Restricted water area ship collision avoidance decision-making method and device based on uncertainty modeling

Abstract

The invention discloses a limited water area ship collision avoidance decision-making method and device based on uncertainty modeling. The method comprises the steps that a ship motion model and random variables representing ship maneuverability uncertainty are combined to construct a motion model considering the ship maneuverability uncertainty; gaussian distribution is adopted to construct a ship observation model considering observation uncertainty; gaussian distribution is adopted to simulate the steering and speed change of the ship so as to construct an uncertainty model of the ship obeying rule; capturing the motion uncertainty of other ships during navigation through a depth Gaussian process model, and constructing an other ship trajectory prediction model considering the motion uncertainty; and constructing a ship collision avoidance behavior decision optimization model based on the uncertainty model, and solving the optimal collision avoidance behavior decision of the ship in the sailing process. According to the method, various uncertain factors encountered when the ship sails in a complex and dynamic limited water area are comprehensively considered, and the ship collision avoidance behavior decision optimization model with high adaptability is established, so that ship collision avoidance decision is effectively made.

Description

Technical Field

[0001] The present invention relates to the technical field of ship collision avoidance, and in particular to a ship collision avoidance decision-making method and device in restricted waters based on uncertainty modeling. Background Art

[0002] As the global demand for maritime transport increases, navigation challenges in confined waters are becoming increasingly severe, and collision accidents have become a major safety issue. Data shows that more than 40% of global ship accidents are caused by collisions each year, of which nearly 80% are related to human decision-making errors. In confined waters, environmental conditions such as wind speed, wind direction, visibility and tides change frequently, seriously affecting ship control. In addition, the movement of other ships is full of uncertainty, and equipment failures, route adjustments and human errors make accurate predictions difficult. Sensors that ships rely on (such as radar and GPS) may also lead to misjudgments due to measurement errors or information delays, further increasing the risk of accidents. Therefore, navigation planning and collision avoidance decisions must fully consider these uncertainties to improve safety and efficiency.

[0003] Although existing collision avoidance algorithms have made some progress, the following problems still exist: (1) Ship maneuvers are affected by environmental factors and their own parameters. Many algorithms fail to fully consider these uncertainties, affecting the reliability of decision-making. (2) Most studies rely on linear prediction of the current motion state and fail to accurately capture the complex uncertainties in the ship's motion, resulting in delayed collision avoidance decisions. (3) Many algorithms assume that ships always comply with navigation rules, but in emergency situations, ships may flexibly adjust their behavior, and existing algorithms fail to effectively deal with this situation. Summary of the invention

[0004] The purpose of the present invention is to overcome the above-mentioned defects and problems existing in the prior art, and to provide a method and device for ship collision avoidance decision-making in restricted waters based on uncertainty modeling, which comprehensively considers the various uncertain factors encountered by ships when navigating in complex and dynamic restricted waters, and establishes a highly adaptable ship collision avoidance behavior decision-making optimization model to effectively make ship collision avoidance decisions.

[0005] To achieve the above objectives, the technical solution of the present invention is: a collision avoidance decision method for ships in restricted waters based on uncertainty modeling, comprising:

[0006] The ship motion model is combined with the random variables representing the uncertainty of ship maneuverability to construct a motion model that takes into account the uncertainty of ship maneuverability.

[0007] Gaussian distribution is used to construct a ship observation model that takes observation uncertainty into account;

[0008] Gaussian distribution is used to simulate the ship's turning and speed changes to build an uncertainty model for ship compliance with regulations;

[0009] The motion uncertainty of other ships during navigation is captured through the deep Gaussian process model, and a trajectory prediction model of other ships considering the motion uncertainty is constructed;

[0010] Based on the motion model considering the uncertainty of ship maneuverability, the ship observation model considering the uncertainty of observation, the uncertainty model of ship compliance with rules and the trajectory prediction model of other ships considering the uncertainty of motion, a ship collision avoidance behavior decision optimization model is constructed to solve the optimal collision avoidance behavior decision of the ship during navigation.

[0011] The ship collision avoidance behavior decision optimization model is:

[0012] ;

[0013] ;

[0014] ;

[0015] ;

[0016] ;

[0017] ;

[0018] In the formula, is the objective function; is the control variable for each time step; is the time step; is the minimum distance between the ship and the static obstacle; is the safety distance threshold between the ship and the obstacle; is the minimum distance between the ship and the dynamic obstacle; For the ship The status of the step; For the ship The status of the step; A motion model that takes into account the maneuverability of the ship; For ships in The mean of the control variable of the step; For the control variable The mean of the steps is And the variance is Gaussian distribution of For the control variable The variance of the step; is the objective function of the error between the current state and the target state; is the objective function of the control variable; The objective function is to avoid static obstacles; The objective function for avoiding dynamic obstacles; For ships in The estimated state of the step; is the expected trajectory; is the state weight matrix; for The transpose of is the control variable weight matrix; For the The minimum distance between a mobile ship and a static obstacle; is the number of dynamic obstacles; For ships and obstacles The minimum distance between the predicted positions at the moment.

