Ship efficient obstacle avoidance guidance and control method with sparse sea ice anti-collision mechanism

Abstract

The invention discloses a ship efficient obstacle avoidance guidance and control method with a sparse sea ice anti-collision mechanism. The ship efficient obstacle avoidance guidance and control method comprises the steps that a ship three-degree-of-freedom nonlinear iterative mathematical model is constructed; a heading angle guidance switching model with a sparse sea ice anti-collision mechanism is constructed so as to be used for a nonlinear iteration mathematical model to achieve floating ice collision avoidance; based on a heading angle guidance switching model, constructing a ship virtual control law according to a heading angle tracking error and a position tracking error; introducing a filter to carry out filtering processing on the ship virtual control law to obtain a ship dynamic error; and constructing a ship iterative learning controller of a nonlinear iterative mathematical model, and performing obstacle avoidance guidance and control with a sparse sea ice anti-collision mechanism on the ship according to the ship iterative learning controller. The problems that efficient obstacle avoidance guidance cannot be achieved when a traditional method faces a random irregular polygonal obstacle, and high-stability control cannot be achieved by accumulating previous control experience for the high-precision control requirement of a polar region ship path tracking control task are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of ship motion control, and particularly to a method for efficient obstacle avoidance guidance and control of ships with a sparse sea ice anti-collision mechanism. Background Art

[0002] With the accelerating rate of polar glacier disintegration and ablation, the Arctic shipping route has shown great economic and strategic value. However, the complex polar navigation environment has greatly increased the difficulty of ship maneuvering. For example, large-sized floating ice is randomly distributed, the signal transmission channel is limited, and the actuator response is slow. The above difficulties are likely to cause problems such as ice entrapment and navigation system failures. Therefore, an anti-collision mechanism for sparse sea ice obstacle avoidance guidance and high-precision integrated control strategy is crucial for ensuring polar navigation safety.

[0003] The existing ice area obstacle avoidance guidance and polar ship path tracking control strategies have the following obvious defects:

[0004] (1) For large-sized randomly distributed floating ice, as Figure 2 shown, the traditional obstacle avoidance algorithm that plans obstacles as circular or grid areas is likely to cause unnecessary maneuvering problems, which increases the control difficulty to a certain extent and violates the navigation requirements of polar ships to safely and quickly leave the ice area;

[0005] (2) During the autonomous navigation of polar ships, they face objective difficulties such as maneuvering in restricted waters and avoiding floating ice. At the same time, they need to overcome the control difficulties caused by their own large inertia, strong coupling, and long time delay. A method that meets the high-precision control requirements of polar ship path tracking tasks remains to be explored. Summary of the Invention

[0006] The present invention provides a method for efficient obstacle avoidance guidance and control of ships with a sparse sea ice anti-collision mechanism to overcome the above technical problems.

[0007] To achieve the above object, the technical solution of the present invention is:

[0008] A method for efficient obstacle avoidance guidance and control of ships with a sparse sea ice anti-collision mechanism specifically includes the following steps:

[0009] S1: Construct a non-linear iterative mathematical model of the ship's three degrees of freedom;

[0010] S2: Obtain a reference position signal according to the set logical virtual ship, and obtain a position tracking error according to the non-linear iterative mathematical model to obtain a reference heading angle signal;

[0011] S3: Construct a heading angle guidance switching model with a sparse sea ice anti-collision mechanism according to the logical virtual ship;

[0012] S4: Based on the heading angle guidance switching model, obtain the heading angle tracking error according to the reference heading angle signal, and construct the virtual control law of the ship according to the heading angle tracking error and the position tracking error;

[0013] S5: Introduce a first-order filter to filter the virtual control law of the ship to obtain the ship dynamics error; construct a ship iterative learning controller for the nonlinear iterative mathematical model according to the ship dynamics error, and perform obstacle avoidance guidance and control with a sparse sea ice anti-collision mechanism on the ship according to the ship iterative learning controller.

[0014] Furthermore, for the nonlinear iterative mathematical model of the three degrees of freedom of the ship constructed in S1, its model expression is

[0015]

[0016] where: k represents the iteration number; x k represents the abscissa of the ship's position; y k represents the ordinate of the ship's position; ψ k represents the heading angle of the ship; u k represents the forward speed of the ship; v k represents the transverse drift speed of the ship; r k represents the yaw angular velocity of the ship; m u represents the added hydrodynamic mass of the ship in the forward degree of freedom; m v represents the added hydrodynamic mass of the ship in the transverse drift degree of freedom; m r represents the added hydrodynamic mass of the ship in the yaw degree of freedom; f u (v) represents the nonlinear term of the ship in the forward degree of freedom; f v (v) represents the nonlinear term of the ship in the transverse drift degree of freedom; f r (v) represents the nonlinear term of the ship in the yaw degree of freedom; v represents the transpose matrix of the motion acceleration of the ship in the three degrees of freedom direction and v = [u k , v k , r k T ; X |u|u represents the hydrodynamic coefficient related to the ship's heaving motion in the heaving degree of freedom; X vr represents the hydrodynamic coefficient related to the ship's swaying and yawing motions in the heaving degree of freedom; X vv represents the hydrodynamic coefficient related to the ship's swaying motion in the heaving degree of freedom; X rr represents the hydrodynamic coefficient related to the ship's yawing motion in the heaving degree of freedom; Y v represents the hydrodynamic coefficient related to the first power of the ship's swaying speed in the swaying degree of freedom; Y |v|v ​Represents the hydrodynamic coefficient related to the square of the sway velocity of the ship in the sway degree of freedom; Y r Represents the hydrodynamic coefficient related to the first power of the yaw velocity of the ship in the sway degree of freedom; Y |r|r Represents the hydrodynamic coefficient related to the square of the yaw velocity of the ship in the sway degree of freedom; Y vvr Represents the hydrodynamic coefficient related to the square of the sway velocity and the first power of the yaw velocity of the ship in the sway degree of freedom; Y vrr Represents the hydrodynamic coefficient related to the square of the sway velocity and the first power of the yaw velocity of the ship in the sway degree of freedom; N v Represents the hydrodynamic coefficient related to the first power of the sway velocity of the ship in the yaw degree of freedom; N r Represents the hydrodynamic coefficient related to the first power of the yaw velocity of the ship in the yaw degree of freedom; N |v|v Represents the hydrodynamic coefficient related to the square of the lateral drift velocity of the ship in the yaw degree of freedom; N |r|r Represents the hydrodynamic coefficient related to the square of the yaw velocity of the ship in the yaw degree of freedom; N vvr Represents the hydrodynamic coefficient related to the square of the sway velocity and the first power of the yaw motion velocity of the ship in the yaw degree of freedom; N vrr Represents the hydrodynamic coefficient related to the first power of the sway velocity and the square of the yaw velocity of the ship in the yaw degree of freedom; d wi (v), i = u, v, r are the disturbing forces and moments acting on the ship in the forward, lateral drift, and yaw degrees of freedom due to ocean environmental disturbances; n represents the main engine speed passing through the servo system of the steering gear; δ represents the rudder angle passing through the servo system of the steering gear; T u (·) represents the actuator gain corresponding to the main engine speed; F r (·) represents the actuator gain corresponding to the rudder angle; |·| represents the absolute value of the variable; Represents x k , y k , ψ k , u k , v k , r k The first derivative of

[0017] Furthermore, the S2 specifically includes the following steps:

[0018] S21: Obtain a reference position signal according to the set logical virtual ship VS;

[0019] And the expression of the reference position signal is

[0020]

[0021] In the formula: x d Represents the longitudinal position coordinate of VS; y dRepresents the horizontal position coordinate of VS; ψ d Represents the heading angle of VS; u d Represents the speed of VS in the forward degree of freedom; Represents x d , y d The first derivative of;

[0022] S22: Based on the reference position signal, obtain the position tracking error according to the non-linear iterative mathematical model to obtain the reference heading angle signal;

[0023] And the expression of the reference heading angle signal is

[0024]

[0025] In the formula: x e,k Represents the lateral position tracking error; y e,k Represents the longitudinal position tracking error; sign(·) represents the sign function; arctan(·) represents the arctangent function; ψ r,k Represents the reference heading angle signal.

