calculation function and an added mass calculation function; an environment map and ship scale information input and processing step, establishing a grid map of a berthing operation area, inputting a ship berthing starting point and a terminal position, inputting main scale parameters of the ship, generating a collision detection area of the ship and establishing a kinematics equation of the ship; a path planning step based on an automatic berthing path planning algorithm, checking, inquiring and solving collision, environment information, a ship motion equation and a maneuvering performance calculation function in the planning process according to an algorithm flow, obtaining a preliminary ship berthing path, and obtaining a final berthing path curve through berthing path smoothing, speed planning and other calculations, so that the ship berthing process can be safely and smoothly carried out.

Description

Technical Field

[0001] The present invention relates to the technical field of ship berthing, and in particular to a method for planning an automatic ship berthing path taking into account the ship's maneuverability. Background Art

[0002] Ship intelligence is a hot topic among marine engineering researchers worldwide. The process of intelligent navigation encompasses research achievements in ship position and obstacle perception, trajectory planning, and ship motion control. Automatic berthing, a key application scenario within intelligent navigation, encompasses all the key technologies necessary for intelligent navigation. Furthermore, due to the limited navigational space in the waterway, ships are affected by wind, waves, currents, oncoming vessels, obstructions, and shallow water effects. This makes berthing more challenging than normal navigation, placing higher demands on ship motion planning and control. Solving the problem of automatic berthing essentially solves most of the challenges of intelligent navigation. Therefore, research into automatic berthing technology is of great practical significance.

[0003] Common path planning algorithms include artificial potential field method, genetic algorithm, neural network, simulated annealing algorithm, ant colony algorithm, etc. However, due to the complex dock environment where the ship is located, the ship's speed is low and the steering efficiency is poor during the berthing stage. The path planned by the above methods does not take into account the ship's maneuverability during berthing, which may increase the safety risk of the ship during berthing. Summary of the Invention

[0004] In response to the above problems and technical requirements, the inventors have proposed a method for planning an automatic berthing path for a ship that takes into account the ship's maneuverability. The technical solution of the present invention is as follows:

[0005] A method for planning an automatic berthing path for a ship taking into account the maneuverability of the ship comprises the following steps:

[0006] Rasterize the ship berthing area map and determine the indexes of the berthing starting point and berthing end point positions under the rasterized map;

[0007] Initialize the to-be-searched set and the searched set, and add the docking starting point as the initial node of the search process to the to-be-searched set;

[0008] The optimal node to be searched is obtained from the set to be searched. If the posture index of the optimal node to be searched is different from the posture index of the berthing end point, the optimal node to be searched is placed in the searched set, and the minimum maneuvering turning radius of the ship of the optimal node to be searched is determined; based on the postures of the optimal node to be searched and the berthing end point and the minimum maneuvering turning radius of the ship, it is determined whether there is a shortest navigable curve between the two points without collision; if so, the berthing end point is placed in the searched set as the end point of the search process, and the optimal node to be searched is used as the parent node of the end point, and the steps of connecting its parent nodes in sequence from the end point until the initial node is reached are executed; if not, based on the posture of the optimal node to be searched and the ship motion equation, the Runge-Kutta method is used to solve the next berthing node that the ship can reach without collision under different rudder angles, and the optimal node to be searched is used as the parent node of the berthing node; the berthing node that is not in the searched set is placed in the set to be searched, and the step of obtaining the optimal node to be searched from the set to be searched is re-executed;

[0009] If the pose index of the optimal node to be searched is the same as the pose index of the berthing end point, the optimal node to be searched is placed in the searched set as the end point of the search process; in the searched set, starting from the end point, its parent nodes are connected in sequence until the initial node is reached to obtain the optimal berthing planning path.

