An intelligent ship safety autonomous berthing control system

Abstract

The application discloses a kind of offshore intelligent ship safety autonomous berthing control systems, including kinematic control module, speed optimization module, extended state observer module and dynamics control module;Kinematic control module according to the desired berthing position of intelligent ship and the current position of intelligent ship obtains the desired speed and heading required for intelligent ship to reach target position;Speed optimization module calculates the optimized speed of intelligent ship according to the desired speed required for intelligent ship to reach target position and the distance constraint between obstacle and intelligent ship during navigation;Extended state observer module is used to obtain the disturbance estimation value of the environment suffered by intelligent ship;Dynamics control module is used to obtain the torque value required for intelligent ship according to the optimized speed and the disturbance estimation value suffered by intelligent ship.This method does not need to plan the path in advance and does not need a large amount of data learning, while it can effectively avoid obstacles to ensure the safety and reliability of autonomous berthing.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of offshore intelligent ship control, and particularly relates to a safe and autonomous berthing control system for offshore intelligent ships. BACKGROUND

[0002] Offshore intelligent ships are considered to be the main means of future maritime transportation, as they significantly improve the efficiency of maritime transportation and increase safety, and according to the International Maritime Organization (IMO) convention, offshore intelligent ships can sail independently to a certain extent without human intervention, thereby greatly reducing the human resource cost of maritime transportation. The berthing work of a ship is generally operated by experienced crew members. However, for offshore intelligent ships, it is difficult to independently and autonomously berth from open waters to the berthing area of a port without human operation, and therefore it is necessary to study safe and autonomous berthing of intelligent ships.

[0003] Autonomous berthing is a challenging task, which includes controlling the operation of a ship to sail to a target berth and maintain a desired heading while avoiding collisions with other obstacles in the presence of environmental disturbances and spatial constraints. Existing researches usually apply reinforcement learning, deep learning, artificial neural networks, fuzzy logic systems, model predictive control, adaptive dynamic programming and other methods to the field of autonomous berthing. From the perspective of control, the existing methods for autonomous berthing of intelligent ships still have the following shortcomings:

[0004] First, the existing learning-based autonomous berthing control methods based on reinforcement learning and deep learning require a large amount of data for training, which increases the computational burden, and the control method relies on the support of high-performance computing hardware, and most low-cost hardware may not be able to achieve real-time control, or even fail to achieve the expected control effect.

[0005] Second, the existing control methods of artificial neural networks and fuzzy logic mostly need to know the internal situation of the port in advance and plan the path, and if the ship needs to berth in a completely new port, it is impossible to make reasonable planning without knowing the internal situation of the port.

[0006] Third, the existing autonomous berthing control methods usually only consider the problem of the last transient state of berthing, and do not consider the problem of collision between ships entering and leaving the port, and do not consider the problem of collision between the ship and the berth. SUMMARY

[0007] The present application provides a safe and autonomous berthing control system for offshore intelligent ships to overcome the technical problems of difficulty in autonomous berthing of large ships and easy collision between ships entering and leaving the port.

[0008] In order to achieve the above-mentioned purpose, the technical scheme of the present application is:

[0009] An intelligent marine vessel safety autonomous berthing control system, characterized in that it comprises a kinematics control module, a speed optimization module, an extended state observer module and a dynamics control module;

[0010] The kinematics control module obtains the desired speed and heading of the intelligent marine vessel to reach the target position according to the desired berthing position of the intelligent marine vessel and the current position of the intelligent marine vessel;

[0011] The speed optimization module calculates the optimized speed of the intelligent marine vessel according to the desired speed of the intelligent marine vessel to reach the target position and the distance constraint between the obstacles and the intelligent marine vessel during navigation;

[0012] The extended state observer module is used to obtain the disturbance estimation value of the environment suffered by the intelligent marine vessel;

[0013] The dynamics control module is used to obtain the torque value required by the intelligent marine vessel according to the optimized speed and the disturbance estimation value suffered by the intelligent marine vessel.

