A ship route planning auxiliary method based on intelligent algorithm and optimization theory

Abstract

The present invention discloses a ship route planning auxiliary method based on intelligent algorithm and optimization theory, which specifically relates to the field of meteorological navigation, including S1: real-time perception of meteorological and ocean information, S2: analysis of ship characteristics, S3: intelligent fusion algorithm, S4: generation of dynamic ship route planning, S5: evaluation of multi-dimensional risks, S6: optimization of ship route planning and S7: human-computer interaction. A ship route planning auxiliary method based on intelligent algorithm and optimization theory integrates improved genetic algorithm and particle swarm optimization algorithm to obtain a ship route planning auxiliary model, thereby realizing personalized ship route planning; by obtaining a motion strategy based on first and second state characteristics and a route planning auxiliary model, the dynamic and real-time performance of ship route planning is improved; by real-time local adjustment of the motion strategy through the third state characteristic constraint, the comprehensive performance of route planning is improved, better route planning is realized, and unnecessary resource consumption is reduced.

Description

Technical Field

[0001] The present invention relates to the field of meteorological navigation technology, and more specifically, to a ship route planning auxiliary method based on intelligent algorithm and optimization theory. Background Art

[0002] Driven by the wave of global economic development, global trade is becoming more frequent. As an important link in international trade, the scale of maritime transport is also expanding continuously. The realization of safe and efficient navigation route planning has become a key issue that needs to be urgently addressed in the field of meteorological navigation technology. Accurate route planning is a solid guarantee for the safety of ship navigation, and can also play a key role in reducing fuel consumption and operating costs, thereby greatly improving the economic benefits and market competitiveness of shipping companies.

[0003] At present, route planning technology mainly relies on traditional meteorological and oceanographic prediction models and simple path planning algorithms. By collecting meteorological and oceanographic forecast data such as wind speed, wind direction, waves, and currents, and combining them with the basic performance parameters of the ship, an algorithm with established rules is used to preliminarily plan a theoretically relatively safe route.

[0004] However, there are still some shortcomings in its actual use. For example, the traditional fixed rule algorithm lacks flexibility. Faced with rapidly changing meteorological and oceanic conditions, especially sudden severe weather such as hurricanes and severe convection, the pre-planned route is likely to lose safety in an instant and cannot be adjusted in time. Traditional route planning technology lacks in-depth analysis of the individual characteristics of ships and cannot achieve personalized route planning, resulting in low resource utilization and failure to give full play to the optimal performance of the ship. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides a ship route planning auxiliary method based on intelligent algorithm and optimization theory, and solves the problems raised in the above-mentioned background technology through the following scheme.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A ship route planning auxiliary method based on intelligent algorithm and optimization theory, comprising:

[0008] S1: Real-time perception of meteorological and oceanographic information: Obtain environmental information of the target ship's working conditions through a group of meteorological and oceanographic sensors to generate the first state feature;

[0009] S2: Analyze ship characteristics: synchronize and collect the actual operating data of the target ship's working conditions to generate the second state characteristics;

[0010] S3: Intelligent fusion algorithm: Based on the historical status characteristics of the target ship, an intelligent algorithm fusion framework is constructed to obtain a route planning auxiliary model. The intelligent algorithm fusion framework integrates the improved genetic algorithm and particle swarm optimization algorithm, and introduces dynamic inertia weight and adaptive mutation strategy;

[0011] S4: Generate dynamic ship route planning: Based on the first state characteristics and the second state characteristics corresponding to the target ship, the probability distribution of the target ship's movement direction is analyzed through the route planning auxiliary model to obtain the target ship's movement strategy;

[0012] The motion strategy of the target ship is based on the operating state equation of the target ship in the time dimension and the artificial potential field corresponding to the target ship in the space dimension. The target motion strategy is formed through the state-potential field coupling relationship. The target motion strategy consists of the operating state and the artificial potential field. The operating state is the operating state equation of the target ship in unit time. The operating state equation of the target ship includes the longitudinal motion state equation , lateral motion state equation And the vertical motion equation of state , the vertical motion state equation, i.e., the change of the heading angle of the target ship with time;

[0013] The collision avoidance planning is performed by the artificial potential field method, the environment of the target ship per unit range is regarded as the artificial potential field, and the artificial potential field corresponding to the target ship per unit range is calculated according to the running state and the first state characteristics corresponding to the running state equation of the target ship in unit time; the artificial potential field includes the gravitational potential field and repulsive potential field , the gravitational potential field is used to guide the target ship to move towards the target point, and the repulsive potential field is used to guide the ship to avoid obstacles;

[0014] S5: Evaluate multi-dimensional risk: obtain a multi-objective reward function, perform risk assessment on the target ship's motion strategy based on the multi-objective reward function, and obtain the third state feature;

[0015] S6: Optimize ship path planning: Based on the constraints of the third state characteristics, the movement strategy of the target ship is locally adjusted in real time to obtain the route optimization result corresponding to the movement strategy of the target ship;

