Unmanned ship obstacle avoidance path planning method and system based on global optimum

Abstract

The invention discloses an unmanned ship obstacle avoidance path planning method and system based on global optimum, and belongs to the field of unmanned ship navigation obstacle avoidance. Comprising the steps that S1, a kinetic model of the unmanned ship is constructed, a motion coordinate system is defined, and the unmanned ship has corresponding speeds in the surging direction, the swaying direction and the yawing direction; s2, improving and simplifying the dynamic model of the unmanned ship through the influence of the water environment influence factors on the dynamic characteristics of the unmanned ship; s3, formulating a collision avoidance strategy based on predictive control, performing discretization processing, calculating parameters encountering other ships and a collision risk, and determining a target function in the collision avoidance strategy; and S4, determining a final collision avoidance path method, and searching a path global optimal solution of the unmanned ship through a loop iteration mode. According to the technical scheme provided by the invention, the problem that unmanned ship path planning and obstacle avoidance cannot be accurately and efficiently carried out in the prior art is effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned ship navigation and obstacle avoidance, and in particular to an unmanned ship obstacle avoidance path planning method and system based on global optimization. Background Art

[0002] Ship collision avoidance has always been a hot topic in the field of maritime safety. Traditional path generation methods include A* and its improved algorithms and artificial potential field APF and its improved algorithms. These path planning algorithms have the problem of insufficient consideration of ship inertia and nonlinear dynamics; intelligent optimization methods that have emerged in recent years, such as fuzzy mathematics, neural networks, swarm intelligence and machine learning, have considered the ship collision risk index and the International Regulations for Preventing Collisions at Sea (COLREGs), but have insufficient consideration of ship maneuverability; real-time optimization methods have the problem of being unable to achieve local optimality when setting static constraints, and have not considered the real-time impact of the water environment. Summary of the invention

[0003] In view of the defects of related technologies, the present invention provides an unmanned ship obstacle avoidance path planning method and system based on global optimization, aiming to solve the problem in the prior art that the path planning of unmanned ship collision avoidance cannot be achieved efficiently and accurately based on water environment conditions.

[0004] The technical solution is as follows:

[0005] A global optimal unmanned ship obstacle avoidance path planning method comprises the following steps:

[0006] Step S1: Construct the dynamic model of the unmanned ship and define the inertial coordinate system O O -x o y o and the motion coordinate system of the ship coordinate system O-xy, in which the unmanned ship has corresponding velocities u, v, r in the three directions of sway, swing and pitch, thereby constructing the dynamic model of the unmanned ship;

[0007] Step S2: improving the dynamic model of the unmanned ship by considering the influence of water environment factors on the dynamic characteristics of the unmanned ship, obtaining an improved dynamic model of the unmanned ship and simplifying it;

[0008] Step S3: Formulate a collision avoidance strategy based on predictive control, discretize the improved and simplified unmanned ship dynamics model, use geometric methods to calculate the parameters of encountering other ships, calculate the risk of collision based on fuzzy mathematics, and determine the objective function in the collision avoidance strategy;

[0009] Step S4: Determine the final collision avoidance path method, and find the global optimal solution for the path of the unmanned ship through a loop iteration method.

[0010] An unmanned ship obstacle avoidance path planning system based on global optimization, the system comprising:

[0011] The unmanned ship dynamics model building module is used to define the inertial coordinate system O O -x o y o and the motion coordinate system of the ship coordinate system O-xy, in which the unmanned ship has corresponding velocities u, v, r in the three directions of sway, swing and pitch, thereby constructing the dynamic model of the unmanned ship;

[0012] The unmanned ship dynamics model improvement and simplification module is used to improve the unmanned ship dynamics model by considering the influence of water environment factors on the unmanned ship dynamics characteristics, obtain the improved unmanned ship dynamics model and simplify it;

[0013] The collision avoidance strategy module based on predictive control is used to discretize the improved and simplified unmanned ship dynamics model, calculate the parameters of encountering other ships using geometric methods, calculate the risk of collision based on fuzzy mathematics, and determine the objective function in the collision avoidance strategy;

[0014] The collision avoidance path planning module is used to find the global optimal solution for the path of the unmanned ship through a loop iteration method.