[0019] The ship motion model is combined with a random variable representing the uncertainty of ship maneuverability to construct a motion model that takes into account the uncertainty of ship maneuverability, including:

[0020] Construct the following ship motion model:

[0021] ;

[0022] ;

[0023] In the formula, for The ship's motion status at all times; , , , , They are The ordinate position, abscissa position, heading angle, speed, and yaw angular velocity of the ship at the moment; is the state transfer matrix; and They are The derivatives of the ship's speed and yaw rate at time instant; is the time interval;

[0024] Define the control variable as ,in, and They are The turning angular speed and rudder angle of the ship at each moment;

[0025] The relationship between the derivative of the ship's yaw angular velocity and the rudder angle is expressed as:

[0026] ;

[0027] In the formula, and They are the ship following index and the ship maneuverability index respectively;

[0028] Derivative of ship speed for:

[0029] ;

[0030] ;

[0031] ;

[0032] In the formula, For the quality of the ship; for The coefficient of for The coefficient of is the fluid mechanics derivative; and propeller thrust coefficient and drag coefficient respectively;

[0033] Control Function for:

[0034] ;

[0035] The control variable Combined with the ship motion model, a motion model that takes into account the ship's maneuverability is constructed:

[0036] ;

[0037] In the formula, is the control input matrix, which represents the influence of the control variables on the ship state;

[0038] The motion model considering the ship's maneuverability is simplified as:

[0039] ;

[0040] Using random variables to represent the uncertainty of ship operation, the motion model considering the uncertainty of ship maneuverability is:

[0041] ;

[0042] In the formula, Indicates the mean And the covariance matrix is Gaussian distribution.

[0043] The ship observation model considering observation uncertainty is:

[0044] ;

[0045] ;

[0046] ;

[0047] In the formula, for The observation status of the ship obtained from AIS at all times; for The ship's motion status at all times; is the observation matrix; Indicates the mean And the covariance matrix is Gaussian distribution.

[0048] The uncertainty model of ship compliance is:

[0049] ;

[0050] In the formula, is the control variable; and are the mean and variance of the ship’s turning angular velocity, respectively; and are the mean and variance of the ship’s rudder angle, respectively.

[0051] The deep Gaussian process model is used to capture the motion uncertainty of other ships during navigation, and a trajectory prediction model of other ships considering the motion uncertainty is constructed, including:

[0052] Define a sliding window and use the mean and variance of the ship trajectory to normalize the ship's longitude and latitude data;

[0053] A deep Gaussian process model including a first-layer Gaussian process model and a second-layer Gaussian process model is constructed; the first-layer Gaussian process model is:

[0054] ;

[0055] ;

[0056] In the formula, is the predicted value of the ship’s trajectory; Ship track point The coordinates of is the first-layer Gaussian process model; is the mean function; is a kernel function used to measure the correlation between input points; is the signal variance; is the length scale; Ship track point The coordinates of

[0057] Given an input trajectory point And the corresponding output , the posterior distribution of the Gaussian process is calculated by Bayes' rule, and the posterior distribution is:

[0058] ;

[0059] In the formula, is a hyperparameter; is the predicted mean; is the predicted covariance matrix;

[0060] Predicted mean for:

[0061] ;

[0062] In the formula, is the covariance matrix of the input trajectory points;

[0063] The predicted covariance matrix for:

[0064] ;

[0065] In the formula, is the newly input trajectory point;

[0066] Hyperparameters It is determined by maximizing the marginal likelihood function, which has the form:

[0067] ;

[0068] In the formula, is the number of samples of input trajectory points;

[0069] The second-layer Gaussian process model uses the predicted output of the first-layer Gaussian process model as input for modeling;

[0070] Calculate the uncertainty estimate of the two-layer Gaussian process model and merge the uncertainties. The combined uncertainty estimate is for:

[0071] ;

[0072] In the formula, The uncertainty of the prediction for the first-layer Gaussian process model; The uncertainty of the prediction for the second-level Gaussian process model;

[0073] On each sliding window, the trajectory of other ships is predicted by the deep Gaussian process model.

[0074] The method of solving the optimal collision avoidance behavior decision of a ship during navigation includes:

[0075] Given the initial control variables, the IPOPT solver is used for iterative optimization to gradually approach the optimal solution of the objective function of the ship collision avoidance behavior decision optimization model.