[0026] Furthermore, the construction method of the heading angle guidance switching model with a sparse sea ice anti-collision mechanism in S3 specifically includes the following steps:

[0027] S31: Based on navigation experience, set a navigation target point in the sea area with sparse sea ice, and represent the virtual ship's heading angle of the logical virtual ship as;

[0028]

[0029] In the formula: ψ d Represents the heading angle of VS; x t , y t Represents the abscissa and ordinate of the position of the navigation target point;

[0030] S32: Based on the set navigation target point, construct the latest collision avoidance distance dist between the ship and the sea ice floes in the sparse sea ice sea area according to the logical virtual ship vo , specifically including:

[0031] S321: Through satellite remote sensing or unmanned aerial vehicle imaging technology, obtain the known irregular sea ice floes in the sparse sea ice sea area, and define them as a polygon obstacle with n vertices along the outer contour of the irregular sea ice floe in a clockwise direction;

[0032] Among them, the vertex coordinates of the polygon obstacle are defined as o1(x o1 , y o1 ), o2(x o2 , y o2 ),..., on (x on , y on ); o n (x on , y on ) represents the coordinate position of the nth vertex;

[0033] S322: Represent the relative distance from the current position of the logical virtual ship to each vertex coordinate in the clockwise order of the polygonal obstacle as d vi , i = 1...n; Represent the side length of each side of the polygonal obstacle as d i(i+1) , i = 1...n, and its expression is

[0034]

[0035] In the formula: x oi represents the abscissa of the ith vertex of the polygonal obstacle; y oi represents the ordinate of the ith vertex of the polygonal obstacle; x o(i+1) represents the abscissa of the (i + 1)th vertex of the polygonal obstacle; y o(i+1) represents the ordinate of the (i + 1)th vertex of the polygonal obstacle;

[0036] S323: According to the relative distance d vi and the side length d i(i+1) of the polygonal obstacle, obtain the latest collision avoidance distance between the ship and the sea ice floes in the sparse sea ice area;

[0037] And the expression of the latest collision avoidance distance dist vo is

[0038]

[0039] In the formula: dist1, dist2,..., dist (n-1) , dist n represents the shortest distance from the current position of the logical virtual ship to each side of the polygonal obstacle; dist i , i = 1...n represents the collective term of the shortest distances from the current position of the logical virtual ship to each side of the polygonal obstacle; min{·} represents the minimum value; α represents the connection line between the current position of the logical virtual ship and the ith vertex o i , i = 1...n of the polygonal obstacle, and the angle between the side o i o i+1 , i = 1...n of the polygonal obstacle; β represents the connection line between the current position of the logical virtual ship and the (i + 1)th vertex o i+1 , i = 1...n of the polygonal obstacle, and the angle between the side o i o i+1, the included angle between i = 1...n; ∧ represents the logical symbol "AND"; ∨ represents the logical symbol "OR"; d i(i+1) represents the length of the i-th side of the polygonal obstacle; d vi represents the distance from the current position of the logical virtual ship to the i-th vertex of the polygonal obstacle; d v(i+1) represents the distance from the current position of the logical virtual ship to the (i + 1)-th vertex of the polygonal obstacle;

[0040] S33: Obtain the azimuth angle ψ of the line connecting the current position of the logical virtual ship and the i-th vertex of the polygonal obstacle oi , i = 1...n, to obtain the maximum azimuth angle and the minimum azimuth angle of the logical virtual ship and the polygonal obstacle, and its expression is

[0041]

[0042] In the formula: x oi represents the abscissa of the i-th obstacle of the polygonal obstacle in the geodetic coordinate system; y oi represents the ordinate of the i-th obstacle of the polygonal obstacle in the geodetic coordinate system; ψ oimax represents the maximum azimuth angle of the line connecting the current position of the virtual ship and each vertex of the polygonal obstacle; ψ oimin represents the minimum azimuth angle of the line connecting the current position of the virtual ship and each vertex of the polygonal obstacle; max{·} represents the maximum value;

[0043] S34: Redefine the vertex coordinates of the polygonal obstacle corresponding to the maximum azimuth angle ψ oimax and the minimum azimuth angle ψ oimin as the coordinate points o rmax (x ormax , y ormax ) and the coordinate point o rmin (x ormin , y ormin );

[0044] And according to the coordinate point o rmax (x ormax , y ormax ) and the coordinate point o rmin (x ormin , y ormin ), set an obstacle avoidance circular area with a radius of l set to construct the maximum obstacle avoidance bow angle ψ ormax and the minimum obstacle avoidance bow angle ψ ormin of the logical virtual ship and the polygonal obstacle, and its expression is

[0045]

[0046] In the formula: arcsin(·) represents the arcsine function;

[0047] S35: Based on the virtual bow angle, according to the maximum obstacle avoidance bow angle ψ ormax and the minimum obstacle avoidance bow angle ψ ormin , construct and obtain the bow angle guidance switching model for the ship with a sparse sea ice anti-collision mechanism;

[0048] The bow angle guidance switching model is specifically

[0049] If the latest collision avoidance distance dist vo ≤l set and ψ ormin ≤ψ d ≤ψ ormax at this time,

[0050] The expression of the bow angle guidance switching model is

[0051]

[0052] In the formula: Δψ max represents the angle between the logical virtual bow angle and the maximum obstacle avoidance bow angle ψ ormax ; Δψ min represents the angle between the logical virtual bow angle and the minimum obstacle avoidance bow angle ψ ormin ;

[0053] Otherwise, the guidance bow angle of the bow angle guidance switching model is the virtual bow angle.

[0054] Furthermore, the constructed virtual control law of the ship, its expression is

[0055]

[0056] In the formula: z e,k represents the position tracking error variable; ψ e,k represents the bow angle tracking error; α u,k represents the virtual control law corresponding to the position tracking error; α r,k represents the virtual control law corresponding to the bow angle tracking error; [[ID=6L]] represents the positive design parameter used to stabilize the position tracking error; represents the positive design parameter used to stabilize the bow angle tracking error; δ xy represents a parameter designed according to engineering requirements to ensure that the actual ship always follows behind the virtual ship.

[0057] Furthermore, the said S5 specifically includes the following steps:

[0058] S51: Introduce a first-order filter to filter the virtual control law of the ship, and its expression is It should be noted that there is an error in the tag "ID=6L" in the original text, which should be "ID=61". This translation has been processed according to the correct content.

[0059]

[0060] wherein: represents a first-order filter corresponding to the position tracking error; represents a first-order filter corresponding to the heading angle tracking error; represents a positive design parameter in the first-order filter corresponding to the position tracking error; represents a positive design parameter in the first-order filter corresponding to the heading angle tracking error; β u,k (0) represents the initial value of the first-order filter corresponding to the position tracking error; β r,k (0) represents the initial value of the first-order filter corresponding to the heading angle tracking error; α u,k (0) represents the initial value of the virtual control law corresponding to the position tracking error; α r,k (0) represents the initial value of the virtual control law corresponding to the heading angle tracking error;

[0061] S52: Based on step S51 and combined with the non-linear iterative mathematical model, obtain the ship dynamics error;

[0062] The expression of the ship dynamics error is

[0063]

[0064] wherein: u e,k represents the surge acceleration dynamics error; r e,k represents the yaw angular velocity dynamics error;

[0065] And based on the ship dynamics error and formula (1), the differential term of the ship kinematics error can be obtained, and its expression is

[0066]

[0067] wherein: σ u represents the gain in the differential term of the surge acceleration error and σ u =-m v ; σ r represents the gain in the differential term of the yaw angular velocity error and σ r =-(m u -m v ); ζ r (·) represents the variable product in the differential term of the surge acceleration error and ζ r (·)=v k r k ; ζ u (·) represents the variable product in the differential term of the yaw angular velocity error and ζ u (·)=u k vk ; N represents the actual ship main engine control input and N = |n|n;

[0068] S53: Approximate the uncertain nonlinear terms and unknown ocean environmental disturbances in the differential term of the ship kinematic error, and its expression is

[0069]