[0010] The beneficial technical effects of the present invention are:

[0011] The automatic berthing path planning method for ships proposed in this application obtains a reliable optimal berthing planning path by taking into account the constraints of the maneuvering turning radius, additional mass and ship speed during the berthing process when searching for berthing path nodes; and further smoothes the nodes that constitute the optimal berthing planning path so that the generated path is more in line with the actual berthing environment. Finally, a ship motion controller is designed according to the geographical location and berthing speed of each smoothed node to control the rudder angle and propeller speed of the ship, so that the ship can navigate according to the optimal berthing planning path, ensuring the safe and smooth progress of the ship's berthing process. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 This is a flow chart of the method for automatic ship berthing path planning provided by this application.

[0013] Figure 2 This is a neural network structure diagram for generating a calculation function for a ship maneuvering turning radius provided by this application.

[0014] Figure 3 It is the docking node provided by this application n Logical relationship diagram with the parent node parentnode.

[0015] Figure 4 This is a schematic diagram of a navigable curve from the current node to the berthing destination provided by this application.

[0016] Figure 5 It is a schematic diagram of the circular collision area s of a ship provided in this application.

[0017] Figure 6 This is a schematic diagram of ship navigation route search under different rudder angles provided by this application. DETAILED DESCRIPTION

[0018] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0019] like Figure 1 As shown, this embodiment discloses a method for planning an automatic berthing path for a ship taking into account the maneuverability of the ship, comprising the following steps:

[0020] Step 1: Establish a calculation function for the ship's berthing maneuverability performance.

[0021] Among them, the ship's berthing maneuverability includes the maneuvering turning radius and the additional mass formed by the water acting on the ship during berthing.

[0022] When establishing the calculation function of the berthing maneuverability of these two types of ships, firstly, the ship's berthing speed V, water depth h, and rudder angle deg are obtained through tank model tests or numerical calculations. rudder The data of maneuvering turning radius and additional mass under different conditions are used to establish the maneuvering performance data set for the ship berthing process. Optionally, the required data can be obtained by fixing two variables and changing one variable in each test. Secondly, the neural network learning method is used to obtain the maneuvering performance data set with berthing speed V, water depth h, ship rudder angle deg rudder The calculation function of ship maneuvering turning radius fr(·) with water depth as variable and the calculation function of additional mass with water depth as variable are As shown in formulas (1) and (2).

[0023] Formula (1): r = fr (V, h, deg rudder )

[0024] Formula (2):

[0025] Where r is the minimum maneuvering radius of the ship; is the calculation function of the longitudinal added mass, is the calculation function of the lateral additional mass, is the calculation function of the additional mass in the bow direction, is the longitudinal additional mass of the ship, is the transverse additional mass of the ship, is the bow additional mass of the ship. These functions are all learned through neural network. The neural network has a three-layer network structure. The number of nodes in the network is dynamically adjusted according to the amount of learning data. Figure 2 Compared with the traditional database fitting method, the additional mass calculation function and the ship maneuvering turning radius calculation function obtained by neural network learning can effectively reduce the data dimension and improve data calculation efficiency.

[0026] Step 2: Rasterize the ship berthing operation area map and determine the indexes of the berthing starting point and berthing end point positions under the rasterized map.

[0027] Step 2.1: When rasterizing, set the resolution of the rasterized map. map 、Ship rudder angle resolution gre rudder , ship heading resolution gre yaw and ship motion resolution gre motion , and determine the position of obstacles in the rasterized map so that it can be used for automatic berthing path search, where: the ship rudder angle resolution gre rudder and ship motion resolution gre motion Used for dynamic planning of ship berthing process, ship heading resolution gre yaw Used for heading discretization during automatic berthing path search.

[0028] Step 2.2: Determine the starting position of the berthing point, including the geographical location of the berthing starting point in the geodetic coordinate system (posx start ,posy start ) and the ship's heading yaw start The berthing endpoint posture includes the geographical location of the berthing endpoint in the geodetic coordinate system (posx end ,posy end ) and the ship's heading yaw end The geographical location of the berthing starting point is the current position of the ship, the ship's bow direction yaw start Take the current heading of the ship and ensure that the berthing starting point and berthing end point will not collide with obstacles.