[0015] The present invention has the following beneficial effects:

[0016] The present invention provides an unmanned ship obstacle avoidance path planning method and system based on global optimization. The power model of the unmanned ship is improved and simplified by considering the interference factors of the water environment, thereby improving the precision and accuracy of the unmanned ship modeling. During collision prediction, discretization processing, geometric calculation and optimal objective function are used to ensure the safety of the ship and minimize unnecessary operations and resource consumption. In the iterative process of the path planning stage, the influence of the water environment is also given priority consideration to minimize the collision risk and the change of the rudder angle, so as to achieve safe and efficient collision avoidance. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flow chart of a method for unmanned ship obstacle avoidance path planning based on global optimization provided by an embodiment of the present invention;

[0018] Figure 2 It is a flowchart of improving and simplifying the unmanned ship dynamics model provided by an embodiment of the present invention;

[0019] Figure 3 is a flow chart of formulating a collision avoidance strategy based on predictive control provided by an embodiment of the present invention;

[0020] Figure 4 is a collision avoidance path planning flow chart provided by an embodiment of the present invention;

[0021] Figure 5 It is a structural diagram of an unmanned ship obstacle avoidance path planning system based on global optimization provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0023] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and / or" used herein includes all or any unit and all combinations of one or more associated listed items.

[0024] See also Figure 1 The present invention provides a global optimal unmanned ship path cooperative control method, comprising the following steps:

[0025] Step S1: Construct the dynamic model of the unmanned ship and define the inertial coordinate system O O -x o y o and the motion coordinate system of the ship coordinate system O-xy, in which the unmanned ship has corresponding velocities u, v, r in the three directions of sway, swing and pitch, thereby constructing the dynamic model of the unmanned ship;

[0026] Step S2: improving the dynamic model of the unmanned ship by considering the influence of water environment factors on the dynamic characteristics of the unmanned ship, obtaining an improved dynamic model of the unmanned ship and simplifying it;

[0027] Step S3: Formulate a collision avoidance strategy based on predictive control, discretize the improved and simplified unmanned ship dynamics model, use geometric methods to calculate the parameters of encountering other ships, calculate the risk of collision based on fuzzy mathematics, and determine the objective function in the collision avoidance strategy;

[0028] Step S4: Determine the final collision avoidance path method, and find the global optimal solution for the path of the unmanned ship through a loop iteration method.

[0029] Step S1: Construct the dynamic model of the unmanned ship and define the inertial coordinate system O O -x o y o And the motion coordinate system of the ship coordinate system O-xy, in which the unmanned ship has corresponding velocities u, v, r in the three directions of sway, swing and pitch, thereby constructing the dynamic model of the unmanned ship, specifically:

[0030]

[0031] Among them, m is the total mass of the unmanned ship, x G is the longitudinal coordinate of the center of gravity of the unmanned ship in the longitudinal direction, I Z is the inertia matrix, and X, Y, and N are the hydrodynamic forces of the unmanned ship in the three directions of longitudinal, transverse, and bow pitch when there is no environmental interference.

[0032] When an unmanned ship is traveling in an actual water environment, it will inevitably be affected by the water environment. The unmanned ship dynamic model that only considers the unmanned ship without environmental interference will obviously have a large error; and the impact of the water environment is mainly reflected in the changes in the plane position and angle of the unmanned ship, which is related to the time variable t, and the speed of change in this plane position and angle is what we need to focus on. It is necessary to convert the velocity vectors of the unmanned ship in the three directions of longitudinal, transverse and bow roll in its own coordinate system to the inertial coordinate system. This conversion is related to the heading angle ψ of the unmanned ship. In order to further improve the accuracy of the modeling, it is necessary to adopt the following step S2 to improve the unmanned ship dynamic model under the influence of the water environment.