[0070] In the formula: θ u represents the upper bound function gain term for approximating the uncertain nonlinear term corresponding to the surge acceleration; ρ u,k (v) represents the upper bound function for approximating the uncertain nonlinear term corresponding to the surge acceleration; represents the ocean environmental disturbance d corresponding to the surge acceleration wu (v) upper bound; θ r represents the upper bound function gain term for approximating the uncertain nonlinear term corresponding to the yaw angular acceleration; ρ r,k (v) represents the upper bound function for approximating the uncertain nonlinear term corresponding to the yaw angular acceleration; represents the ocean environmental disturbance d corresponding to the yaw angular acceleration wr (v) upper bound;

[0071] S54: According to formulas (17) and (18), construct the hyperbolic tangent function split term for controller design, and its expression is

[0072]

[0073] In the formula: Ψ u,k represents the function split term corresponding to the surge acceleration and Ψ r,k represents the function split term corresponding to the yaw angular velocity and τ u represents a positive arbitrarily small quantity corresponding to the surge acceleration; τ r represents a positive arbitrarily small quantity corresponding to the yaw angular velocity; tanh(·) represents the hyperbolic tangent function;

[0074] S55: Based on formula (19), design the candidate Lyapunov function of the ship dynamics error as to obtain the differential term of the candidate Lyapunov function as

[0075]

[0076] In the formula: represents V u,k , V r,k 's first derivative; η u,k represents the term to be processed in the differential of the candidate Lyapunov function corresponding to the surge acceleration, ηr,k Denote the item to be processed in the differential of the candidate Lyapunov function corresponding to the yaw angular velocity. Using the Young's inequality theory and introducing the radial basis neural network technology, process the items η u,k and η r,k , and its expression is

[0077]

[0078] In the formula: Denote the upper bound of the gain in the differential term of the surge acceleration error; Denote the upper bound of the gain in the differential term of the yaw angular velocity error; Denote the upper bound of the variable product in the differential term of the surge acceleration error and Denote the upper bound of the variable product in the differential term of the yaw angular velocity error and Denote the ideal weight of the radial basis neural network corresponding to the surge acceleration; Denote the transpose of the ideal weight of the radial basis neural network corresponding to the surge acceleration; Denote the ideal weight of the radial basis neural network corresponding to the yaw angular velocity; Denote the transpose of the ideal weight of the radial basis neural network corresponding to the yaw angular velocity; Φ u (u e,k ) Denote the Gaussian function term of the radial basis neural network corresponding to the surge acceleration; Φ r (r e,k ) Denote the Gaussian function term of the radial basis neural network corresponding to the yaw angular velocity; Is the upper bound of the approximation error of the radial basis neural network corresponding to the surge acceleration; Is the upper bound of the approximation error of the radial basis neural network corresponding to the yaw angular velocity;

[0079] S53: Construct a ship iterative learning controller for ensuring the stability of the differential term of the candidate Lyapunov function according to the ship dynamics error;

[0080] And the expression of the ship iterative learning controller is

[0081]

[0082] In the formula: Denote the gain adaptive estimation term corresponding to the surge acceleration in the k-th iteration; Denote the gain adaptive estimation term corresponding to the surge acceleration in the (k - 1)-th iteration; α N,k Denote the control law corresponding to the surge acceleration in the k-th iteration; Denote the estimation term of the control law corresponding to the surge acceleration in the k-th iteration; Denote the estimated term of the control law corresponding to the surge acceleration at the (k - 1)-th iteration; Denote the intermediate control law corresponding to the surge acceleration; θ u Denote the positive design parameter in the adaptation law corresponding to the surge acceleration; Denote the positive design parameter in the control law corresponding to the surge acceleration; Denote the positive design parameter in the intermediate control law corresponding to the surge acceleration; Denote the estimated term of the ideal weight of the radial basis neural network corresponding to the surge acceleration; γ u Denote the positive design parameter in the estimated term of the ideal weight corresponding to the surge acceleration; g u Denote the positive design parameter related to the initial value of the estimated term of the ideal weight corresponding to the surge acceleration; Denote the initial value of the estimated term of the ideal weight corresponding to the surge acceleration; Denote the gain adaptation estimation term corresponding to the yaw angular velocity at the k-th iteration; Denote the gain adaptation estimation term corresponding to the yaw angular velocity at the (k - 1)-th iteration; α δ,k Denote the control law corresponding to the yaw angular velocity at the k-th iteration; Denote the estimated term of the control law corresponding to the yaw angular velocity at the k-th iteration; Denote the estimated term of the control law corresponding to the yaw angular velocity at the (k - 1)-th iteration; Denote the intermediate control law corresponding to the yaw angular velocity; θ r Denote the positive design parameter in the adaptation law corresponding to the yaw angular velocity; Denote the positive design parameter in the control law corresponding to the yaw angular velocity; Denote the positive design parameter in the intermediate control law corresponding to the yaw angular velocity; Denote the estimated term of the ideal weight of the radial basis neural network corresponding to the yaw angular velocity; γ r Denote the positive design parameter in the estimated term of the ideal weight corresponding to the yaw angular velocity; g r Denote the positive design parameter related to the initial value of the estimated term of the ideal weight corresponding to the yaw angular velocity; Denote the initial value of the estimated term of the ideal weight corresponding to the yaw angular velocity; Denote The first derivative of;

[0083] S54: According to the ship iterative learning controller, conduct obstacle avoidance guidance and control for the ship with a sparse sea ice anti-collision mechanism.

[0084] The present invention provides a method for efficient obstacle avoidance guidance and control of a ship with a sparse sea ice anti-collision mechanism, and the beneficial effects are as follows:

[0085] (1) The present invention constructs a heading angle guidance switching model with a sparse sea ice anti-collision mechanism based on a logical virtual ship, so as to standardize the obstacle avoidance time of the ship and sea ice obstacles in the form of the latest obstacle avoidance distance, and solves the unnecessary maneuvering problem caused by the traditional obstacle avoidance technology that plans the obstacle as a circular area.

[0086] (2) Considering the high-precision requirements of polar ship path tracking control, the present invention obtains the virtual control law of the ship based on the constructed nonlinear iterative mathematical model of the three degrees of freedom of the ship, and filters the virtual control law of the ship through the introduction of a first-order filter to obtain the ship dynamics error, so as to construct a ship iterative learning controller for the nonlinear iterative mathematical model. The ship iterative learning controller can improve the tracking accuracy by accumulating control experience in previous iterations. In addition, in the design process of the ship iterative learning controller, aiming at the problem of uncertain nonlinear terms in the nonlinear iterative mathematical model, the present invention constructs a hyperbolic tangent function splitting term, combines the robust neural damping technology, and ensures that the radial basis neural network only estimates the hyperbolic tangent function term with convergence properties, so as to obtain better nonlinear approximation performance, solve the singularity problem caused by the unknown upper bound function, and further effectively improve the accuracy and efficiency of obstacle avoidance guidance and control of the ship with a sparse sea ice anti-collision mechanism. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0088] Figure 1 It is a flowchart of an efficient obstacle avoidance guidance and control method for a ship with a sparse sea ice anti-collision mechanism according to the present invention;

[0089] Figure 2 It is a schematic diagram of ice floe obstacle avoidance guidance based on the velocity obstacle method in this embodiment;

[0090] Figure 3 It is a schematic diagram of the calculation principle of the latest collision avoidance distance from the virtual ship to the polygonal obstacle in this embodiment;

[0091] Figure 4 It is an effect diagram of obstacle avoidance guidance and path tracking control in this embodiment;

[0092] Figure 5 It is a simulation diagram of the tracking error results under the action of two algorithms in this embodiment;

[0093] Figure 6Simulation diagram of the control input results under the action of two algorithms in this embodiment;

[0094] Figure 7 Simulation diagram of the adaptive estimation results under the action of two algorithms in this embodiment. Specific implementation manner

[0095] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0096] This embodiment provides an efficient obstacle avoidance guidance and control method for ships with a sparse sea ice anti-collision mechanism, as Figure 1 described, specifically including the following steps:

[0097] S1: Construct a non-linear iterative mathematical model of the ship's three degrees of freedom;

[0098] Specifically, in this embodiment, based on the Newton-Lagrange mechanism, a non-linear iterative mathematical model of the ship's three degrees of freedom is constructed, and its model expression is

[0099]