[0014] Further, the specific steps of the kinematics control module to obtain the speed required by the intelligent marine vessel to reach the target position are:

[0015] Step 1.1, according to the intelligent marine vessel pose information and the desired berthing position of the intelligent marine vessel, a tracking error is obtained, and the specific calculation formula of the tracking error is:

[0016]

[0017]

[0018] Where x and y are the horizontal and vertical coordinate positions of the intelligent marine vessel, and ψ is the yaw angle information of the intelligent marine vessel in the earth coordinate system; χ d ,y d ,ψ d are the horizontal and vertical coordinate positions and the heading information of the desired berthing position of the intelligent marine vessel, R T (ψ d ) is the transpose of R(ψ d ), s is the forward tracking error of the intelligent marine vessel, e is the lateral tracking error of the intelligent marine vessel, ψ e is the angle tracking error of the intelligent marine vessel, and R(ψ) is the coordinate conversion matrix, which rotates the position in the earth coordinate system to the ship coordinate system;

[0019] Step 1.2, the speed required by the intelligent marine vessel to reach the target position is obtained by using the tracking error, and the specific calculation formula of the speed required by the intelligent marine vessel to reach the target position is:

[0020]

[0021] wherein u c is the forward velocity required for the intelligent ship to reach the target position, v c is the lateral velocity required for the intelligent ship to reach the target position, r c is the angular velocity required for the intelligent ship to reach the target position, are real numbers, and k1, k2, k3, Δ1, Δ2, Δ3 are design constants.

[0022] Further, the speed optimization module calculates the specific calculation steps of the optimized speed of the intelligent ship according to the expected speed required for the intelligent ship to reach the target position and the distance constraint between the obstacle and the intelligent ship during navigation, which are as follows:

[0023] Step 2.1, the obstacle avoidance function h j (p, p j ) is obtained according to the current position of the intelligent ship and the position of the obstacle during navigation. j The specific calculation formula of the obstacle avoidance function h j (p, p T ) is as follows:

[0024]

[0025] wherein p = (x, y) j is the current position of the intelligent ship, p j = (x j , y T ) j is the position of the obstacle, x j , y j are the horizontal coordinate position and the vertical coordinate position of the obstacle, and d j is the safety distance between the intelligent ship and the obstacle.

[0026] Step 2.2, the safety speed constraint is obtained by deriving the obstacle avoidance function h j (p, p j ), and the specific formula of the constraint is as follows:

[0027]

[0028]

[0029]

[0030] wherein is the derivative of the obstacle avoidance function h j (p, p j ), and (ψ) is a coordinate rotation matrix, γj is the design coefficient, is p j derivative;

[0031] Step 2.3, the speed required for the intelligent ship to reach the target position is optimized by using the obstacle avoidance function, and the specific optimization formula is:

[0032] argmin||q s -q c || 2

[0033]

[0034] A j =-2(p-p j ) T R (ψ)

[0035] wherein q s is the speed after adding the collision avoidance constraint optimization, q c =[u c ,v c ] is the expected speed of the intelligent ship to reach the target position, A j is the expression of the constraint condition, wherein q s =[u s ,v s ], u s is the forward speed after optimization, v s is the transverse speed after optimization, and s.t. represents the constraint content.

[0036] Further, the specific calculation formula of the extended state observer module obtaining the disturbance estimation value of the environment suffered by the intelligent ship is:

[0037]

[0038] wherein v=[u,v,r] is the actual speed of the ship, K1,K2 is the control gain, σ is the disturbance value, is the estimation value of σ, is the estimation value of v, M is a matrix composed of weight inertia and hydrodynamic additional inertia, represents the longitudinal velocity, transverse velocity and yaw angular velocity of the controlled intelligent ship in the ship reference frame.