[0076] In each iteration, the optimization direction is evaluated by calculating the gradient information of the objective function and judging whether the convergence condition meets the convergence criteria;

[0077] When the change of the objective function meets the convergence criteria, the optimal control variables of the current step are output to ensure that the ship can safely avoid obstacles.

[0078] A device for ship collision avoidance decision-making in restricted waters based on uncertainty modeling, which is applied to the above-mentioned method, and comprises:

[0079] The multi-dimensional uncertainty model building module is used to combine the ship motion model with the random variables representing the uncertainty of the ship's maneuverability to build a motion model that takes into account the uncertainty of the ship's maneuverability; Gaussian distribution is used to build a ship observation model that takes into account the observation uncertainty; Gaussian distribution is used to simulate the ship's turning and speed changes to build an uncertainty model for the ship's compliance with the rules; the deep Gaussian process model is used to capture the motion uncertainty of other ships during navigation, and a trajectory prediction model for other ships that takes into account the motion uncertainty is built;

[0080] The ship collision avoidance behavior decision generation module is used to construct a ship collision avoidance behavior decision optimization model based on the motion model considering the uncertainty of ship maneuverability, the ship observation model considering the uncertainty of observation, the uncertainty model of ship compliance with rules and the other ship trajectory prediction model considering the uncertainty of motion, and solve the optimal collision avoidance behavior decision of the ship during navigation.

[0081] A ship collision avoidance decision-making device in restricted waters based on uncertainty modeling, comprising a memory and a processor;

[0082] The memory is used to store computer program code and transmit the computer program code to the processor;

[0083] The processor is used to execute the method according to the instructions in the computer program code.

[0084] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described above is implemented.

[0085] Compared with the prior art, the present invention has the following beneficial effects:

[0086] In the method and device for ship collision avoidance decision-making in restricted waters based on uncertainty modeling of the present invention, a comprehensive multi-dimensional uncertainty model is constructed by quantifying the uncertainty of ship motion characteristics, surrounding environment and other ship behaviors. Compared with the traditional single-factor modeling method, this model can more comprehensively and accurately reflect the various uncertain factors that may be encountered in actual navigation, and improve the accuracy of the decision-making process. Combining the uncertainty model with the multi-objective optimization algorithm can adjust the ship's maneuvering behavior in real time, flexibly optimize the ship's heading and speed based on real-time navigation data and environmental feedback, make the ship's collision avoidance behavior more intelligent and adaptable, and greatly reduce the risk of collision caused by environmental changes. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] Figure 1 The present invention is a flow chart of a method for ship collision avoidance decision-making in restricted waters based on uncertainty modeling.

[0088] Figure 2 Schematic diagram of a ship coordinate system in an embodiment of the present invention.

[0089] Figure 3 It is a schematic diagram of ship avoidance rules in an embodiment of the present invention.

[0090] Figure 4 is a flow chart of a deep Gaussian process in an embodiment of the present invention.

[0091] Figure 5 It is a multi-objective optimization principle diagram in an embodiment of the present invention.

[0092] Figure 6 The present invention is a structural block diagram of a collision avoidance decision-making device for ships in restricted waters based on uncertainty modeling.

[0093] Figure 7 The present invention is a structural block diagram of a ship collision avoidance decision-making device in restricted waters based on uncertainty modeling. DETAILED DESCRIPTION

[0094] The present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0095] See also Figure 1 , a collision avoidance decision method for ships in restricted waters based on uncertainty modeling, comprising:

[0096] S1. Combine the ship motion model with the random variables representing the uncertainty of the ship's maneuverability to construct a motion model that takes into account the uncertainty of the ship's maneuverability. The maneuverability of the ship is constrained by the dynamic equations, especially in confined waters. Considering its shallow water effect, a ship motion model is established, and random variables are introduced to represent the uncertainty in the maneuvering process. These random variables are combined with the dynamic model to establish a ship motion model that includes uncertainty, so that the future navigation status of the ship can be predicted more accurately.

[0097] Gaussian distribution is used to construct a ship observation model that takes into account observation uncertainty. Shore-based monitoring equipment, ship-borne radar, AIS data, and inter-ship communications have certain errors, resulting in certain uncertainties in the ship observation data. Gaussian distribution is used to integrate these uncertainties with the observation model.

[0098] Gaussian distribution is used to simulate the ship's turning and speed changes to build an uncertainty model of the ship's compliance with the rules. In an emergency, the ship may not fully comply with the rules. In order to simulate this behavior, Gaussian distribution is used to describe the ship's turning and speed changes.

[0099] The motion uncertainty of other ships during navigation is captured by the deep Gaussian process model, and a trajectory prediction model of other ships considering the motion uncertainty is constructed. The motion uncertainty of other ships during navigation is captured by constructing a deep Gaussian process model. This model is used to capture the basic characteristics of other ship motion and preliminarily quantify its uncertainty, thereby providing a more accurate trajectory prediction of other ships.