[0100] In the formula: k represents the number of iterations; x k represents the abscissa of the ship's position; y k represents the ordinate of the ship's position; ψ k represents the heading angle of the ship; u k represents the forward speed of the ship; v k represents the transverse drift speed of the ship; r k represents the yaw angular velocity of the ship; m u represents the added hydrodynamic mass of the ship in the forward degree of freedom; m v represents the added hydrodynamic mass of the ship in the transverse drift degree of freedom; m r represents the added hydrodynamic mass of the ship in the yaw degree of freedom; f u (v) represents the non-linear term of the ship in the forward degree of freedom; f v (v) represents the non-linear term of the ship in the transverse drift degree of freedom; f r (v) represents the non-linear term of the ship in the yaw degree of freedom; v represents the transposed matrix of the motion acceleration of the ship in the three degrees of freedom direction and v = [u k , v k , r k T; X |u|u represents the hydrodynamic coefficient related to the surge motion of the ship in the surge degree of freedom; X vr represents the hydrodynamic coefficient related to the sway and yaw motions of the ship in the surge degree of freedom; X vv represents the hydrodynamic coefficient related to the sway motion of the ship in the surge degree of freedom; X rr represents the hydrodynamic coefficient related to the yaw motion of the ship in the surge degree of freedom; Y v represents the hydrodynamic coefficient related to the first power of the sway speed of the ship in the sway degree of freedom; Y |v|v represents the hydrodynamic coefficient related to the second power of the sway speed of the ship in the sway degree of freedom; Y r represents the hydrodynamic coefficient related to the first power of the yaw speed of the ship in the sway degree of freedom; Y |r|r represents the hydrodynamic coefficient related to the second power of the yaw speed of the ship in the sway degree of freedom; Y vvr represents the hydrodynamic coefficient related to the second power of the sway speed and the first power of the yaw speed of the ship in the sway degree of freedom; Y vrr represents the hydrodynamic coefficient related to the second power of the sway speed and the first power of the yaw speed of the ship in the sway degree of freedom; N v represents the hydrodynamic coefficient related to the first power of the sway speed of the ship in the yaw degree of freedom; N r represents the hydrodynamic coefficient related to the first power of the yaw speed of the ship in the yaw degree of freedom; N |v|v represents the hydrodynamic coefficient related to the second power of the side drift speed of the ship in the yaw degree of freedom; N |r|r represents the hydrodynamic coefficient related to the second power of the yaw speed of the ship in the yaw degree of freedom; N vvr represents the hydrodynamic coefficient related to the second power of the sway speed and the first power of the yaw motion speed of the ship in the yaw degree of freedom; N vrr represents the hydrodynamic coefficient related to the first power of the sway speed and the second power of the yaw speed of the ship in the yaw degree of freedom; d wi (v), i = u, v, r are the disturbing forces and moments acting on the ship in the forward, side drift, and yaw degrees of freedom due to ocean environmental disturbances; n represents the main engine speed passing through the steering gear servo system; δ represents the rudder angle passing through the steering gear servo system; T u (·) represents the actuator gain corresponding to the main engine speed; F r (·) represents the actuator gain corresponding to the rudder angle; |·| represents the absolute value of the variable; represents x k , y k , ψ k , u k , v k , r k the first derivative of.

[0101] S2: Obtain a reference position signal according to the set logical virtual ship, and obtain a position tracking error according to the non-linear iterative mathematical model to obtain a reference heading angle signal;

[0102] Specifically, it includes the following steps:

[0103] S21: Obtain a reference position signal according to the set logical virtual ship VS;

[0104] And the expression of the reference position signal is

[0105]

[0106] In the formula: x d represents the longitudinal position coordinate of VS; y d represents the lateral position coordinate of VS; ψ d represents the heading angle of VS; u d represents the speed of VS in the forward degree of freedom; represents x d , y d the first derivative of;

[0107] S22: Based on the reference position signal, obtain a position tracking error according to the non-linear iterative mathematical model to obtain a reference heading angle signal;

[0108] And the expression of the reference heading angle signal is

[0109]

[0110] In the formula: x e,k represents the lateral position tracking error; y e,k represents the longitudinal position tracking error; sign(·) represents the sign function; arctan(·) represents the arctangent function; ψ r,k represents the reference heading angle signal;

[0111] S3: According to the logical virtual ship, construct a heading angle guidance switching model with a sparse sea ice anti-collision mechanism for the non-linear iterative mathematical model to achieve iceberg collision avoidance;

[0112] Specifically, it includes the following steps:

[0113] S31: In this embodiment, by setting a dangerous radius threshold R d , and defining the shortest distance from the virtual ship to the irregular iceberg as the latest collision avoidance distance dist vo , and based on navigation experience, set a navigation target point in the sea area with sparse sea ice, and represent the virtual ship's heading angle of the logical virtual ship as;

[0114]

[0115] Where: ψ d represents the heading angle of VS; x t , y t represent the abscissa and ordinate of the position of the navigation target point;

[0116] S32: Based on the set navigation target point, according to the logical virtual ship, construct the latest collision avoidance distance dist between the ship and the sea ice floes in the sparse sea ice area vo , such as Figure 3 shown, specifically including:

[0117] S321: In the case of obtaining high-precision ice condition information through technologies such as satellite remote sensing or UAV imaging, that is, obtaining the irregular sea ice floes in the known sparse sea ice area, and defining the outer contour of the irregular sea ice floes along the clockwise direction as a polygon obstacle with n vertices;

[0118] Among them, the vertex coordinates of the polygon obstacle are defined as o1(x o1 , y o1 ), o2(x o2 , y o2 ),..., o n (x on , y on ); o<^ n (x on , y on ) represents the coordinate position of the nth vertex;

[0119] S322: Express the relative distance between the current position of the logical virtual ship and each vertex coordinate in the clockwise order of the polygon obstacle as d vi , i = 1... n; Express the side length of each side of the polygon obstacle as d i(i+1) , i = 1... n, and its expression is

[0120]

[0121] Where: x oi represents the abscissa of the ith vertex of the polygon obstacle; y oi represents the ordinate of the ith vertex of the polygon obstacle; x o(i+1) represents the abscissa of the (i + 1)th vertex of the polygon obstacle; y o(i+1) represents the ordinate of the (i + 1)th vertex of the polygon obstacle;

[0122] S323: According to the relative distance d vi and the side length d i(i+1) of the polygon obstacle, obtain the latest collision avoidance distance between the ship and the sea ice floes in the sparse sea ice area;

[0123] and the latest collision avoidance distance dist vo The expression is

[0124]

[0125] In the formula: dist1, dist2, …, dist (n-1) , dist n represent the shortest distances from the current position of the logical virtual ship to each side of the polygonal obstacle; dist i , i = 1...n represents the collective term of the shortest distances from the current position of the logical virtual ship to each side of the polygonal obstacle; min{·} represents the minimum value; α represents the connection line between the current position of the logical virtual ship and the i-th vertex o i , i = 1...n of the polygonal obstacle, and the angle between the connection line and the side o i o i+1 , i = 1...n of the polygonal obstacle; β represents the connection line between the current position of the logical virtual ship and the (i + 1)-th vertex o i+1 , i = 1...n of the polygonal obstacle, and the angle between the connection line and the side o i o i+1 , i = 1...n of the polygonal obstacle; ∧ represents the logical symbol "and"; ∨ represents the logical symbol "or"; d i(i+1) represents the length of the i-th side of the polygonal obstacle; d vi represents the distance from the current position of the logical virtual ship to the i-th vertex of the polygonal obstacle; d v(i+1) represents the distance from the current position of the logical virtual ship to the (i + 1)-th vertex of the polygonal obstacle;

[0126] S33: Obtain the azimuth angle ψ oi , i = 1...n of the connection line between the current position of the logical virtual ship and the i-th vertex of the polygonal obstacle, so as to obtain the maximum azimuth angle and the minimum azimuth angle between the logical virtual ship and the polygonal obstacle, and its expression is

[0127]

[0128] In the formula: x oi represents the abscissa of the i-th obstacle of the polygonal obstacle in the geodetic coordinate system; y oi represents the ordinate of the i-th obstacle of the polygonal obstacle in the geodetic coordinate system; ψ oimax represents the maximum azimuth angle of the connection line between the current position of the virtual ship and each vertex of the polygonal obstacle; ψ oimin represents the minimum azimuth angle of the connection line between the current position of the virtual ship and each vertex of the polygonal obstacle; max{·} represents the maximum value;