[0039] Further, the calculation formula of the dynamics control module for deriving the torque value required by the intelligent ship according to the optimized speed and the disturbance value suffered by the intelligent ship is:

[0040]

[0041] s.t.τ min ≤τ≤τ max

[0042] where K c is the control gain matrix, is the velocity error, v s = [u s , v s , r c ] is the optimized velocity, i.e., q s .

[0043] Advantages:

[0044] First, compared with the existing learning-based berthing control method, the berthing control method proposed in the present application can calculate the required torque for stopping according to the existing ship position, target position and obstacle position, does not rely on a large amount of data learning, saves a large amount of hardware resources, and improves the real-time performance of the control.

[0045] Second, compared with the existing intelligent ship path planning method, the control method based on optimization designed in the present application does not need to be familiar with the internal environment of the berthing port in advance to plan the path, and can also guarantee the safety and stability of berthing, and is more universal.

[0046] Third, compared with the existing intelligent ship collision avoidance method, the obstacle avoidance method designed in the present application not only considers the influence of inherent terrain obstacles, but also considers the collision avoidance between ships in a complex port environment, and has simple structure and high safety. It makes the operation of the unmanned mode of the intelligent ship more reliable. BRIEF DESCRIPTION OF DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0048] Figure 1 is a structure diagram of a safe autonomous berthing control system of an intelligent ship at sea;

[0049] Figure 2 is a motion trajectory simulation diagram of a safe autonomous berthing control system of an intelligent ship at sea;

[0050] Figure 3 is a simulation diagram of expected velocity, actual velocity, estimated velocity and optimized velocity in the longitudinal direction of autonomous berthing;

[0051] Figure 4is a simulation diagram of the expected speed, actual speed, estimated speed and optimized speed of the autonomous berthing in the sway direction;

[0052] Figure 5 is a simulation diagram of the expected speed, actual speed, estimated speed and optimized speed of the autonomous berthing in the yaw direction;

[0053] Figure 6a is a simulation diagram of the thrust provided by the ship in the surge direction during the autonomous berthing;

[0054] Figure 6b is a simulation diagram of the thrust provided by the ship in the sway direction during the autonomous berthing;

[0055] Figure 6c is a simulation diagram of the thrust provided by the ship in the yaw direction during the autonomous berthing;

[0056] Figure 7a is a simulation diagram of the distance between the intelligent ship and the first obstacle during the autonomous berthing;

[0057] Figure 7b is a simulation diagram of the distance between the intelligent ship and the second obstacle during the autonomous berthing;

[0058] Figure 8a is a simulation diagram of the wind disturbance on the ship in the surge direction during the autonomous berthing.

[0059] Figure 8b is a simulation diagram of the wind disturbance on the ship in the sway direction during the autonomous berthing.

[0060] Figure 8c is a simulation diagram of the wind disturbance on the ship in the yaw direction during the autonomous berthing. DETAILED DESCRIPTION

[0061] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0062] The present embodiment provides a safe autonomous berthing control system for an intelligent ship at sea, as shown in Figure 1 , which comprises a kinematic control module, a speed optimization module, an extended state observer module and a dynamic control module.

[0063] The kinematic control module derives the desired speed and heading of the intelligent ship to reach the target position according to the desired berthing position of the intelligent ship and the current position of the intelligent ship;

[0064] The kinematic control module presets the kinematic model and the dynamic model of the intelligent ship;

[0065] The kinematic model of the intelligent ship is represented as:

[0066]

[0067] The dynamic model is represented as:

[0068] wherein respectively represent the position and the yaw angle information of the controlled intelligent ship in the earth coordinate system under the earth reference system; represent the surge velocity, the sway velocity and the yaw angle velocity of the controlled intelligent ship under the ship body reference system; M is a matrix composed of weight inertia and water dynamic additional inertia; C(v) is a centripetal force and Coriolis force coefficient matrix; D(v) is a damping matrix; τ is a moment; w is a disturbance, wherein the disturbance is designed to simulate the disturbance of wind, and the structure is as follows:

[0069]

[0070] wherein, is a parameter related to the wind speed, is the included angle between the ship and the wind direction, ρ a is the air density, C X ,C Y ,C N is the wind coefficient, A FW ,A LW is the front and lateral projection area of the ship, L oa is the overall length of the ship.