[0033] See also Figure 2 , Step S2: According to the influence of water environment factors on the dynamic characteristics of the unmanned ship, the dynamic model of the unmanned ship is improved, and the improved dynamic model of the unmanned ship is obtained and simplified, specifically:

[0034] Step S21: Define parameters It reflects how fast the plane position and angle of the unmanned ship change when it is affected by the water environment, and the conversion matrix R(ψ(t))·[u(t),v(t),r(t)] is used to convert the velocity vectors of the unmanned ship in the three directions of sway, swing and pitch in its own coordinate system to the inertial coordinate system. The following formula is used to express the influence of the water environment on dynamics:

[0035]

[0036] Among them, M is the center of gravity matrix of the unmanned ship in three directions; τ(t) represents the control input torque vector generated by the rudder force and propeller thrust on the unmanned ship; C(V(t)) represents the force matrix generated by the rotational motion of the unmanned ship, which is related to the speed change of the unmanned ship; D(V(t)) represents the damping force matrix generated by the viscosity of water and the friction between the ship and water, which is also related to the speed change of the unmanned ship; b(t) represents the interference force generated by environmental factors such as wind, waves, and water flow in the environment on the ship;

[0037] Step S22: The calculation of C(V(t)) is derived based on the physical formula of centripetal force, which is:

[0038]

[0039] For the calculation of D(V(t)), since it represents the viscosity of water and the damping force generated by the friction between the boat and water, it is necessary to experimentally measure the damping coefficient of different unmanned boat materials in water. u d v and d r Then the calculation of D(V(t)) can be defined as:

[0040]

[0041] For the calculation of b(t), the influence of wind speed, wave height and water flow speed needs to be considered, which is defined as:

[0042]

[0043] Among them, v w (t) is the wind speed, θ w is the angle between the wind direction and the coordinate axis of the unmanned ship, C w is the force coefficient of wind on the unmanned ship; v c (t) Water flow velocity, θ c is the angle between the water flow and the coordinate axis of the unmanned ship, C c is the force coefficient of the water flow on the unmanned boat; h l (t) is the wave height, T l is the wave cycle, C lis the force coefficient of the wave on the unmanned ship; a, b, and c are the angle constant coefficients related to the steering characteristics of the unmanned ship and the motion characteristics in the waves;

[0044] Step S23: Improve the dynamic model of the unmanned ship. Substitute the formulas that affect the speed of the unmanned ship in three directions in step S22 into formula (1) for mathematical derivation to obtain the improved dynamic model of the unmanned ship. At the same time, in order to facilitate the calculation of the subsequent predictive control collision avoidance strategy and the final collision avoidance path method, the physical parameters of the improved dynamic model are simplified to obtain:

[0045]

[0046] Among them, X s It is a vector representation containing the motion state information of the unmanned ship. u, v, r are still the speeds of the unmanned ship in three directions. x, y are the coordinates of the unmanned ship. ψ is the heading angle. f is X s The derivative function of X b , Y b 、N b It is the hydrodynamic expression of the unmanned ship in the three directions of sway, swing and pitch after the influence of various forces under the condition of introducing the environmental disturbance of step S22; A X , A Y , is the constant coefficient matrix of the model state variables, which is used to describe the relationship between the state variables and their influence on the motion, and θ is the representation of the evidence formed by extracting the diagonal elements of the required matrix, which is substituted into f to represent the influence of the state variables on the motion state information of the unmanned ship; and D x , D y , D ψ It is a variable related to the position and heading of the unmanned ship, which is calculated through the speed and angle of the unmanned ship, and reflects the motion state and direction of the unmanned ship in the plane. Through such simplified processing, the simplicity and efficiency of the subsequent predictive control collision avoidance strategy and the final collision avoidance path method calculation are greatly improved.