[0129] S34: Compare the maximum azimuth angle ψ oimax with the minimum azimuth angle ψoimin The vertex coordinates of the corresponding polygonal obstacle are redefined as the coordinate point o rmax (x ormax , y ormax ) and the coordinate point o rmin (x ormin , y ormin );

[0130] And based on the coordinate point o rmax (x ormax , y ormax ) and the coordinate point o rmin (x ormin , y ormin ), a collision avoidance circular area with a radius of l set is set to construct the maximum collision avoidance bow angle ψ ormax and the minimum collision avoidance bow angle ψ ormin of the logical virtual ship and the polygonal obstacle, and their expression is

[0131]

[0132] In the formula: arcsin(·) represents the arcsine function;

[0133] S35: Based on the virtual ship bow angle, according to the maximum collision avoidance bow angle ψ ormax and the minimum collision avoidance bow angle ψ ormin , construct and obtain the bow angle guidance switching model for the ship with a sparse sea ice anti-collision mechanism;

[0134] The specific form of the bow angle guidance switching model is

[0135] If the latest collision avoidance distance dist vo ≤ l set and ψ ormin ≤ ψ d ≤ ψ ormax at this time,

[0136] The expression of the bow angle guidance switching model is

[0137]

[0138] In the formula: Δψ max represents the angle between the virtual ship bow angle and the maximum collision avoidance bow angle ψ ormax ; Δψ min represents the angle between the virtual ship bow angle and the minimum collision avoidance bow angle ψ ormin ;

[0139] Otherwise, the guidance bow angle of the bow angle guidance switching model is the virtual ship bow angle;

[0140] In this embodiment, in order to make the actual ship path converge to the desired path with high precision, the present invention designs an adaptive iterative learning controller. The specific controller design process is as follows:

[0141] S4: Based on the heading angle guidance switching model, obtain the heading angle tracking error according to the reference heading angle signal, and based on the heading angle tracking error and the position tracking error. And in order to stabilize the heading angle tracking error and the position tracking error in Equation (12), design and construct the virtual control law of the ship based on the Back steeping method;

[0142] And the expression of the virtual control law of the ship is

[0143]

[0144] where: z e,k represents the position tracking error variable; ψ e,k represents the heading angle tracking error; α u,k represents the virtual control law corresponding to the position tracking error; α r,k represents the virtual control law corresponding to the heading angle tracking error; represents the positive design parameter used to stabilize the position tracking error; represents the positive design parameter used to stabilize the heading angle tracking error; δ xy represents a parameter designed according to engineering requirements to ensure that the actual ship always follows behind the virtual ship;

[0145] S5: In this embodiment, in order to avoid adding unnecessary computational load during the differentiation process of the virtual control laws α u,k and α r,k filter the virtual control law of the ship by introducing a first-order filter to obtain the ship dynamics error; construct a ship iterative learning controller for the nonlinear iterative mathematical model according to the ship dynamics error, and perform obstacle avoidance guidance and control with a sparse sea ice anti-collision mechanism for the ship according to the ship iterative learning controller;

[0146] Specifically, it includes the following steps:

[0147] S51: Introduce a first-order filter to filter the virtual control law of the ship, and its expression is

[0148]

[0149] where: represents the first-order filter corresponding to the position tracking error; represents the first-order filter corresponding to the heading angle tracking error; represents the positive design parameter in the first-order filter corresponding to the position tracking error; Denote the positive design parameter in the first-order filter corresponding to the heading angle tracking error; β u,k (0) Denote the initial value of the first-order filter corresponding to the position tracking error; β r,k (0) Denote the initial value of the first-order filter corresponding to the heading angle tracking error; α u,k (0) Denote the initial value of the virtual control law corresponding to the position tracking error; α r,k (0) Denote the initial value of the virtual control law corresponding to the heading angle tracking error;

[0150] S52: Based on Step S51 and combined with the nonlinear iterative mathematical model, obtain the ship dynamics error;

[0151] The expression of the ship dynamics error is

[0152]

[0153] In the formula: u e,k Denote the surge acceleration dynamics error; r e,k Denote the yaw angular velocity dynamics error;

[0154] And based on the ship dynamics error and Equation (1), the differential term of the ship kinematics error can be obtained, and its expression is

[0155]

[0156] In the formula: σ u Denote the gain in the differential term of the surge acceleration error and σ u =-m v ; σ r Denote the gain in the differential term of the yaw angular velocity error and σ r =-(m u -m v ); ζ r (·) Denote the variable product in the differential term of the surge acceleration error and ζ r (·)=v k r k ; ζ u (·) Denote the variable product in the differential term of the yaw angular velocity error and ζ u (·)=u k v k ; N denote the actual ship main engine control input and N = |n|n;

[0157] S53: For Equation (16), use the assumed upper bound function and Young's inequality to handle the uncertain nonlinear terms and unknown ocean environmental disturbances, and then approximate the uncertain nonlinear terms and unknown ocean environmental disturbances in the differential term of the ship kinematics error. Its expression is

[0158]

[0159] where: θ u represents the upper - bound function gain term for approximating the uncertain nonlinear term corresponding to the surge acceleration; ρ u,k (v) represents the upper - bound function for approximating the uncertain nonlinear term corresponding to the surge acceleration; represents the ocean environmental disturbance d corresponding to the surge acceleration wu (v)'s upper - bound; θ r represents the upper - bound function gain term for approximating the uncertain nonlinear term corresponding to the yaw angular acceleration; ρ r,k (v) represents the upper - bound function for approximating the uncertain nonlinear term corresponding to the yaw angular acceleration; represents the ocean environmental disturbance d corresponding to the yaw angular acceleration wr (v)'s upper - bound;

[0160] S54: Since the unknown upper - bound function cannot be directly used to design the controller, according to formulas (17) and (18), a hyperbolic tangent function split term for controller design is constructed to solve the above - mentioned problem, and its expression is

[0161]

[0162] where: Ψ u,k represents the function split term corresponding to the surge acceleration and Ψ r,k represents the function split term corresponding to the yaw angular velocity and τ u represents a positive arbitrary small quantity corresponding to the surge acceleration; τ r represents a positive arbitrary small quantity corresponding to the yaw angular velocity; tanh(·) represents the hyperbolic tangent function;

[0163] S55: Based on formula (19), the candidate Lyapunov function for the ship dynamics error is designed as to obtain the differential term of the candidate Lyapunov function as

[0164]

[0165] where: represents the first - order derivative of V u,k ,V r,k η u,k represents the term to be processed in the differential of the candidate Lyapunov function corresponding to the surge acceleration, η r,k represents the term to be processed in the differential of the candidate Lyapunov function corresponding to the yaw angular velocity and using the Young's inequality theory, and at the same time introducing the radial basis neural network technology, for the terms to be processed η u,k and η r,kProcessed with the expression

[0166]

[0167] Where: Represents the upper bound of the gain in the differential term of the surge acceleration error; Represents the upper bound of the gain in the differential term of the yaw angular velocity error; Represents the upper bound of the product of variables in the differential term of the surge acceleration error and Represents the upper bound of the product of variables in the differential term of the yaw angular velocity error and Represents the ideal weight of the radial basis neural network corresponding to the surge acceleration; Represents the transpose of the ideal weight of the radial basis neural network corresponding to the surge acceleration; Represents the ideal weight of the radial basis neural network corresponding to the yaw angular velocity; Represents the transpose of the ideal weight of the radial basis neural network corresponding to the yaw angular velocity; Φ u (u e,k ) represents the Gaussian function term of the radial basis neural network corresponding to the surge acceleration; Φ r (r e,k ) represents the Gaussian function term of the radial basis neural network corresponding to the yaw angular velocity; Is the upper bound of the approximation error of the radial basis neural network corresponding to the surge acceleration; Is the upper bound of the approximation error of the radial basis neural network corresponding to the yaw angular velocity;

[0168] S53: Construct a ship iterative learning controller for ensuring the stability of the differential term of the candidate Lyapunov function according to the ship dynamics error;

[0169] And the expression of the ship iterative learning controller is

[0170]