[0071] The speed optimization module calculates the optimized speed of the intelligent ship according to the desired speed of the intelligent ship to reach the target position and the distance constraint between the obstacles and the intelligent ship during the navigation process;

[0072] The extended state observer module is used to obtain the disturbance estimation value of the environment suffered by the intelligent ship;

[0073] The dynamic control module is used to derive the moment value required by the intelligent ship according to the optimized speed and the disturbance estimation value suffered by the intelligent ship.

[0074] In specific embodiments, the specific steps of the kinematic control module to derive the speed required by the intelligent ship to reach the target position are:

[0075] Step 1.1, according to the intelligent ship pose information and the intelligent ship expected berthing position, a tracking error is obtained as the input of the kinematics control module, and the specific calculation formula of the tracking error is:

[0076]

[0077]

[0078] Wherein, x, y are the horizontal coordinate position and the longitudinal coordinate position of the intelligent ship, and ψ is the yaw angle information of the intelligent ship in the earth coordinate system; χ d , y d , ψ d are the horizontal coordinate position and the longitudinal coordinate position and the heading information of the expected berthing position of the intelligent ship, R T (ψ d ) is the transpose of R(ψ d ), s is the forward tracking error of the intelligent ship, e is the lateral tracking error of the intelligent ship, ψ e is the angle tracking error of the intelligent ship, and R(ψ) is the coordinate conversion matrix, which rotates the position in the earth coordinate system to the ship body coordinate system;

[0079] Step 1.2, the tracking error is used to obtain the speed required for the intelligent ship to reach the target position, and the specific calculation formula of the speed required for the intelligent ship to reach the target position is:

[0080]

[0081] Wherein, u c is the forward speed required for the intelligent ship to reach the target position, v c is the lateral speed required for the intelligent ship to reach the target position, r c is the angle speed required for the intelligent ship to reach the target position, is a real number, and k1, k2, k3, Δ1, Δ2, Δ3 are design constants.

[0082] In specific embodiments, the speed optimization module calculates the specific calculation steps of the optimized speed of the intelligent ship according to the expected speed required for the intelligent ship to reach the target position and the distance constraint between the obstacle and the intelligent ship during navigation.

[0083] Step 2.1, according to the current position of the intelligent ship and the position of the obstacle during navigation, an obstacle avoidance function h j (p, p j ) is defined, and the specific calculation formula of the obstacle avoidance function h j (p, p j ) is:

[0084]

[0085] where p = (x, y) T is the current position of the intelligent ship, p j = (x j , y j ) T is the position of the obstacle, x j , y j is the horizontal and vertical position of the obstacle, d j is the safety distance between the intelligent ship and the obstacle;

[0086] Step 2.2, derive the safety speed constraint by the obstacle avoidance function h j (p, p j ), the specific formula of the constraint is:

[0087]

[0088]

[0089]

[0090] where is the derivative of the obstacle avoidance function h j (p, p j ), R (ψ) is the coordinate rotation matrix, γ j is the design coefficient, is the derivative of p j , where is a static obstacle, is a dynamic obstacle;

[0091] Step 2.3, optimize the speed required for the intelligent ship to reach the target position using the obstacle avoidance function, the specific optimization formula is:

[0092] argmin||q s -q c || 2

[0093]

[0094] A j = -2(p-p j ) T R (ψ)

[0095] where q s is the speed after adding the collision avoidance constraint optimization, q c = [u c , vc ] is the desired speed of the intelligent ship to reach the target position, A j is the expression of the constraint condition, where q s = [u s ,v s ], u s is the optimized forward speed, v s is the optimized lateral speed, and s.t. represents the constraint content.