[0029] Step 2.3: The corresponding calculation formula for converting the pose of the berthing start and end points into the index under the rasterized map is:

[0030] Formula (3): posxidx=ceil(posx t / gre map +0.5)-1

[0031] posyidx=ceil(posy t / gre map +0.5)-1

[0032] yawidx=ceil[(yaw t +π) / gre yaw ]

[0033] Wherein, ceil(·) indicates rounding up a floating-point number, and subscript t = start or end.

[0034] Step 3: Initialize the set to be searched openset = {} and the searched set closeset = {}, add the docking starting point startpoint as the initial node node0 of the search process to the set to be searched, and now openset = {node0}.

[0035] In this embodiment, a node node contains the following attributes:

[0036] node{posxidx,posyidx,yawidx,posx,posy,yaw,velx,vely,velyaw,parentnode,nodevalue}

[0037] Among them, posx, posy, yaw are the geographical location of the node in the geodetic coordinate system and the heading of the ship respectively; velx, vely, velyaw are the longitudinal velocity, transverse velocity and heading velocity of the node respectively; parentnode is the parent node of the node, indicating that the node is searched by the parent node. The relationship between parentnode and node is as follows: Figure 3 As shown; nodevalue is the cost value of the node.

[0038] Step 4: Get the optimal node to be searched from the openset opt , specifically including:

[0039] Step 4.1: Generate a random number num between 0 and 1 rand ;

[0040] Step 4.2: Calculate the cost value nodevalue of each node in the openset to be searched, and sort them from low to high to obtain the nodevalue sequence. The calculation formula is:

[0041] Formula (4): nodevalue = d posxidx,posyidx +distance n2p +cost steer ·|degrudder |

[0042] Where, d posxidx,posyidx The distance from the grid index (posxidx, posyidx) of the current node position to the berthing end position index, distance n2p Cost is the distance from the grid index of the current node position to the index of its parent node position. steer is the penalty coefficient for steering;

[0043] Step 4.3: To avoid the algorithm from falling into local search, if num rand >0.2, then the node with the smallest value nodevalue is selected as the optimal node to be searched. opt ;

[0044] Step 4.4: If num rand ≤0.2, a node is randomly selected from the first 1 / 3 nodes in the nodevalue sequence as the optimal node to be searched. opt .

[0045] Step 5: Use formula (3) to calculate the optimal node to be searched opt The pose index of the optimal search node node opt The pose index of the endpoint is different from the pose index of the berthing endpoint, then the optimal node to be searched is opt Put it into the searched set closeset and execute step 6. If the optimal node to be searched is node opt If the pose index of is the same as the pose index of the berthing endpoint, it is considered that a path from the berthing starting point to the berthing endpoint is found, and the optimal node to be searched is node opt As the end node of the search process end Put it into the searched set closeset and execute step 11.

[0046] Step 6: Determine the optimal node to be searched opt The minimum maneuvering radius of the ship r node .

[0047] Step 6.1: Use inequality C as the selection condition to determine the optimal node to be searched opt Target speed V node :

[0048] Formula (5):

[0049] C:(posx node-posx end ) 2 +(posy node -posy end ) 2 ≤(aL) 2

[0050] Where, (posx node ,posy node ) is the optimal node to be searched opt The geographical location in the geodetic coordinate system (posx end ,posy end ) is the geographical location of the berthing destination in the geodetic coordinate system, d n2e is the distance from the optimal search node to the berthing end point, L is the ship length, and a is the proportional coefficient. Formula (5) can comprehensively consider the berthing operation experience and the direction of the environmental force to accurately calculate the ideal berthing speed of the ship at the specified node position.

[0051] Step 6.2: The optimal node to be searched opt Water depth at the geographical location h node Target speed V node And the current ship rudder angle deg rudder Input into the ship maneuvering turning radius calculation function fr(·), that is, formula (1), and get the optimal node to be searched node opt The minimum maneuvering radius of the ship r node .

[0052] Step 7: According to the optimal node to be searched opt The position of the berthing endpoint and the minimum maneuvering radius r of the ship node and yaw end , determine whether there is a shortest navigable curve between two points without collision.