[0047] See also Figure 3 ,Step S3: Formulate a collision avoidance strategy based on predictive control, discretize the improved and simplified unmanned ship dynamics model, use geometric methods to calculate the parameters of encountering other ships, and calculate the risk of collision based on fuzzy mathematics, and determine the objective function in the collision avoidance strategy, specifically:

[0048] Step S31: discretizing the improved and simplified unmanned ship dynamics model, including:

[0049]

[0050] Among them, X s,nis the state vector at the time step, Δt is the time step, and f is still the derivative function;

[0051] Step S32: The discretization process not only processes the current unmanned ship but also the target ship. The encounter parameters of the current unmanned ship and the target ship are calculated by geometric methods, and the state of the current unmanned ship is set to X. s , the state of the target ship is X T , then the relative position, relative speed, relative speed direction, closest encounter distance DCPA and time to reach the closest encounter point TCPA between the two are expressed by the following set of equations:

[0052]

[0053] Among them, Δα is the relative velocity direction, Δv x and Δv y is the component; TCPA can be solved by Come get;

[0054] These encounter parameters are crucial for assessing the risk of collision between two ships. DCPA represents the closest distance that the two ships may reach in the future. If DCPA is less than a certain safety threshold, it means that there is a risk of collision between the two ships. TCPA represents the time required for the two ships to reach the closest encounter point, which can help determine whether the two ships will collide at some point in the future.

[0055] Step S33: Calculate the risk of collision based on fuzzy mathematics, and determine the objective function in the collision avoidance strategy, setting the collision risk as F CRI (t), then:

[0056]

[0057] Among them, ω1~ω5 are the membership functions of various parameters f DCPA 、f TCPA 、f v , The weight, v C Represents the speed ratio of the two;

[0058] At the same time, the objective function is set to J, and the actual rudder angle δ is selected as the control variable. The goal is to minimize the predicted collision risk and rudder angle change. The objective function is established as:

[0059] J=ω risk |F CRI (t)|+ω δ |Δδ|; (10)

[0060] Among them, ω risk and ω δare weight coefficients, which are used to adjust the importance of predicting collision risk and rudder angle change in the objective function; |*| is the norm operation; Δδ is the rudder angle change value of this unmanned ship.

[0061] The calculation of collision risk adopts fuzzy mathematics method. By setting membership functions and weights for each encounter parameter, the influence of multiple factors on collision risk is comprehensively considered. The membership function is used to describe the fuzzy relationship between each parameter and collision risk. For example, the smaller the DCPA, the larger the corresponding membership function value, indicating a higher collision risk. The weight is used to adjust the importance of each parameter in the calculation of collision risk. In this way, the collision risk between the two ships can be evaluated more accurately and dynamic prediction can be performed within the prediction time domain.

[0062] A reasonable objective function is the key to optimizing collision avoidance decisions; the objective function needs to comprehensively consider safety costs and economic costs in order to ensure ship safety while minimizing unnecessary operations and resource consumption.

[0063] See also Figure 4 , Step S4: Determine the final collision avoidance path method, and find the global optimal solution of the path of the unmanned ship through a loop iteration method, specifically:

[0064] Step S41: The actual rudder angle δ is the control variable, the number of initial search control variables is determined, and the maximum number of iterations T is set. max , initial step size h0, step size adjustment factor adj and step size lower limit ε;

[0065] During the search process, the search step size is dynamically adjusted according to the feedback information of the algorithm. The search step size determines the distance that the algorithm moves in each iteration in the search space. If the step size is too large, the area where the optimal solution is located may be skipped. If the step size is too small, the search speed will be too slow. Therefore, the step size adjustment factor adj and the step size lower limit ε are set. In the early stage of the search, a relatively large initial step size is set. When the search falls into the vicinity of the local optimum, it is judged by comparing the objective function value of the current individual (that is, the current rudder angle sequence) with the surrounding individuals. If the objective function value does not improve significantly after several consecutive iterations, it is considered that it may fall into the local optimum. At this time, the step size is reduced to (where is the current step size) to improve the search accuracy and explore the current local area more carefully. When the search range is too large and no effective solution is found, that is, in several consecutive iterations, the newly generated individual objective function value is greater than the objective function value of the current optimal individual, and the difference exceeds a certain threshold, the step size is increased to speed up the search and expand the search range. At the same time, in order to avoid the search being unable to proceed due to too small a step size, the step size lower limit is set to the set step size.