[0171] Where: Represents the gain adaptive estimation term corresponding to the surge acceleration at the k-th iteration; Represents the gain adaptive estimation term corresponding to the surge acceleration at the (k - 1)-th iteration; α N,k Represents the control law corresponding to the surge acceleration at the k-th iteration; Represents the estimated term of the control law corresponding to the surge acceleration at the k-th iteration; Represents the estimated term of the control law corresponding to the surge acceleration at the (k - 1)-th iteration; Represents the intermediate control law corresponding to the surge acceleration; θ u Represents the positive design parameter in the adaptive law corresponding to the surge acceleration; represents the positive design parameter in the control law corresponding to the surge acceleration; represents the positive design parameter in the intermediate control law corresponding to the surge acceleration; represents the estimated term of the ideal weight of the radial basis neural network corresponding to the surge acceleration; γ u represents the positive design parameter in the estimated term of the ideal weight corresponding to the surge acceleration; g u represents the positive design parameter related to the initial value of the estimated term of the ideal weight corresponding to the surge acceleration; represents the initial value of the estimated term of the ideal weight corresponding to the surge acceleration; represents the gain adaptive estimation term corresponding to the yaw angular velocity at the k-th iteration; represents the gain adaptive estimation term corresponding to the yaw angular velocity at the (k - 1)-th iteration; α δ,k represents the control law corresponding to the yaw angular velocity at the k-th iteration; represents the estimated term of the control law corresponding to the yaw angular velocity at the k-th iteration; represents the estimated term of the control law corresponding to the yaw angular velocity at the (k - 1)-th iteration; represents the intermediate control law corresponding to the yaw angular velocity; θ r represents the positive design parameter in the adaptive law corresponding to the yaw angular velocity; represents the positive design parameter in the control law corresponding to the yaw angular velocity; represents the positive design parameter in the intermediate control law corresponding to the yaw angular velocity; represents the estimated term of the ideal weight of the radial basis neural network corresponding to the yaw angular velocity; γ r represents the positive design parameter in the estimated term of the ideal weight corresponding to the yaw angular velocity; g r represents the positive design parameter related to the initial value of the estimated term of the ideal weight corresponding to the yaw angular velocity; represents the initial value of the estimated term of the ideal weight corresponding to the yaw angular velocity; ... represents the first derivative of

[0172] S54: According to the ship iterative learning controller, conduct obstacle avoidance guidance and control for the ship with a sparse sea ice anti-collision mechanism.

[0173] To verify the effectiveness of the method described in this embodiment, a simulation numerical experiment is carried out in the polar environment. 40 irregular random floating ice obstacles are generated in a sea area of 3000m×3000m. A container ship with a length of 175m, a width of 25.4m, and an average draft of 8m is selected as the controlled object for the experiment. Set the target point (x t , y t ) = (3000m×3000m), and the obstacle avoidance radius threshold R d= 290 m, the radius l of the circular obstacle avoidance area set = 50 m. Set the initial value of the virtual ship state variable to x d (0) = -100 m, y d (0) = -100 m, ψ d (0) = 90 deg, u d (0) = 5 m / s; the initial value of the real ship's state variable is x k (0) = -300 m, y k (0) = -300 m, ψ k (0) = 90 deg, φ k (0) = 0 deg, u k (0) = 5 m / s, v k (0) = 0 m / s, r k (0) = 0 deg / s, p k (0) = 0 deg / s, n k (0) = 35 rpm, δ k (0) = 0 deg. Select the controller design parameters as θ u = 0.080, θ r = 0.200, γ u = 0.060, γ r = 0.800, g u = 0.080, g r = 0.600.

[0174] As Figure 4 shown, the obstacle avoidance results and path tracking control effects under the proposed strategy are presented. The pink solid line represents the reference path planned by the proposed obstacle avoidance algorithm, and the yellow solid line represents the real ship trajectory under the action of the proposed iterative learning control algorithm. It can be seen from the figure that the virtual ship can achieve refined obstacle avoidance path planning, and the velocity obstacle avoidance algorithm constructed for polygon obstacles has been effectively verified, effectively alleviating the non-necessary maneuvering problem. In addition, the real ship trajectory converges to the desired path with high accuracy. Although there are lag problems caused by the large inertia of the polar ship itself, they are all within the allowable range of marine engineering. Figure 5 shows the tracking error results under the proposed adaptive iterative learning control algorithm and the traditional iterative learning control algorithm. It can be seen from the figure that as the number of iterations increases, the tracking error has an obvious decreasing trend. In addition, the control algorithm proposed in the present invention has a faster convergence rate compared with the traditional iterative learning control algorithm. Figure 6 shows the actual control input results. It can be seen from the figure that as the number of iterations increases, the tracking accuracy gradually improves, and the actual control input will also gradually increase as the tracking error decreases.Figure 7 The adaptive law curve is shown. As can be seen from the figure, as the number of iterations increases, the gain estimation value will gradually become more accurate.

[0175] By analyzing the existing technical means and combining the simulation test results, it can be known that the obstacle avoidance guidance and iterative learning control scheme with a sparse sea ice anti-collision mechanism proposed in this embodiment, that is, the efficient obstacle avoidance guidance and control method for ships with a sparse sea ice anti-collision mechanism has the following two obvious advantages:

[0176] (1) By constructing a heading angle guidance switching model with a sparse sea ice anti-collision mechanism according to the logical virtual ship, the present invention standardizes the obstacle avoidance time between the ship and the sea ice obstacle in the way of the latest obstacle avoidance distance, and solves the unnecessary maneuver problem caused by the traditional obstacle avoidance technology that plans the obstacle as a circular area.

[0177] (2) Considering the high-precision requirements of polar ship path tracking control, the present invention obtains the ship virtual control law based on the constructed three-degree-of-freedom nonlinear iterative mathematical model of the ship, and filters the ship virtual control law through the introduction of a first-order filter to obtain the ship dynamics error, so as to construct a ship iterative learning controller for the nonlinear iterative mathematical model. The ship iterative learning controller can improve the tracking accuracy by accumulating the control experience in the previous iterations. In addition, in the design process of the ship iterative learning controller, aiming at the problem of uncertain nonlinear terms in the nonlinear iterative mathematical model, the present invention constructs a hyperbolic tangent function splitting term, combines the robust neural damping technology, and ensures that the radial basis neural network only estimates the hyperbolic tangent function term with convergence properties, so as to obtain better nonlinear approximation performance, solve the singularity problem caused by the unknown upper bound function, and further effectively improve the accuracy and efficiency of the obstacle avoidance guidance and control of the ship with a sparse sea ice anti-collision mechanism.

[0178] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or equivalently replace some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for efficient obstacle avoidance guidance and control of a ship with a sparse sea ice anti-collision mechanism, characterized in that, Specifically, it includes the following steps: S1: Construct a non-linear iterative mathematical model of the ship's three degrees of freedom; S2: Obtain the reference position signal according to the set logical virtual ship, and obtain the position tracking error according to the non-linear iterative mathematical model to obtain the reference heading angle signal; S3: Based on the logical virtual ship, construct a heading angle guidance switching model with a sparse sea ice anti-collision mechanism; S4: Based on the heading angle guidance switching model, obtain the heading angle tracking error according to the reference heading angle signal, and construct the ship virtual control law according to the heading angle tracking error and the position tracking error; S5: Introduce a first-order filter to filter the ship virtual control law to obtain the ship dynamics error; construct a ship iterative learning controller for the non-linear iterative mathematical model according to the ship dynamics error, and perform obstacle avoidance guidance and control with a sparse sea ice anti-collision mechanism on the ship according to the ship iterative learning controller.