[0100] S2. Based on the motion model considering the uncertainty of ship maneuverability, the ship observation model considering the uncertainty of observation, the uncertainty model of ship compliance with rules and the other ship trajectory prediction model considering the uncertainty of motion, a ship collision avoidance behavior decision optimization model is constructed to solve the optimal collision avoidance behavior decision of the ship during navigation.

[0101] The present invention focuses on considering various uncertain factors in the navigation process. By modeling the uncertainties in ship operation, observation, rule compliance and the movement of other ships, random variables, Gaussian distribution and deep Gaussian processes are introduced respectively to establish a ship motion model, observation model and rule compliance model containing uncertainties, especially to simulate the incomplete rule compliance behavior of ships in emergency situations. Based on these uncertainties, a multi-objective optimization problem is constructed to generate the optimal collision avoidance decision. The optimization objectives include a comprehensive trade-off of energy consumption, navigation economy, safety and collision risk, thereby improving the safety of ships in navigation in restricted waters and their ability to cope with uncertainties.

[0102] Furthermore, the ship motion model is combined with a random variable representing the uncertainty of ship maneuverability to construct a motion model taking into account the uncertainty of ship maneuverability, including:

[0103] The three-degree-of-freedom model is used to model the motion of the ship in three degrees of freedom: surge, sway and yaw. The ship coordinate system is as follows: Figure 2 As shown, the ship at any time The state can be expressed as:

[0104] ;

[0105] The continuous dynamic equation of the ship is converted into discrete form, and the following ship motion model is constructed:

[0106] ;

[0107] In the formula, for The ship's motion status at all times; , , , , They are The ordinate position, abscissa position, heading angle, speed, and yaw angular velocity of the ship at the moment; is the state transfer matrix; and They are The derivative of the ship speed and yaw angular velocity at the moment, in actual operation, is the propeller's rotation angular velocity and rudder angle To change and ; is the time interval;

[0108] Define the control variable as ,in, and They are The turning angular speed and rudder angle of the ship at each moment;

[0109] The MMG equation for a ship is expressed as:

[0110] ;

[0111] In the formula, For the quality of the ship; is the moment of inertia about the Z axis (in the coordinate system of the attached figure, pointing to the center of the earth); , , , , , , , , , is the hydrodynamic derivative of the hull in the fixed coordinate system.

[0112] Separating the pitch motion from the roll and yaw motion, the relationship between the derivative of the ship's yaw angular velocity and the rudder angle is expressed as follows through Laplace transform:

[0113] ;

[0114] In the formula, and They are the ship following index and the ship maneuverability index respectively, which can be calculated through the data of the actual ship Z-shaped steering experiment.

[0115] The above relationship is called the KT maneuvering response model; considering the uncertainty of ship maneuvering, even for the same ship, and It will also change in different environments. Especially in narrow waters, under the influence of shallow water effect, the ship's resistance increases, the turning diameter is larger than in deep water, and the ship's maneuverability is poor; at the same rudder angle, the ship's turning speed decreases.

[0116] The acceleration of the ship is controlled by the propeller thrust and water resistance and can be expressed as:

[0117] ;

[0118] Derivative of ship speed for:

[0119] ;

[0120] ;

[0121] ;

[0122] In the formula, For the quality of the ship; for The coefficient of for The coefficient of is the fluid mechanics derivative; and propeller thrust coefficient and drag coefficient respectively;

[0123] Control Function for:

[0124] ;

[0125] The control variable Combined with the ship motion model, a motion model that takes into account the ship's maneuverability is constructed:

[0126] ;

[0127] In the formula, is the control input matrix, which represents the influence of the control variables on the ship state. It is a diagonal matrix with the last two diagonal elements equal to , other elements are equal to 0;

[0128] The motion model considering the ship's maneuverability is simplified as:

[0129] ;

[0130] Considering that there is a certain uncertainty in ship maneuvering, random variables are used to represent the uncertainty of ship operation. The motion model considering the uncertainty of ship maneuverability is:

[0131] ;

[0132] In the formula, Indicates the mean And the covariance matrix is Gaussian distribution.

[0133] Furthermore, the ship observation model considering observation uncertainty is constructed through the following steps:

[0134] The position of the ship can be observed by the sensor system, and its observation model is expressed as:

[0135] ;

[0136] ;

[0137] Taking into account that sensor equipment such as AIS has certain errors, the observation model of the ship is expressed as:

[0138] ;

[0139] In the formula, for The observation status of the ship obtained from AIS at all times; for The ship's motion status at all times; is the observation matrix; Indicates the mean And the covariance matrix is Gaussian distribution.