[0066] Step S42: for each individual in the initial search control variable, i.e., the rudder angle sequence, the corresponding objective function value is calculated according to the ship dynamics model and the definition of the objective function;

[0067] Step S43: iteratively process the calculated objective function value. In each iteration, determine the search direction according to the individual information and the environmental interference force vector in the current control variable, calculate the difference between the objective function value and the surrounding individuals, so as to determine the direction in which the objective function value decreases; and combine the environmental influence vector at the same time. If the angle between the objective function value decreasing direction and the environmental influence vector is small, determine to search in this direction;

[0068] Unmanned ships are affected by environmental interference force vectors (such as wind direction, water flow direction, etc.) during actual navigation. These environmental factors have an important impact on the motion trajectory and optimal collision avoidance strategy of the ship. Incorporating environmental information into the algorithm's search direction determination can enable the algorithm to search for the optimal rudder angle sequence in a direction that is more conducive to collision avoidance.

[0069] First, the environmental interference force vector is quantified and standardized. For example, the wind direction information is converted into a unit vector to represent the direction of the wind, and the water flow direction information is also converted into a unit vector. Then, according to the current position and speed of the unmanned ship and the relevant information of the target ship, a comprehensive environmental impact vector is calculated, which can be the weighted sum of the wind direction vector and the water flow vector. The weight is determined according to the importance of the environmental factors on the ship's motion. For example, if the water flow has a greater impact on the ship's motion than the wind direction, the weight of the water flow vector can be set larger. Finally, when determining the search direction, the environmental impact vector is combined with the change trend of the objective function value of individuals in the current population. If the objective function value decreases as the search moves in a certain direction, and the angle between the direction and the environmental impact vector is small, it is more inclined to search in that direction to take advantage of the favorable effects of environmental factors and improve the collision avoidance effect.

[0070] Step S44: The search step size is adjusted according to the dynamic step size adjustment mechanism. According to the determined search direction and the adjusted step size, a new control variable, i.e., the rudder angle sequence and the objective function set, is generated. When the number of iterations reaches the maximum number of iterations, the algorithm stops searching. At this time, the optimal rudder angle sequence is output as the final solution. This optimal rudder angle sequence will be used for the actual collision avoidance operation of the unmanned ship to minimize the collision risk and the change of the rudder angle, and to achieve safe and efficient collision avoidance.

[0071] See also Figure 5 , an unmanned ship obstacle avoidance path planning system 100 based on global optimization, the system comprising:

[0072] The unmanned ship dynamics model building module 101 is used to define an inertial coordinate system O O -xo y o and the motion coordinate system of the ship coordinate system O-xy, in which the unmanned ship has corresponding velocities u, v, r in the three directions of sway, swing and pitch, thereby constructing the dynamic model of the unmanned ship;

[0073] The unmanned ship dynamics model improvement and simplification module 102 is used to improve the unmanned ship dynamics model by considering the influence of water environment factors on the unmanned ship dynamics characteristics, obtain the improved unmanned ship dynamics model and simplify it;

[0074] The collision avoidance strategy module 103 based on predictive control is used to discretize the improved and simplified unmanned ship dynamics model, calculate the parameters of encountering other ships using geometric methods, calculate the risk of collision based on fuzzy mathematics, and determine the objective function in the collision avoidance strategy;

[0075] The collision avoidance path planning module 104 is used to find the global optimal solution for the path of the unmanned ship through a loop iteration method.

Claims

1. A global optimal unmanned ship obstacle avoidance path planning method, characterized in that: The following steps are involved: Step S1: construct a dynamic model of the unmanned ship, define a motion coordinate system including an inertial coordinate system and a ship-following coordinate system, in which the unmanned ship has corresponding velocities u, v, r in the three directions of sway, swing and pitch, thereby constructing a dynamic model of the unmanned ship; Step S2: improving the dynamic model of the unmanned ship by considering the influence of water environment factors on the dynamic characteristics of the unmanned ship, obtaining an improved dynamic model of the unmanned ship and simplifying it; Step S3: Formulate a collision avoidance strategy based on predictive control, discretize the improved and simplified unmanned ship dynamics model, use geometric methods to calculate the parameters of encountering other ships, calculate the risk of collision based on fuzzy mathematics, and determine the objective function in the collision avoidance strategy; Step S4: Determine the final collision avoidance path method, and find the global optimal solution for the path of the unmanned ship through a loop iteration method.