[0053] Step 7.1: Optimal search node opt The geographical location in the geodetic coordinate system (posx node ,posy node ), Yaw of the ship node , minimum maneuvering radius of the ship r node , and the geographical location of the berthing endpoint in the geodetic coordinate system (posx end ,posy end ), Yaw of the ship end , minimum maneuvering radius of the ship r end For input data, search for the optimal node to be searched based on the geometric relationship between the two pointsopt Is there a navigable curve between the endpoint and the berthing end point? ship If there are multiple navigable curves ship , then select the navigable curve with the shortest path and execute step 7.2; if there is no navigable curve ship , then go to step 8.

[0054] like Figure 4 As shown, from the current node node to the destination node end Navigable curve ship There are multiple possibilities to meet the constraints, and only one is given as an example. The ship moves along the navigable curve with the turning center O and the turning radius R. ship Travel, where the turning radius R satisfies R≤r node And R≤r end , the ship should meet the constraints of position pos and heading yaw at each node.

[0055] Step 7.2: Based on the ship's circular collision area s, check whether the ship will collide with an obstacle while navigating along the shortest navigable curve. That is, whether the collision area s overlaps with the obstacle area. If so, a collision has occurred, and the process proceeds to Step 8. If not, no collision has occurred, and the process proceeds to Step 10.

[0056] The circular collision area s of the ship is generated according to the input ship scale information, such as Figure 5 As shown in the figure, when static and dynamic obstacles enter the collision area s, it is considered that there is a risk of collision, which needs to be avoided during the automatic berthing path search process. The diameter of the collision area s is greater than the ship length L, usually about 2L.

[0057] Step 8: If the optimal node to be searched is opt There is no shortest navigable curve without collision between the endpoint and the berthing end point, then based on the optimal node to be searched opt The posture and ship motion equations The Runge-Kutta method is used to solve the next berthing node that the ship can reach without collision under different rudder angles. i (i=1,2,…,n), and the optimal node to be searched is node opt As a docking node i , and then go to step 9.

[0058] <1> Establishing the ship's motion equations The specific expression is:

[0059] Formula (6): V s =[velx vely velyaw]

[0060] P s =[posx posy yaw]

[0061]

[0062] Where V s is the velocity vector of the ship, including the longitudinal velocity velx, the transverse velocity vely, and the bowing velocity velyaw; P s is the position vector of the ship, including the geographical location (posx, posy) and the heading yaw in the geodetic coordinate system; R(yaw) is the rotation matrix, m is the mass of the ship, and the · symbol above the parameter indicates the derivative of the variable; F X is the longitudinal force acting on the ship, F Y M is the force acting on the ship in the transverse direction, which includes environmental force, rudder force and propeller force; N is the longitudinal torque acting on the ship, which also includes the environmental torque, rudder torque and propeller torque. The expression is:

[0063] Formula (7): F X =X H0 (V,d,h)+X H (velx,velyaw)+X P +X R (deg rudder )

[0064]

[0065] F Y =Y H (vely,velyaw)+Y R (deg rudder )

[0066] M N =N H (vely,velyaw)+N P +N R (deg rudder )

[0067] Where, X H0 (V, d, h) is a function with ship speed V, draft d, and water depth h as variables, and the output is the straight-line resistance encountered by the ship; X H The output (velx, velyaw) is the hydrodynamic force on the longitudinal motion of the ship, Y H(vely, velyaw) output is the hydrodynamic force on the ship's lateral motion, N H (vely, velyaw) is the hydrodynamic force exerted on the ship’s bow motion; X R (·) is the resistance caused by the rudder, Y R (·) is the lateral force provided by the rudder, N R (·) is the yawing moment caused by the rudder; X P is the thrust of the propeller, N P is the yaw moment caused by the propeller, and R0 is the basic resistance in still water.

[0068] <2> Berthing nodes that can be reached at different rudder angles i The total number n and the lth rudder angle deg rudder,l The calculation formulas are expressed as follows:

[0069] Formula (8):

[0070] Where maxdeg rudder is the maximum allowed rudder angle, and int(·) is the rounding of floating point numbers.