[0096] In specific embodiments, the specific calculation formula of the disturbance estimation value of the environment suffered by the intelligent ship obtained by the extended state observer module is as follows:

[0097]

[0098] wherein v = [u, v, r] is the actual speed of the ship, K1, K2 are control gains, and s is the disturbance value, is the estimation value of s, is the estimation value of v, M is a matrix composed of weight inertia and hydrodynamic additional inertia, represents the longitudinal velocity, lateral velocity, and yaw angular velocity of the controlled intelligent ship in the ship reference system.

[0099] In specific embodiments, the calculation formula of the required torque value of the intelligent ship derived by the dynamics control module according to the optimized speed and the disturbance value suffered by the intelligent ship is as follows:

[0100]

[0101] s.t. t min ≤ t ≤ t max

[0102] wherein K c is a control gain matrix, is a speed error, v s = [u s ,v s ,r c ] is the optimized speed, i.e., q s .

[0103] In specific embodiments, the specific parameters of the intelligent ship model are selected as follows:

[0104]

[0105]

[0106] The initial positions of the controlled intelligent ship and the target berth are set as and (x j1 ,yj1 ) = (10m, 9m), (x j2 ,y j2 ) = (55m, 33m) are the positions of static obstacles, d j1 = 6, d j2 = 12 are the safety distances for collision avoidance, and other specific control parameters are chosen as follows:

[0107]

[0108] p a = 1.224 is the air density at 20°C, V w = 20, A Fw = 10, A Lw = 30, L oa = 3.

[0109] CX = - (A0+ A1*2*ALw / Loa 2 +A2*2*AFw / B 2 +A3*(Loa / B) ;

[0110] CY = B0+B1*2*ALw / Loa 2 +B2*(Loa / B) +B3*(C / Loa) ;

[0111] CN = C0+C1*2*ALw / Loa 2 +C2*(C / Loa) ;

[0112] wherein:

[0113] A0 = 0.1*sin(t / 10) + 2.1; A1 = 4.5 + 0.5*sin(t / 10); A2 = 0.05*rand + 0.2; A3 = -0.02*rand - 0.15;

[0114] B0 = 0.01*sin(t / 20) + 0.01; B1 = 0.7*sin(t / 15) + 0.75; B2 = 0.0125*sin(t / 20) + 0.0125; B3 = -0.2 - 0.15*sin(t / 20);

[0115] C0 = 0.01*sin(t / 20); C1 = 0.01*sin(t / 20); C2 = -0.0001*t*1.3; t is the time of simulation running, and B = 1 is the ship width.

[0116] k1 = 1.2, k2 = 0.35, k3 = 0.15

[0117] Δ1 = 150, Δ2 = 150, Δ3 = 20, γ j=0.3

[0118] τ max =[10*10^5,2*10^5,5*10^7],τ min =-τ max

[0119] Simulation results are as follows Figure 2 As shown in Figure -8. Figure 2 The trajectory diagram of the autonomous berthing motion of an intelligent ship shows that the controlled intelligent ship can effectively avoid obstacles and reach the designated destination. Figures 3-5 The speeds in the sway, drift, and yaw directions during the movement of an intelligent ship can be calculated. The optimized speed is basically consistent with the estimated speed and the actual speed, and the intelligent ship can sail at the desired speed. Figure 6a , Figure 6b , Figure 6c Forces in three directions provided during the navigation of intelligent ships; Figure 7a , Figure 7b Given the distance between an intelligent ship and two different obstacles during navigation, it can be determined that the minimum distance between the ship and the obstacle is within the safety constraints. Figure 8a , Figure 8b , Figure 8c To simulate the actual and estimated forces exerted by wind on a ship during wind disturbances, this invention provides a good estimate of the impact of wind on a ship.

[0120] Simulation results show that the intelligent ship safe autonomous berthing control method and system of the present invention can accurately and reliably reach the target position and effectively avoid the set obstacles. At the same time, it takes into account the impact of wind on ship navigation and performs anti-disturbance processing on its own, which verifies the effectiveness of the method.