[0140] Furthermore, for inland navigation, when two ships meet, except in special circumstances, they should each turn right to avoid each other; when crossing, if there is another ship on the starboard side of the ship, it should give way to the other ship. The specific circumstances are as follows: Figure 3 As shown in the figure, the Gaussian distribution is set to simulate the ship's turning and speed changes to reflect the uncertainty of the ship's compliance with the rules. The uncertainty model of the ship's compliance with the rules is:

[0141] ;

[0142] In the formula, is the control variable; and are the mean and variance of the ship’s turning angular velocity, respectively; and are the mean and variance of the ship’s rudder angle, respectively.

[0143] For the right turn mentioned in the above rules, Set to 0.05 radians, taking into account that ships may violate regulations in emergency situations, the variance It is set to a relatively large value, which is set to 0.1 in this embodiment, to increase the possibility of the ship turning left in an emergency.

[0144] Furthermore, the motion uncertainty of other ships during navigation is captured by the deep Gaussian process model, and a trajectory prediction model of other ships considering the motion uncertainty is constructed. The whole process is as follows: Figure 4 As shown in the figure, the model consists of two layers of Gaussian processes. The first layer is used to capture the basic characteristics of other ship motion and preliminarily quantify its uncertainty. The second layer further captures more complex patterns in the ship's navigation intentions and their corresponding uncertainties by nonlinearly mapping the output of the first layer, thereby providing a more accurate prediction of other ship trajectories. The specific implementation process is as follows:

[0145] Define a sliding window for online modeling, and use the mean and variance of the ship trajectory to normalize the ship's longitude and latitude data;

[0146] A deep Gaussian process model including a first-layer Gaussian process model and a second-layer Gaussian process model is constructed; the first-layer Gaussian process model is:

[0147] ;

[0148] ;

[0149] In the formula, is the predicted value of the ship’s trajectory; Ship track point The coordinates of is a two-dimensional vector; is the first-layer Gaussian process model; is the mean function, usually assumed to be zero; is a kernel function used to measure the correlation between input points; is the signal variance; is the length scale; Ship track point The coordinates of

[0150] Given an input trajectory point And the corresponding output , the posterior distribution of the Gaussian process is calculated by Bayes' rule, and the posterior distribution is:

[0151] ;

[0152] In the formula, is a matrix; is a hyperparameter; is the predicted mean; is the predicted covariance matrix;

[0153] Predicted mean for:

[0154] ;

[0155] In the formula, is the covariance matrix of the input trajectory points;

[0156] The predicted covariance matrix for:

[0157] ;

[0158] In the formula, is the newly input trajectory point;

[0159] Hyperparameters (The length scale of the kernel function and the signal variance) are determined by maximizing the marginal likelihood function (the purpose is to learn the most appropriate hyperparameters , thus obtaining a Gaussian process model that can fit the data well), which is in the form of:

[0160] ;

[0161] In the formula, is the number of samples of input trajectory points.

[0162] The second-layer Gaussian process model uses the predicted output of the first-layer Gaussian process model as input for modeling; by processing the predicted results of the first-layer Gaussian process model, especially the errors and residuals, the second-layer Gaussian process model can capture more details and complex associations in the data, thereby further improving the prediction accuracy and outputting more accurate prediction results and uncertainty estimates.

[0163] Calculate the uncertainty estimates of the two-layer Gaussian process model and merge the uncertainties. The combined uncertainty estimates can provide a credible error range for each point in the trajectory prediction. Combined uncertainty estimates for:

[0164] ;

[0165] In the formula, The uncertainty of the prediction for the first-layer Gaussian process model; The uncertainty of the prediction for the second-level Gaussian process model;

[0166] On each sliding window, the trajectory of the ship is predicted by the deep Gaussian process model and denormalized. This step restores the standardized output to the original coordinate space for better understanding and application of the prediction results. At the same time, the denormalization process also takes into account the prediction uncertainty to ensure the validity of the final result.

[0167] Furthermore, when generating ship collision avoidance behavior decisions, a multi-objective optimization problem is constructed based on the uncertainty of ship maneuvering, observation, rule compliance, and other ship motions, such as Figure 5 As shown in the figure, the aim is to solve the best collision avoidance decision of the ship during navigation. The optimization goal is to reduce energy consumption, improve the economy and safety of navigation, and reduce the risk of collision. The specific implementation process is:

[0168] Path tracking error. By calculating the error between the current state of the ship and the target state and taking a weighted sum of them, the ship is made to move along the desired path as much as possible. Specifically, the state error is added to the objective function in a quadratic form:

[0169] ;

[0170] Control input energy consumption. In order to minimize the energy of the control variables for maneuvering the ship, the cost term of the control variables is added to the objective function, which is also expressed in quadratic form:

[0171] ;