[0071] <3> Solve the next berthing node of the ship under different rudder angles i Then, according to the optimal node to be searched opt To the i-th docking node node i Sailing distance S i and ship motion resolution gre motion , the sailing distance S i Divide into multiple lengths of g i,j =gre motion The small path and node n i,j , where i=1,2,...,n,j=1,2,...,int(S i / gre motion ),like Figure 6 shown.

[0072] <4> If the optimal node to be searched is node opt To the i-th docking node node i Each node n between i,j If there is no collision with any obstacle (i.e. the area of ​​the obstacle overlaps), the i-th berthing node is retained. i ; If any node n i,j If all obstacles collide, the i-th docking node is discarded. i .like Figure 6 As shown, with the nth docking node node nNode n on the relevant navigable path i,int(Si / gre motion)-1 A collision with an obstacle (shaded in the figure) occurs, so the node is discarded. n node.

[0073] It should be noted that based on the optimal node to be searched opt The posture and ship motion equations The Runge-Kutta method is used to solve the next berthing node when the ship does not collide under different rudder angles. i The specific implementation method is a conventional technical means in this field, and therefore is not the focus of this application and will not be described in detail here.

[0074] Step 9: Determine the docking nodes in turn i (i=1,2,…,n) Whether it is in the searched set closeset, if it is already in the closeset, then abandon the node; if not, then dock node node i Put it into the set to be searched, openset, and execute step 4 again.

[0075] Step 10: If the optimal node to be searched is opt If there is a shortest navigable curve between the endpoint and the berthing end point without collision, the berthing end point will be used as the end node of the search process. end Put it into the searched set closeset and put the optimal node to be searched node opt As the end node end The parent node of , goes to step 11.

[0076] Step 11: In the searched set closeset, from the end node node end Start connecting to its parent node parentnode in sequence until reaching the initial node node0, and get the optimal berthing planning path. The node sequence is represented as {node end ,node end-1 ,…,node1,node0}.

[0077] Step 12: Smooth the geographical locations of the nodes that make up the optimal berthing planning path in the geodetic coordinate system.

[0078] According to the geographical location (posx, posy), heading yaw and velocity attributes velx, vely, velyaw of each node in the node sequence, the position and velocity sequence of the ship during berthing are further obtained as follows:

[0079] Pose sequence: POSX = {posx end ,posx end-1 ,…,posx1,posx0}

[0080] POSY={posy end ,posy end-1 ,…,posy1,posy0}

[0081] YAW={yaw end ,yaw end-1 ,…,yaw1,yaw0}

[0082] Velocity sequence: VELX = {velx end ,velx end-1 ,…,velx1,velx0}

[0083] VELY={vely end ,vely end-1 ,…,vely1,vely0}

[0084] VELYAW={velyaw end ,velyaw end-1 ,…,velyaw1,velyaw0}

[0085] According to the position sequence POSX, POSY, apply formula (9) to smooth each berthing node in the sequence, and record the smoothed path sequence as POSX h POSY h .

[0086] Formula (9):

[0087] Where J is the path smoothing objective function, w o is the penalty weight coefficient of the obstacle distance of the optimal berthing planning path, x u is the geographical location x of the uth node on the optimal berthing planning path in the geodetic coordinate system u =(posx u ,posy u ), N is the total number of nodes on the optimal berthing planning path, o u is the geographical location of the obstacle closest to the u-th node, dis max is the maximum allowable distance between the node and the obstacle, w κ is the penalty weight coefficient of the curvature term of the optimal berthing planning path, κ u is the curvature of the u-th node, κ max is the maximum allowed curvature of the node, w sis the penalty coefficient of the smoothing term, Δx u The position vector of the current node, expressed as Δx u =x u -x u-1 By taking the derivative of the path smoothing objective function J, we can apply Perform iterative correction on each node, where k is the number of iterations.

[0088] Step 13: Determine the berthing speed of each node based on the smoothed geographic location and speed attributes of each node.