2. An efficient obstacle avoidance guidance and control method for ships with a sparse sea ice anti-collision mechanism according to claim 1, characterized in that, The non-linear iterative mathematical model of the ship's three degrees of freedom constructed in S1, its model expression is where: k represents the number of iterations; x k represents the abscissa of the ship's position; y k represents the ordinate of the ship's position; ψ k represents the heading angle of the ship; u k represents the forward speed of the ship; v k represents the transverse drift speed of the ship; r k represents the yaw angular velocity of the ship; m u represents the added hydrodynamic mass of the ship in the forward degree of freedom; m v represents the added hydrodynamic mass of the ship in the transverse drift degree of freedom; m r represents the added hydrodynamic mass of the ship in the yaw degree of freedom; f u (v) represents the non - linear term of the ship in the forward degree of freedom; f v (v) represents the non - linear term of the ship in the transverse drift degree of freedom; f r (v) represents the non - linear term of the ship in the yaw degree of freedom; v represents the transpose matrix of the ship's three - degree - of - freedom motion acceleration and v = [u k ,v k ,r k T ; X |u|u represents the hydrodynamic coefficient related to the ship's surge motion in the surge degree of freedom; X vr represents the hydrodynamic coefficient related to the ship's sway and yaw motions in the surge degree of freedom; X vv represents the hydrodynamic coefficient related to the ship's sway motion in the surge degree of freedom; X rr represents the hydrodynamic coefficient related to the ship's yaw motion in the surge degree of freedom; Y v represents the hydrodynamic coefficient related to the first power of the ship's sway speed in the sway degree of freedom; Y |v|v represents the hydrodynamic coefficient related to the second power of the ship's sway speed in the sway degree of freedom; Y r represents the hydrodynamic coefficient related to the first power of the ship's yaw speed in the sway degree of freedom; Y |r|r represents the hydrodynamic coefficient related to the second power of the ship's yaw speed in the sway degree of freedom; Y vvr represents the hydrodynamic coefficient related to the second power of the ship's sway speed and the first power of the yaw speed in the sway degree of freedom; Y vrr represents the hydrodynamic coefficient related to the second power of the ship's sway speed and the first power of the yaw speed in the sway degree of freedom; N v represents the hydrodynamic coefficient related to the first power of the ship's sway speed in the yaw degree of freedom; N r represents the hydrodynamic coefficient related to the first power of the ship's yaw speed in the yaw degree of freedom; N |v|v represents the hydrodynamic coefficient related to the second power of the ship's transverse drift speed in the yaw degree of freedom; N​ |r|r Denotes the hydrodynamic coefficient related to the square of the yaw velocity of the ship in the yaw degree of freedom; N vvr Denotes the hydrodynamic coefficient related to the square of the sway velocity of the ship and the first power of the yaw motion velocity in the yaw degree of freedom; N vrr Denotes the hydrodynamic coefficient related to the first power of the sway velocity of the ship and the square of the yaw velocity in the yaw degree of freedom; d wi (v), i = u, v, r are the disturbing forces and moments acting on the ship in the forward, sway, and yaw degrees of freedom due to ocean environmental disturbances; n represents the main engine speed passing through the steering gear servo system; δ represents the rudder angle passing through the steering gear servo system; T u (·) represents the actuator gain corresponding to the main engine speed; F r (·) represents the actuator gain corresponding to the rudder angle; |·| represents the absolute value of a variable; represents x k , y k , ψ k , u k , v k , r k the first derivative of.

3. A method for efficient obstacle avoidance guidance and control of a ship with a sparse sea ice anti-collision mechanism according to claim 2, characterized in that, The specific steps of S2 are as follows: S21: Obtain the reference position signal according to the set logical virtual ship VS; And the expression of the reference position signal is where: x d represents the longitudinal position coordinate of VS; y d represents the transverse position coordinate of VS; ψ d represents the heading angle of VS; u d represents the speed of VS in the forward degree of freedom; represents the first derivative of x d , y d ; S22: Based on the reference position signal, obtain the position tracking error according to the non-linear iterative mathematical model to obtain the reference heading angle signal; And the expression of the reference heading angle signal is where: x e,k represents the lateral position tracking error; y e,k represents the longitudinal position tracking error; sign(·) represents the sign function; arctan(·) represents the arctangent function; ψ r,k represents the reference heading angle signal.

4. A method for efficient obstacle avoidance guidance and control of a ship with a sparse sea ice anti-collision mechanism according to claim 3, characterized in that, The construction method of the heading angle guidance switching model with a sparse sea ice anti-collision mechanism in S3 specifically includes the following steps: S31: Based on navigation experience, set a navigation target point in the sea area with sparse sea ice, and represent the virtual ship's heading angle of the logical virtual ship as; where: ψ d represents the heading angle of VS; x t , y t represent the abscissa and ordinate of the position of the navigation target point; S32: Based on the set navigation target point, construct the latest collision avoidance distance dist between the ship and the sea ice floes in the sparse sea ice area according to the logical virtual ship vo , which specifically includes: S321: Through satellite remote sensing or UAV imaging technology, obtain the irregular sea ice floating on the sea in the known sparse sea ice sea area, and define it as a polygonal obstacle with n vertices along the outer contour of the irregular sea ice in the clockwise direction; Among them, the vertex coordinates of the polygonal obstacle are defined as o1(x o1 , y o1 ), o2(x o2 , y o2 ),..., o n (x on , y on ); o n (x on , y on ) represents the coordinate position of the nth vertex; S322: Represent the relative distance of the current position of the logical virtual ship from each vertex coordinate in the clockwise order with respect to the polygonal obstacle as d vi , where i = 1...n; represent the side length of each side of the polygonal obstacle as d i(i+1) , where i = 1...n, and its expression is Where: x oi represents the abscissa of the i-th vertex of the polygonal obstacle; y oi represents the ordinate of the i-th vertex of the polygonal obstacle; x o(i+1) represents the abscissa of the (i + 1)-th vertex of the polygonal obstacle; y o(i+1) represents the ordinate of the (i + 1)-th vertex of the polygonal obstacle; S323: According to the relative distance d vi and the side length d of the polygonal obstacle i(i+1) , obtain the latest collision avoidance distance between the ship and the sea ice floes in the sparse sea ice area; and the expression of the latest collision avoidance distance dist vo is as follows where: dist1, dist2, …, dist (n-1) , dist n represent the shortest distances from the current position of the logical virtual ship to each side of the polygonal obstacle; dist i , i = 1...n represents the collective term for the shortest distances from the current position of the logical virtual ship to each side of the polygonal obstacle; min{·} represents the minimum value; α represents the angle between the line connecting the current position of the logical virtual ship and the i-th vertex o i , i = 1...n of the polygonal obstacle and the side o i o i+1 , i = 1...n of the polygonal obstacle; β represents the angle between the line connecting the current position of the logical virtual ship and the (i + 1)-th vertex o i+1 , i = 1...n of the polygonal obstacle and the side o i o i+1 , i = 1...n of the polygonal obstacle; ∧ represents the logical symbol "AND"; ∨ represents the logical symbol "OR"; d i(i+1) represents the length of the i-th side of the polygonal obstacle; d vi represents the distance from the current position of the logical virtual ship to the i-th vertex of the polygonal obstacle; d v(i+1) represents the distance from the current position of the logical virtual ship to the (i + 1)-th vertex of the polygonal obstacle; S33: Obtain the azimuth angle ψ of the line connecting the current position of the logical virtual ship and the i-th vertex of the polygonal obstacle oi , where i = 1...n, to obtain the maximum azimuth angle and the minimum azimuth angle between the logical virtual ship and the polygonal obstacle, and its expression is where: x oi represents the abscissa of the i-th obstacle of the polygonal obstacle in the geodetic coordinate system; y oi represents the ordinate of the i-th obstacle of the polygonal obstacle in the geodetic coordinate system; ψ oimax represents the maximum azimuth angle of the line connecting the current position of the virtual ship to each vertex of the polygonal obstacle; ψ oimin represents the minimum azimuth angle of the line connecting the current position of the virtual ship to each vertex of the polygonal obstacle; max{·} represents the maximum value; S34: Redefine the vertex coordinates of the polygonal obstacle corresponding to the maximum azimuth angle ψ oimax and the minimum azimuth angle ψ oimin as the coordinate points o rmax (x ormax , y ormax ) and the coordinate point o rmin (x ormin , y ormin ), respectively; And based on the coordinate point o rmax (x ormax , y ormax ) and the coordinate point o rmin (x ormin , y ormin ), set an obstacle avoidance circular area with a radius of l set to construct the maximum obstacle avoidance bow angle ψ ormax and the minimum obstacle avoidance bow angle ψ ormin , and its expression is In the formula: arcsin(·) represents the arcsine function; S35: Based on the virtual bow angle, construct and obtain a bow angle guidance switching model for the ship with a sparse sea ice anti-collision mechanism according to the maximum anti-collision bow angle ψ ormax and the minimum anti-collision bow angle ψ ormin . The specific heading angle guidance switching model is If the latest collision avoidance distance dist vo ≤l set and ψ ormin ≤ψ d ≤ψ ormax when The expression of the heading angle guidance switching model is where: Δψ max represents the angle between the logical virtual bow angle and the maximum obstacle avoidance bow angle ψ ormax ; Δψ min represents the angle between the logical virtual bow angle and the minimum obstacle avoidance bow angle ψ ormin ; Otherwise, the guided heading angle of the heading angle guidance switching model is the virtual ship's heading angle.