[0172] Obstacle avoidance cost. To ensure that the ship avoids static obstacles, an additional penalty will be imposed when the ship is not within a safe distance from the obstacle. The specific calculation is as follows:

[0173] ;

[0174] Dynamic obstacles mainly refer to other ships. The dynamic obstacle avoidance cost is achieved by predicting the possible navigation positions of these ships by considering the uncertainty of navigation and calculating the minimum distance between the own ship and these predicted positions. The calculation method of this avoidance cost is similar to that of static obstacles, but it is necessary to consider the predicted positions of obstacles at different time points:

[0175] ;

[0176] The objective function combines path tracking, energy consumption control, static obstacle avoidance, and dynamic obstacle avoidance:

[0177] ;

[0178] Combining the above constraints on ship dynamics, rules and obstacle avoidance, the optimization problem is:

[0179] ;

[0180] In the above formula, is the objective function; is the control variable for each time step; is the time step; is the minimum distance between the ship and the static obstacle; is the safety distance threshold between the ship and the obstacle; is the minimum distance between the ship and the dynamic obstacle; For the ship The status of the step; For the ship The status of the step; A motion model that takes into account the maneuverability of the ship; For ships in The mean of the control variable of the step; For the control variable The mean of the steps is And the variance is Gaussian distribution of For the control variable The variance of the step; is the objective function of the error between the current state and the target state; is the objective function of the control variable; The objective function is to avoid static obstacles; The objective function for avoiding dynamic obstacles; For ships in The estimated state of the first step is the estimated value of the ship obtained by the Kalman filter algorithm using the motion model considering the uncertainty of the ship's maneuverability and the ship observation model considering the uncertainty of observation; is the expected trajectory; is the state weight matrix; for The transpose of is the control variable weight matrix; For the The minimum distance between a mobile ship and a static obstacle; is the number of dynamic obstacles; For ships and obstacles The minimum distance between the predicted positions at the moment.

[0181] Furthermore, the method of solving the optimal collision avoidance behavior decision of the ship during navigation includes:

[0182] Given the initial control variables, the IPOPT solver is used for iterative optimization to gradually approach the optimal solution of the objective function of the ship collision avoidance behavior decision optimization model. In each iteration, the IPOPT solver updates the control variables, calculates the corresponding ship state, and gradually approaches the optimal solution through the gradient descent method to ensure the optimal strategy for collision avoidance behavior.

[0183] In each iteration, the IPOPT solver evaluates the optimization direction by calculating the gradient information of the objective function (such as path tracking error, obstacle avoidance cost), and determines whether the convergence condition meets the convergence criteria;

[0184] When the change of the objective function meets the convergence criteria, the optimal control variables of the current step are output, including the ship's direction, acceleration and other specific maneuvers to ensure that the ship safely avoids obstacles.

[0185] The present invention establishes a highly adaptable model by comprehensively considering the uncertainties encountered by ships when navigating in complex and dynamically restricted waters, so as to make ship collision avoidance decisions more effectively. First, these uncertainty factors are quantified through data analysis and modeling technology, and corresponding mathematical models are constructed. These models not only take into account the motion characteristics of the ship itself, but also take into account the surrounding environment and the behavior of other ships to form a comprehensive multi-dimensional uncertainty model. Then, the uncertainty model is combined with multi-objective optimization, and multiple optimization goals are set. Through the optimization algorithm, the ship's maneuvering behavior is flexibly adjusted according to real-time navigation data and environmental feedback. The model of the present invention has strong robustness and can cope with different types of uncertainties in complex and dynamic navigation environments, especially in restricted waters, to ensure that ships can make appropriate collision avoidance decisions, avoiding the limitations of traditional methods when facing a changing environment.

[0186] See also Figure 6 The present invention also provides a device for ship collision avoidance decision-making in restricted waters based on uncertainty modeling, which is applied to the above-mentioned method for ship collision avoidance decision-making in restricted waters based on uncertainty modeling, and the device comprises:

[0187] The multi-dimensional uncertainty model building module is used to combine the ship motion model with the random variables representing the uncertainty of the ship's maneuverability to build a motion model that takes into account the uncertainty of the ship's maneuverability; Gaussian distribution is used to build a ship observation model that takes into account the observation uncertainty; Gaussian distribution is used to simulate the ship's turning and speed changes to build an uncertainty model for the ship's compliance with the rules; the deep Gaussian process model is used to capture the motion uncertainty of other ships during navigation, and a trajectory prediction model for other ships that takes into account the motion uncertainty is built;

[0188] The ship collision avoidance behavior decision generation module is used to construct a ship collision avoidance behavior decision optimization model based on the motion model considering the uncertainty of ship maneuverability, the ship observation model considering the uncertainty of observation, the uncertainty model of ship compliance with rules and the other ship trajectory prediction model considering the uncertainty of motion, and solve the optimal collision avoidance behavior decision of the ship during navigation.