[0089] Substitute the smoothed geographic location and speed attributes of each node into the following formula to solve the coefficients k3, k2, k1, and k0. The calculation formula is:

[0090] Formula (10):

[0091] Where inv(·) represents the inversion of the matrix; s u is the path length from the u-1th node to the uth node, expressed as s u =||x u -x u-1 ||;v u is the berthing speed of the u-th node, including the longitudinal speed velx u , lateral speed u , calculated as v u =||velx u ,vely u ||;a u is the berthing acceleration of the u-th node, which is usually 0. After obtaining the coefficients k3, k2, k1, and k0, the berthing speeds of the u-th node and the u+1-th node are calculated according to formula (11). The planned speed sequence is VEL = {v end ,v end -1,…,v1,v0}.

[0092] Formula (11):

[0093] Step 14: According to the smoothed geographic location of each node POSX h POSY h The ship motion controller is designed based on the planned berthing speed VEL to control the ship's rudder angle and propeller speed, so that the ship can navigate according to the optimal berthing planned path and finally complete the ship's automatic berthing process.

[0094] The above description is only a preferred embodiment of the present application, and the present invention is not limited to the above embodiment. It is understood that other improvements and variations directly derived or imagined by those skilled in the art without departing from the spirit and concept of the present invention should be considered to be included in the scope of protection of the present invention.

Claims

1. A method for planning a ship's automatic berthing path taking into account the ship's maneuverability, characterized in that: The method comprises: Rasterize the ship berthing area map and determine the indexes of the berthing starting point and berthing end point positions under the rasterized map; Initialize the to-be-searched set and the searched set, and add the docking starting point as the initial node of the search process to the to-be-searched set; Obtain the optimal node to be searched from the set to be searched. If the posture index of the optimal node to be searched is different from the posture index of the berthing end point, the optimal node to be searched is placed in the searched set, and the minimum maneuvering turning radius of the ship of the optimal node to be searched is determined; based on the postures of the optimal node to be searched and the berthing end point and the minimum maneuvering turning radius of the ship, determine whether there is a shortest navigable curve between the two points without collision; if so, place the berthing end point as the end node of the search process into the searched set, and use the optimal node to be searched as the parent node of the end node, and execute the steps of connecting its parent nodes in sequence starting from the end node until reaching the initial node; if not, based on the posture of the optimal node to be searched and the ship motion equation, use the Runge-Kutta method to solve the next berthing node that the ship can reach without collision under different rudder angles, and use the optimal node to be searched as the parent node of the berthing node; place the berthing node that is not in the searched set into the set to be searched, and re-execute the step of obtaining the optimal node to be searched from the set to be searched; If the pose index of the optimal node to be searched is the same as the pose index of the berthing destination, the optimal node to be searched is placed in the searched set as the destination node of the search process; in the searched set, starting from the destination node, its parent nodes are connected in sequence until the initial node is reached to obtain the optimal berthing planning path; The method for determining the minimum maneuvering radius of a ship of the optimal node to be searched comprises: Use inequality C as the selection condition to determine the target speed V of the optimal node to be searched node : C:(posx node -posx end ) 2 +(posy node -posy end ) 2 ≤(aL) 2 ; Where, (posx node ,posy node ) is the geographical location of the optimal node to be searched in the geodetic coordinate system, (posx end ,posy end ) is the geographical location of the berthing destination in the geodetic coordinate system, d n2e is the distance from the optimal search node to the berthing destination, L is the ship length, and a is the proportional coefficient; The water depth h at the geographical location of the optimal node to be searched node Target speed V node And the current ship rudder angle deg rudder Input into the ship maneuvering turning radius calculation function fr(·) to obtain the ship's minimum maneuvering turning radius r of the optimal node to be searched node , r node =fr(V node ,h node ,deg rudder ).