[0121] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A smart ship safety autonomous berthing control system for offshore, characterized in that, The method comprises the following steps: The kinematics control module, the speed optimization module, the extended state observer module and the dynamics control module; The kinematics control module obtains the desired speed and heading of the intelligent ship to reach the target position according to the desired berthing position of the intelligent ship and the current position of the intelligent ship; The speed optimization module calculates the optimized speed of the intelligent ship according to the desired speed of the intelligent ship to reach the target position and the distance constraint between the obstacle and the intelligent ship during navigation; The extended state observer module is used to obtain the disturbance estimation value of the environment suffered by the intelligent ship; The dynamics control module is used to obtain the torque value required by the intelligent ship according to the optimized speed and the disturbance estimation value suffered by the intelligent ship; The specific steps of the kinematics control module to obtain the speed required by the intelligent ship to reach the target position are as follows: Step 1.1, according to the intelligent ship pose information and the desired berthing position of the intelligent ship, the tracking error is obtained, and the specific calculation formula of the tracking error is as follows: wherein, is a lateral position and a longitudinal position of the smart ship, is a yaw angle information of the smart ship in the earth coordinate system; is a desired berthing position of the smart ship, including a lateral position and a longitudinal position and a heading information, is a transpose of is a transpose of is a forward tracking error of the smart ship, is a lateral tracking error of the smart ship, is an angle tracking error of the smart ship, is a coordinate conversion matrix, rotating the position in the earth coordinate system to the ship coordinate system; Step 1.2, the speed required by the intelligent ship to reach the target position is obtained by using the tracking error, and the specific calculation formula of the speed required by the intelligent ship to reach the target position is as follows: wherein, is the forward velocity required for the intelligent ship to reach the target position, is the lateral velocity required for the intelligent ship to reach the target position, is the angular velocity required for the intelligent ship to reach the target position, , is a real number, , , , , , are all design constants.

2. A smart ship safety autonomous berthing control system for offshore use as claimed in claim 1, characterized in that: The specific calculation steps of the speed optimization module to calculate the optimized speed of the intelligent ship according to the desired speed of the intelligent ship to reach the target position and the distance constraint between the obstacle and the intelligent ship during navigation are as follows: Step 2.1, define an obstacle avoidance function according to the current position of the intelligent ship and the position of the obstacle in the navigation process , obtain the obstacle avoidance function The specific calculation formula is: wherein, is the current position of the intelligent ship, is the position of the obstacle, is the horizontal and vertical position of the obstacle, is the safety distance between the intelligent ship and the obstacle; Step 2.2, get the safety velocity constraint by derivation of the obstacle avoidance function The specific formula of the constraint is: wherein is an obstacle avoidance function is a derivative of is a coordinate rotation matrix is a design coefficient is a derivative of is a derivative of Step 2.3, the speed required by the intelligent ship to reach the target position is optimized by using the obstacle avoidance function, and the specific optimization formula is as follows: wherein, is the optimized speed after adding collision avoidance constraints, is the desired speed of the intelligent ship to reach the target position, is the expression of the constraint condition, wherein, , is the optimized forward speed, is the optimized lateral speed, represents the constraint content.

3. A smart ship safety autonomous berthing control system for offshore use as claimed in claim 2, wherein: The specific calculation formula of the extended state observer module to obtain the disturbance estimation value of the environment suffered by the intelligent ship is as follows: wherein, is the actual speed of the ship, is the control gain, is the disturbance value, is the estimated value of is the estimated value of is the matrix of weight inertia and hydrodynamic added inertia, denotes the surge, sway and yaw angular velocities of the controlled intelligent ship in the ship reference frame.

4. A smart ship safety autonomous berthing control system for offshore use as claimed in claim 3, wherein: The calculation formula of the dynamics control module to obtain the torque value required by the intelligent ship according to the optimized speed and the disturbance value suffered by the intelligent ship is as follows: wherein is a control gain matrix, is a speed error, is an optimized speed, i.e. .

Images