2. The unmanned ship obstacle avoidance path planning method based on global optimization as claimed in claim 1, characterized in that: The step S1 specifically includes: Among them, m is the total mass of the unmanned ship, x G is the longitudinal coordinate of the center of gravity of the unmanned ship in the longitudinal direction, I Z is the inertia matrix, and X, Y, and N are the hydrodynamic forces of the unmanned ship in the three directions of longitudinal, transverse, and bow pitch when there is no environmental interference.

3. The unmanned ship obstacle avoidance path planning method based on global optimization as claimed in claim 1, characterized in that: The step S2 specifically includes: Step S21: Define parameters It reflects how fast the plane position and angle of the unmanned ship change when it is affected by the water environment, and the conversion matrix R(ψ(t))·[u(t),v(t),r(t)] is used to convert the velocity vectors of the unmanned ship in the three directions of sway, swing and pitch in its own coordinate system to the inertial coordinate system. The following formula is used to express the influence of the water environment on dynamics: Among them, M is the center of gravity matrix of the unmanned ship in three directions; τ(t) represents the control input torque vector generated by the rudder force and propeller thrust on the unmanned ship; C(V(t)) represents the force matrix generated by the rotational motion of the unmanned ship, which is related to the speed change of the unmanned ship; D(V(t)) represents the damping force matrix generated by the viscosity of water and the friction between the ship and water, which is also related to the speed change of the unmanned ship; b(t) represents the interference force generated by environmental factors such as wind, waves, and water flow in the environment on the ship; Step S22: The calculation of C(V(t)) is derived based on the physical formula of centripetal force, which is: For the calculation of D(V(t)), since it represents the viscosity of water and the damping force generated by the friction between the boat and water, it is necessary to experimentally measure the damping coefficient of different unmanned boat materials in water. u ,d v and d r Then the calculation of D(V(t)) can be defined as: For the calculation of b(t), the influence of wind speed, wave height and water flow speed needs to be considered, which is defined as: Among them, v w (t) is the wind speed, θ w is the angle between the wind direction and the coordinate axis of the unmanned ship, C w is the force coefficient of wind on the unmanned ship; v c (t) Water flow velocity, θ c is the angle between the water flow and the coordinate axis of the unmanned ship, C c is the force coefficient of the water flow on the unmanned boat; h l (t) is the wave height, T l is the wave cycle, C l is the force coefficient of the wave on the unmanned ship; a, b, and c are the angle constant coefficients related to the steering characteristics of the unmanned ship and the motion characteristics in the waves; Step S23: Improve the dynamic model of the unmanned ship. Substitute the formulas that affect the speed of the unmanned ship in three directions in step S22 into formula (1) for mathematical derivation to obtain the improved dynamic model of the unmanned ship. At the same time, in order to facilitate the calculation of the subsequent predictive control collision avoidance strategy and the final collision avoidance path method, the physical parameters of the improved dynamic model are simplified to obtain: Among them, X s It is a vector representation containing the motion state information of the unmanned ship. u, v, r are still the speeds of the unmanned ship in three directions. x, y are the coordinates of the unmanned ship. ψ is the heading angle. f is X s The derivative function of b , Y b 、N b It is the hydrodynamic expression of the unmanned ship in the three directions of sway, swing and pitch after the influence of various forces under the condition of introducing the environmental disturbance of step S22; A X , A Y , is the constant coefficient matrix of the model state variables, which is used to describe the relationship between the state variables and their influence on the motion, and θ is the representation of the evidence formed by extracting the diagonal elements of the required matrix, which is substituted into f to represent the influence of the state variables on the motion state information of the unmanned ship; and D x , D y , D ψ It is a variable related to the position and heading of the unmanned ship. It is calculated through the speed and angle of the unmanned ship and reflects the motion state and direction of the unmanned ship in the plane.