2. The method for planning an automatic berthing path for a ship considering the maneuverability of the ship according to claim 1, characterized in that: The method for obtaining the optimal node to be searched from the set to be searched includes: Calculate the cost value nodevalue of each node in the set to be searched, and sort them from low to high to obtain a nodevalue sequence; where nodevalue=d posxidx,posyidx +distance n2p +cost steer |deg rudder |; If the generated random number is greater than 0.2, the node with the smallest replacement value nodevalue is selected as the optimal node to be searched; if the generated random number is less than or equal to 0.2, a node is randomly selected from the first 1 / 3 nodes in the nodevalue sequence as the optimal node to be searched; wherein, the random number is between 0 and 1; Where, d posxidx,posyidx The distance from the grid index (posxidx, posyidx) of the current node position to the berthing end position index, distance n2p Cost is the distance from the grid index of the current node position to the index of its parent node position. steer is the penalty coefficient for steering, deg rudder is the ship's rudder angle.

3. The method for planning an automatic berthing path for a ship considering the maneuverability of the ship according to claim 1, characterized in that: The method for determining whether there is a shortest navigable curve between two points without collision based on the posture of the optimal node to be searched and the berthing end point and the minimum maneuvering turning radius of the ship includes: The geographical location of the optimal node to be searched in the geodetic coordinate system (posx node ,posy node ), the ship's bow yaw of the optimal node to be searched node , the minimum maneuvering radius r of the ship at the optimal node to be searched node , and the geographical location of the berthing end point in the geodetic coordinate system (posx end ,posy end ), Yaw of the vessel at the berthing destination end , the minimum maneuvering radius of the ship at the berthing end point r end For input data, a graph is constructed based on the geometric relationship between two points to search whether there is a navigable curve between the optimal node to be searched and the berthing end point; If there are multiple navigable curves, the navigable curve with the shortest path is selected, and the circular collision area s of the ship is used to detect whether the ship will collide with the obstacle when navigating along the shortest navigable curve; if no collision occurs, then there is a shortest navigable curve without collision; if a collision occurs, or if no navigable curve exists, then there is no shortest navigable curve without collision; The diameter of the circular collision area s of the ship is greater than the length of the ship.

4. The method for planning an automatic berthing path for a ship considering the maneuverability of the ship according to claim 1, characterized in that: The ship's equation of motion The specific expression is: V s =[velx vely velyaw]; P s =[posx posy yaw]; Where V s is the velocity vector of the ship, including the longitudinal velocity velx, the transverse velocity vely, and the bow velocity velyaw; P s is the position vector of the ship, including the geographical location (posx, posy) and the heading yaw in the geodetic coordinate system; R(yaw) is the rotation matrix, and m is the mass of the ship; is the longitudinal additional mass of the ship, is the transverse additional mass of the ship, is the bow additional mass of the ship, and the additional mass calculation function with water depth as the variable is Calculated; F X is the longitudinal force acting on the ship, F Y is the lateral force acting on the ship, M N is the longitudinal moment acting on the ship, and its expression is: F X =X H0 (V,d,h)+X H (velx,velyaw)+X P +X R (deg rudder ); F Y =Y H (vely,velyaw)+Y R (ten rudder ); M N =N H (vely,velyaw)+N P +N R (deg rudder ); Where, X H0 (V, d, h) is a function with ship speed V, draft d, and water depth h as variables, and the output is the straight-line resistance encountered by the ship; X H The output (velx, velyaw) is the hydrodynamic force on the longitudinal motion of the ship, Y H (vely, velyaw) output is the hydrodynamic force on the ship's lateral motion, N H (vely, velyaw) is the hydrodynamic force exerted on the ship’s bow motion; X R (·) is the resistance caused by the rudder, Y R (·) is the lateral force provided by the rudder, N R (·) is the yawing moment caused by the rudder, deg rudder is the ship's rudder angle; X P is the thrust of the propeller, N P is the yaw moment caused by the propeller, and R0 is the basic resistance in still water.