[0189] See also Figure 7 ,The present invention also provides a ship collision avoidance decision-making device in restricted waters based on uncertainty modeling, including a memory and a processor;

[0190] The memory is used to store computer program code and transmit the computer program code to the processor;

[0191] The processor is used to execute the above-mentioned uncertainty modeling restricted waters ship collision avoidance decision method according to the instructions in the computer program code.

[0192] The present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method for ship collision avoidance decision-making in restricted waters based on uncertainty modeling is implemented.

[0193] Generally speaking, the computer instructions for implementing the method of the present invention may be carried in any combination of one or more computer-readable storage media. Non-transitory computer-readable storage media may include any computer-readable media, except for the signal itself that is temporarily propagating.

[0194] Computer readable storage media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or components, or any combination thereof. More specific examples (non-exhaustive list) of computer readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EKROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or device.

[0195] Computer program code for performing the operation of the present invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages, in particular, Python suitable for neural network computing and platform frameworks based on TensorFlow, PyTorch, etc. can be used. The program code can be executed entirely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer or to an external computer (for example, using an Internet service provider to connect via the Internet) through any type of network, including a local area network (LAN) or a wide area network (WAN).

[0196] The above-mentioned device and non-temporary computer-readable storage medium can be referred to the specific description of a method for ship collision avoidance decision-making in restricted waters based on uncertainty modeling and its beneficial effects, which will not be repeated here.

[0197] Although the embodiments of the present invention have been shown and described above, it should be understood that the above embodiments are exemplary and should not be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A ship collision avoidance decision method in restricted waters based on uncertainty modeling, characterized in that: include: The ship motion model is combined with the random variables representing the uncertainty of ship maneuverability to construct a motion model that takes into account the uncertainty of ship maneuverability. Gaussian distribution is used to construct a ship observation model that takes observation uncertainty into account; Gaussian distribution is used to simulate the ship's turning and speed changes to build an uncertainty model for ship compliance with regulations; The motion uncertainty of other ships during navigation is captured through the deep Gaussian process model, and a trajectory prediction model of other ships considering the motion uncertainty is constructed; Based on the motion model considering the uncertainty of ship maneuverability, the ship observation model considering the uncertainty of observation, the uncertainty model of ship compliance with rules and the trajectory prediction model of other ships considering the uncertainty of motion, a ship collision avoidance behavior decision optimization model is constructed to solve the optimal collision avoidance behavior decision of the ship during navigation.

2. The method for ship collision avoidance decision-making in restricted waters based on uncertainty modeling according to claim 1 is characterized in that: The ship collision avoidance behavior decision optimization model is: ； ； ； ； ； ； In the formula, is the objective function; is the control variable for each time step; is the time step; is the minimum distance between the ship and the static obstacle; is the safety distance threshold between the ship and the obstacle; is the minimum distance between the ship and the dynamic obstacle; For the ship The status of the step; For the ship The status of the step; A motion model that takes into account the maneuverability of the ship; For ships in The mean of the control variable of the step; For the control variable The mean of the steps is And the variance is Gaussian distribution of For the control variable The variance of the step; is the objective function of the error between the current state and the target state; is the objective function of the control variable; The objective function is to avoid static obstacles; The objective function for avoiding dynamic obstacles; For ships in The estimated state of the step; is the expected trajectory; is the state weight matrix; for The transpose of is the control variable weight matrix; For the The minimum distance between a mobile ship and a static obstacle; is the number of dynamic obstacles; For ships and obstacles The minimum distance between the predicted positions at the moment.

3. The method for ship collision avoidance decision-making in restricted waters based on uncertainty modeling according to claim 1 is characterized in that: The ship motion model is combined with a random variable representing the uncertainty of ship maneuverability to construct a motion model that takes into account the uncertainty of ship maneuverability, including: Construct the following ship motion model: ； ； In the formula, for The motion status of the ship at all times; , , , , They are The ordinate position, abscissa position, heading angle, speed, and yaw angular velocity of the ship at the moment; is the state transfer matrix; and They are The derivatives of the ship's speed and yaw rate at time instant; is the time interval; Define the control variable as ,in, and They are The turning angular speed and rudder angle of the ship at each moment; The relationship between the derivative of the ship's yaw angular velocity and the rudder angle is expressed as: ； In the formula, and They are the ship following index and the ship maneuverability index respectively; Derivative of ship speed for: ； ； ； In the formula, For the quality of the ship; for The coefficient of for The coefficient of is the fluid mechanics derivative; and propeller thrust coefficient and drag coefficient respectively; Control Function for: ； The control variable Combined with the ship motion model, a motion model that takes into account the ship's maneuverability is constructed: ； In the formula, is the control input matrix, which represents the influence of the control variables on the ship state; The motion model considering the ship's maneuverability is simplified as: ； Using random variables to represent the uncertainty of ship operation, the motion model considering the uncertainty of ship maneuverability is: ； In the formula, Indicates the mean And the covariance matrix is Gaussian distribution.