5. The efficient obstacle avoidance guidance and control method for a ship with a sparse sea ice anti-collision mechanism according to claim 4, characterized in that, The ship virtual control law constructed in S4, its expression is where: z e,k represents the position tracking error variable; ψ e,k represents the heading angle tracking error; α u,k represents the virtual control law corresponding to the position tracking error; α r,k represents the virtual control law corresponding to the heading angle tracking error; represents the positive design parameter used to stabilize the position tracking error; represents the positive design parameter used to stabilize the heading angle tracking error; δ xy represents a parameter designed according to engineering requirements to ensure that the actual ship always follows behind the virtual ship.

6. The efficient obstacle avoidance guidance and control method for a ship with a sparse sea ice anti-collision mechanism according to claim 4, characterized in that The specific steps of S5 are as follows: S51: Introduce a first-order filter to filter the ship virtual control law, and its expression is In the formula: represents the first-order filter corresponding to the position tracking error; represents the first-order filter corresponding to the heading angle tracking error; represents the positive design parameter in the first-order filter corresponding to the position tracking error; represents the positive design parameter in the first-order filter corresponding to the heading angle tracking error; β u,k (0) represents the initial value of the first-order filter corresponding to the position tracking error; β r,k (0) represents the initial value of the first-order filter corresponding to the heading angle tracking error; α u,k (0) represents the initial value of the virtual control law corresponding to the position tracking error; α r,k (0) represents the initial value of the virtual control law corresponding to the heading angle tracking error; S52: Based on step S51 and combined with the non-linear iterative mathematical model, obtain the ship dynamics error; The expression of the ship dynamics error is Where: u e,k represents the dynamic error of surge acceleration; r e,k represents the dynamic error of yaw angular velocity; And based on the ship dynamics error and formula (1), the differential term of the ship kinematic error can be obtained, and its expression is Where: σ u represents the gain in the differential term of the surge acceleration error, and σ u = -m v ; σ r represents the gain in the differential term of the yaw angular velocity error, and σ r = -(m u - m v ); ζ r (·) represents the variable product in the differential term of the surge acceleration error, and ζ r (·) = v k r k ; ζ u (·) represents the variable product in the differential term of the yaw angular velocity error, and ζ u (·) = u k v k ; N represents the actual ship main engine control input, and N = |n|n; S53: Approximate the uncertain non-linear term and the unknown ocean environment disturbance in the differential term of the ship kinematic error, and its expression is where: θ u represents the upper bound function gain term for approximating the uncertain nonlinear term corresponding to the surge acceleration; ρ u,k (v) represents the upper bound function for approximating the uncertain nonlinear term corresponding to the surge acceleration; represents the ocean environmental disturbance d corresponding to the surge acceleration wu upper bound of (v); θ r represents the upper bound function gain term for approximating the uncertain nonlinear term corresponding to the yaw acceleration; ρ r,k (v) represents the upper bound function for approximating the uncertain nonlinear term corresponding to the yaw acceleration; represents the ocean environmental disturbance d corresponding to the yaw acceleration wr upper bound of (v); S54: According to formulas (17) and (18), construct a hyperbolic tangent function splitting term for controller design, and its expression is where: Ψ u,k represents the function split term corresponding to the surge acceleration and Ψ r,k represents the function split term corresponding to the yaw angular velocity and τ u represents a positive arbitrary small quantity corresponding to the surge acceleration; τ r represents a positive arbitrary small quantity corresponding to the yaw angular velocity; tanh(·) represents the hyperbolic tangent function; S55: The candidate Lyapunov function for the ship dynamics error designed based on formula (19) is to obtain the differential term of the candidate Lyapunov function as In the formula: represents V u,k , V r,k is the first derivative of; η u,k represents the term to be processed in the differential of the candidate Lyapunov function corresponding to the surge acceleration, η r,k represents the term to be processed in the differential of the candidate Lyapunov function corresponding to the yaw angular velocity. By using the Young's inequality theory and introducing the radial basis neural network technology, the terms to be processed η u,k and η r,k are processed, and their expressions are In the formula: represents the upper bound of the gain in the differential term of the surge acceleration error; represents the upper bound of the gain in the differential term of the yaw angular velocity error; represents the upper bound of the product of variables in the differential term of the surge acceleration error and represents the upper bound of the product of variables in the differential term of the yaw angular velocity error and represents the ideal weight of the radial basis neural network corresponding to the surge acceleration; represents the transpose of the ideal weight of the radial basis neural network corresponding to the surge acceleration; represents the ideal weight of the radial basis neural network corresponding to the yaw angular velocity; represents the transpose of the ideal weight of the radial basis neural network corresponding to the yaw angular velocity; Φ u (u e,k ) represents the Gaussian function term of the radial basis neural network corresponding to the surge acceleration; Φ r (r e,k ) represents the Gaussian function term of the radial basis neural network corresponding to the yaw angular velocity; is the upper bound of the approximation error of the radial basis neural network corresponding to the surge acceleration; is the upper bound of the approximation error of the radial basis neural network corresponding to the yaw angular velocity; S53: According to the ship dynamics error, construct a ship iterative learning controller to ensure the stability of the differential term of the candidate Lyapunov function; And the expression of the ship iterative learning controller is where: represents the gain adaptive estimation term corresponding to the surge acceleration at the k-th iteration; represents the gain adaptive estimation term corresponding to the surge acceleration at the (k - 1)-th iteration; α N,k represents the control law corresponding to the surge acceleration at the k-th iteration; represents the estimated term of the control law corresponding to the surge acceleration at the k-th iteration; represents the estimated term of the control law corresponding to the surge acceleration at the (k - 1)-th iteration; represents the intermediate control law corresponding to the surge acceleration; represents the positive design parameter in the adaptive law corresponding to the surge acceleration; represents the positive design parameter in the control law corresponding to the surge acceleration; represents the positive design parameter in the intermediate control law corresponding to the surge acceleration; represents the estimated term of the ideal weight of the radial basis neural network corresponding to the surge acceleration; γ u represents the positive design parameter in the estimated term of the ideal weight corresponding to the surge acceleration; g u represents the positive design parameter related to the initial value of the estimated term of the ideal weight corresponding to the surge acceleration; represents the initial value of the estimated term of the ideal weight corresponding to the surge acceleration; represents the gain adaptive estimation term corresponding to the yaw angular velocity at the k-th iteration; represents the gain adaptive estimation term corresponding to the yaw angular velocity at the (k - 1)-th iteration; α δ,k represents the control law corresponding to the yaw angular velocity at the k-th iteration; represents the estimated term of the control law corresponding to the yaw angular velocity at the k-th iteration; represents the estimated term of the control law corresponding to the yaw angular velocity at the (k - 1)-th iteration; represents the intermediate control law corresponding to the yaw angular velocity; represents the positive design parameter in the adaptive law corresponding to the yaw angular velocity; represents the positive design parameter in the control law corresponding to the yaw angular velocity; represents the positive design parameter in the intermediate control law corresponding to the yaw angular velocity; represents the estimated term of the ideal weight of the radial basis neural network corresponding to the yaw angular velocity; γ r represents the positive design parameter in the estimated term of the ideal weight corresponding to the yaw angular velocity; g r represents the positive design parameter related to the initial value of the estimated term of the ideal weight corresponding to the yaw angular velocity; represents the initial value of the estimated term of the ideal weight corresponding to the yaw angular velocity; represents the first derivative of S54: According to the ship iterative learning controller, perform obstacle avoidance guidance and control on the ship with a sparse sea ice anti-collision mechanism.

Images