4. The unmanned ship obstacle avoidance path planning method based on global optimization as claimed in claim 1, characterized in that: The step S3 specifically includes: Step S31: discretizing the improved and simplified unmanned ship dynamics model, including: Among them, X s,n is the state vector at the time step, Δt is the time step, and f is still the derivative function; Step S32: The discretization process not only processes the current unmanned ship but also the target ship. The encounter parameters of the current unmanned ship and the target ship are calculated by geometric methods, and the state of the current unmanned ship is set to X. s , the state of the target ship is X T , then the relative position, relative speed, relative speed direction, closest encounter distance DCPA and time to reach the closest encounter point TCPA between the two are expressed by the following set of equations: Among them, Δα is the relative velocity direction, Δv x and Δv y is the component; TCPA can be solved by Come get; Step S33: Calculate the risk of collision based on fuzzy mathematics, and determine the objective function in the collision avoidance strategy, setting the collision risk as F CRI (t), then: Among them, ω1~ω5 are the membership functions of various parameters f DCPA 、f TCPA 、f v , The weight, v C Represents the speed ratio of the two; At the same time, the objective function is set to J, and the actual rudder angle δ is selected as the control variable. The goal is to minimize the predicted collision risk and rudder angle change. The objective function is established as: J=ω risk |F CRI (t)|+ω δ |Dδ|; (10) Among them, ω risk and ω δ are weight coefficients, which are used to adjust the importance of predicting collision risk and rudder angle change in the objective function; |*| is the norm operation; Δδ is the rudder angle change value of this unmanned ship.

5. The unmanned ship obstacle avoidance path planning method based on global optimization as claimed in claim 1, characterized in that: The step S4 specifically includes: Step S41: The actual rudder angle δ is the control variable, the number of initial search control variables is determined, and the maximum number of iterations T is set. max , initial step size h0, step size adjustment factor adj and step size lower limit ε; Step S42: for each individual in the initial search control variable, i.e., the rudder angle sequence, the corresponding objective function value is calculated according to the ship dynamics model and the definition of the objective function; Step S43: iteratively process the calculated objective function value. In each iteration, determine the search direction according to the individual information and the environmental interference force vector in the current control variable, calculate the difference between the objective function value and the surrounding individuals, so as to determine the direction in which the objective function value decreases; and combine the environmental influence vector at the same time. If the angle between the objective function value decreasing direction and the environmental influence vector is small, determine to search in this direction; Step S44: The search step size is adjusted according to the dynamic step size adjustment mechanism. According to the determined search direction and the adjusted step size, a new control variable, i.e., the rudder angle sequence and the objective function set, is generated. When the number of iterations reaches the maximum number of iterations, the algorithm stops searching. At this time, the optimal rudder angle sequence is output as the final solution. This optimal rudder angle sequence will be used for the actual collision avoidance operation of the unmanned ship to minimize the collision risk and the change of the rudder angle, and to achieve safe and efficient collision avoidance.

6. An unmanned ship obstacle avoidance path planning system 100 based on global optimization, characterized in that: The system includes: The unmanned ship dynamics model building module 101 is used to define an inertial coordinate system O O -x o y o and the motion coordinate system of the ship coordinate system O-xy, in which the unmanned ship has corresponding velocities u, v, r in the three directions of sway, swing and pitch, thereby constructing the dynamic model of the unmanned ship; The unmanned ship dynamics model improvement and simplification module 102 is used to improve the unmanned ship dynamics model by considering the influence of water environment factors on the unmanned ship dynamics characteristics, obtain the improved unmanned ship dynamics model and simplify it; The collision avoidance strategy module 103 based on predictive control is used to discretize the improved and simplified unmanned ship dynamics model, calculate the parameters of encountering other ships using geometric methods, calculate the risk of collision based on fuzzy mathematics, and determine the objective function in the collision avoidance strategy; The collision avoidance path planning module 104 is used to find the global optimal solution for the path of the unmanned ship through a loop iteration method.

Images