5. The method for planning an automatic berthing path for a ship considering the maneuverability of the ship according to claim 1, characterized in that: In the method of using the Runge-Kutta method to solve the next berthing node that a ship can reach without collision under different rudder angles, wherein: The lth rudder angle deg rudder,l The calculation formula is expressed as: Where maxdeg rudder is the maximum permissible rudder angle, gre rudder is the ship rudder angle resolution, n is the total number of the next docking nodes that can be reached, expressed as: int(·) is used to round floating point numbers; After solving the next berthing node of the ship under different rudder angles, the sailing distance S from the optimal search node to the i-th berthing node is obtained. i and ship motion resolution gre motion , the sailing distance S i Divide into multiple lengths of g i,j =gre motion The small path and node n i,j , where i=1,2,...,n,j=1,2,...,int(S i / gre motion ); If each node n between the optimal search node and the i-th docking node i,j If there is no collision with any obstacle, the i-th berthing node is retained; otherwise, the i-th berthing node is discarded.

6. The method for planning an automatic berthing path for a ship considering the maneuverability of the ship according to claim 1, characterized in that: The method for determining the indexes of the berthing starting point posture and the berthing ending point posture under the rasterized map includes: The berthing starting point position includes the geographical location of the berthing starting point in the geodetic coordinate system (posx start ,posy start ) and the ship's heading yaw start The berthing end position includes the geographical location of the berthing end in the geodetic coordinate system (posx end ,posy end ) and the ship's heading yaw end , which is converted into the corresponding calculation formula for the index under the rasterized map: posxidx=ceil(posx t / gre map +0.5)-1; posyidx=ceil(posy t / gre map +0.5)-1; yawidx=ceil[(yaw t +π) / gre yaw ]: In the formula, ceil(·) means rounding up the floating point number, gre map is the grid map resolution, gre yaw is the ship heading resolution, subscript t = start or end.

7. The method for automatic berthing path planning of a ship considering the maneuverability of the ship according to any one of claims 1 to 6, characterized in that: The method further comprises: The geographical locations of the nodes constituting the optimal berthing planning path in the geodetic coordinate system are smoothed, and the berthing speed of each node is determined based on the smoothed geographical locations and speed attributes of each node. A ship motion controller is designed based on the smoothed geographical locations and berthing speeds of each node to control the ship's rudder angle and propeller speed, so that the ship can navigate according to the optimal berthing planning path.

8. The method for planning an automatic berthing path for a ship considering the maneuverability of the ship according to claim 7, characterized in that: The calculation formula for smoothing the geographical locations of the nodes constituting the optimal berthing planning path in the geodetic coordinate system is: Where J is the path smoothing objective function, w o is the penalty weight coefficient of the obstacle distance of the optimal berthing planning path, x u is the geographical location x of the uth node on the optimal berthing planning path in the geodetic coordinate system u =(posx u ,posy u ), N is the total number of nodes on the optimal berthing planning path, o u is the geographical location of the obstacle closest to the u-th node, dis max is the maximum allowable distance between the node and the obstacle, w κ is the penalty weight coefficient of the curvature term of the optimal berthing planning path, κ u is the curvature of the u-th node, κ max is the maximum allowed curvature of the node, w s is the penalty coefficient of the smoothing term, Δx u The position vector of the current node, expressed as Δx u =x u -x u-1 ; By deriving the path smoothing objective function J, we can apply Perform iterative correction on each node, where k is the number of iterations.

9. The method for planning an automatic berthing path for a ship considering the maneuverability of the ship according to claim 7, characterized in that: The method for determining the berthing speed of each node based on the smoothed geographical location and speed attributes of each node includes: Substitute the smoothed geographic location and speed attributes of each node into the following formula to solve the coefficients k3, k2, k1, and k0. The calculation formula is: Where inv(·) represents the inversion of the matrix; s u is the path length from the u-1th node to the uth node, expressed as s u =||x u -x u-1 ||,x u is the geographical location x of the uth node on the optimal berthing planning path in the geodetic coordinate system u =(posx u ,posy u );v u is the berthing speed of the u-th node, including the longitudinal speed velx u , lateral speed u , calculated as v u =||velx u ,vely u ||;a u is the berthing acceleration of the u-th node; After obtaining the coefficients k3, k2, k1, and k0, the berthing speeds of the u-th node and the u+1-th node are calculated according to the following formula:

Images