4. The method for ship collision avoidance decision-making in restricted waters based on uncertainty modeling according to claim 1 is characterized in that: The ship observation model considering observation uncertainty is: ； ； ； In the formula, for The observation status of the ship obtained from AIS at all times; for The motion status of the ship at all times; is the observation matrix; It means the mean And the covariance matrix is Gaussian distribution.

5. The method for ship collision avoidance decision-making in restricted waters based on uncertainty modeling according to claim 1 is characterized in that: The uncertainty model of ship compliance is: ； In the formula, is the control variable; and are the mean and variance of the ship’s turning angular velocity, respectively; and are the mean and variance of the ship’s rudder angle, respectively.

6. The method for ship collision avoidance decision-making in restricted waters based on uncertainty modeling according to claim 1 is characterized in that: The deep Gaussian process model is used to capture the motion uncertainty of other ships during navigation, and a trajectory prediction model of other ships considering the motion uncertainty is constructed, including: Define a sliding window and use the mean and variance of the ship trajectory to normalize the ship's longitude and latitude data; A deep Gaussian process model including a first-layer Gaussian process model and a second-layer Gaussian process model is constructed; the first-layer Gaussian process model is: ； ； In the formula, is the predicted value of the ship’s trajectory; Ship track point The coordinates of is the first-layer Gaussian process model; is the mean function; is a kernel function used to measure the correlation between input points; is the signal variance; is the length scale; Ship track point The coordinates of Given an input trajectory point And the corresponding output , the posterior distribution of the Gaussian process is calculated by Bayes' rule, and the posterior distribution is: ； In the formula, is a hyperparameter; is the predicted mean; is the predicted covariance matrix; Predicted mean for: ； In the formula, is the covariance matrix of the input trajectory points; The predicted covariance matrix for: ； In the formula, is the newly input trajectory point; Hyperparameters It is determined by maximizing the marginal likelihood function, which has the form: ； In the formula, is the number of samples of input trajectory points; The second-layer Gaussian process model uses the predicted output of the first-layer Gaussian process model as input for modeling; Calculate the uncertainty estimate of the two-layer Gaussian process model and merge the uncertainties. The combined uncertainty estimate is for: ； In the formula, The uncertainty of the prediction for the first-layer Gaussian process model; The uncertainty of the prediction for the second-level Gaussian process model; On each sliding window, the trajectory of other ships is predicted by the deep Gaussian process model.

7. The method for ship collision avoidance decision-making in restricted waters based on uncertainty modeling according to claim 1 is characterized in that: The method of solving the optimal collision avoidance behavior decision of a ship during navigation includes: Given the initial control variables, the IPOPT solver is used for iterative optimization to gradually approach the optimal solution of the objective function of the ship collision avoidance behavior decision optimization model. In each iteration, the optimization direction is evaluated by calculating the gradient information of the objective function and judging whether the convergence condition meets the convergence criteria; When the change of the objective function meets the convergence criteria, the optimal control variables of the current step are output to ensure that the ship can safely avoid obstacles.

8. A ship collision avoidance decision-making device in restricted waters based on uncertainty modeling, characterized in that: The device is applied to the method described in any one of claims 1 to 7, and the device comprises: The multi-dimensional uncertainty model building module is used to combine the ship motion model with the random variables representing the uncertainty of the ship's maneuverability to build a motion model that takes into account the uncertainty of the ship's maneuverability; Gaussian distribution is used to build a ship observation model that takes into account the observation uncertainty; Gaussian distribution is used to simulate the ship's turning and speed changes to build an uncertainty model for the ship's compliance with the rules; the deep Gaussian process model is used to capture the motion uncertainty of other ships during navigation, and a trajectory prediction model for other ships that takes into account the motion uncertainty is built; The ship collision avoidance behavior decision generation module is used to construct a ship collision avoidance behavior decision optimization model based on the motion model considering the uncertainty of ship maneuverability, the ship observation model considering the uncertainty of observation, the uncertainty model of ship compliance with rules and the other ship trajectory prediction model considering the uncertainty of motion, and solve the optimal collision avoidance behavior decision of the ship during navigation.

9. A collision avoidance decision-making device for ships in restricted waters based on uncertainty modeling, characterized in that: including memory and processor; The memory is used to store computer program code and transmit the computer program code to the processor; The processor is configured to execute the method according to any one of claims 1 to 7 according to instructions in the computer program code.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images