[0016] The constraint of the third state feature is based on the motion trajectory of the target ship. and control trajectory , Expressed as planning time horizon, based on navigation safety objective weights , navigation efficiency target weight and weight of navigation economic objectives ,and , the multi-objective reward values ​​corresponding to the third state feature are accumulated to the maximum extent, which can be specifically expressed as:

[0017]

[0018] in, Expressed as the number of time steps, is represented as an index of time steps, The value ranges from 0 to , Expressed as time Reward for safe navigation Expressed as time The navigation efficiency bonus is Expressed as time The economic reward for sailing at that time, Expressed as time The target ship's motion trajectory, including longitudinal position, lateral position and heading angle, Expressed as time The control trajectory of the target ship, including the gravitational potential field and the repulsive potential field;

[0019] S7: Human-computer interaction: Output the route optimization results corresponding to the target ship's motion strategy in the form of a data structure to generate a visualization result.

[0020] Preferably, the steps of constructing the S3, intelligent algorithm fusion framework are as follows:

[0021] Initialize the population: Generate an initial population randomly through a genetic algorithm, and perform particle swarm optimization on the generated initial population. Each particle individual is represented as a route planning scheme, and the particle individual is a route planning candidate scheme in the initial population after the genetic algorithm and the particle swarm optimization algorithm.

[0022] Fitness evaluation: Calculate the fitness value of each individual particle based on the optimal position, speed of each individual particle and the global optimal position and speed;

[0023] Select individual particles: dynamically adjust the inertia weight and use the tournament selection method to select excellent individual particles to enter the next generation;

[0024] Crossover operation: Perform crossover operation on the selected excellent particle individuals to generate new particle individuals;

[0025] Mutation: Adopt an adaptive mutation strategy, adaptively adjust the mutation probability based on the fitness value, and mutate the individual particles after crossover.

[0026] Preferably, the step S4, obtaining the movement strategy of the target ship, specifically includes:

[0027] The motion strategies of multiple target ships are obtained, wherein the motion strategies of the multiple target ships include a target motion strategy, and the target motion strategy is composed of any one of the operating states corresponding to multiple unit times and an artificial potential field corresponding to the operating state in the artificial potential fields corresponding to multiple unit ranges.

[0028] Preferably, the longitudinal motion state equation corresponding to the running state equation of the target ship in S4 is specifically expressed as:

[0029]

[0030] in, It is expressed as the motion state equation of the target ship's longitudinal position in the geographic coordinate system changing with time, Represented as the initial longitude position of the target ship, Expressed as the longitudinal velocity of the target ship, is expressed as the yaw angular velocity of the target ship, Expressed as a unit of time, Expressed as the lateral velocity of the target ship, Represented as the initial heading angle of the target ship.

[0031] Preferably, the lateral motion state equation corresponding to the running state equation of the target ship in S4 is specifically expressed as:

[0032]

[0033] in, It is expressed as the motion state equation of the target ship's lateral position in the geographic coordinate system changing with time, is represented as the initial latitude position of the target ship, Expressed as the longitudinal velocity of the target ship, is expressed as the yaw angular velocity of the target ship, Expressed as a unit of time, Represented as the initial heading angle of the target ship.

[0034] Preferably, the vertical motion state equation corresponding to the running state equation of the target ship in S4 is specifically expressed as:

[0035]

[0036] in, It is expressed as the motion state equation of the target ship's vertical position in the geographic coordinate system changing with time, that is, the change of the target ship's heading angle with time, It is represented as the initial heading angle of the target ship, is expressed as the yaw angular velocity of the target ship, Expressed as a unit of time.

[0037] Preferably, the S4 is a gravitational potential field that guides the target ship to move toward the target point. , specifically expressed as:

[0038]

[0039] in, Expressed as the position vector of the target ship in the geographic coordinate system, Represents the current real-time speed of the target ship. and is expressed as the gravitational coefficient, Represented as the target ship position To the destination The distance It is expressed as the limit of the gravitational range, It is expressed as the target speed of the target ship;

[0040] Repulsive potential field that guides ships to avoid obstacles , specifically expressed as:

[0041]

[0042] in, Expressed as the position vector of the target ship in the geographic coordinate system, Expressed as the repulsion coefficient, Represents the influence range of the obstacle. Represented as ship position To obstacles distance.

[0043] Preferably, the step S5, obtaining the third state feature, specifically includes:

[0044] The optimization target set corresponding to the target motion strategy is obtained, and the optimization target set includes navigation safety target, navigation efficiency target and navigation economic target.

[0045] Technical effects and advantages of the present invention:

[0046] 1. The present invention integrates improved genetic algorithm and particle swarm optimization algorithm to obtain the route planning auxiliary model of the ship under different working conditions, improves the in-depth analysis capability of the individual characteristics of the ship, realizes personalized route planning, improves resource utilization, and gives full play to the best performance of the ship;

[0047] 2. The present invention obtains the motion strategy based on the first and second state characteristics and the route planning auxiliary model, thereby improving the dynamics and real-time performance of route planning, more accurately predicting the ship's operating status, effectively avoiding collisions, ensuring the safety of ship navigation, and further optimizing route planning;

[0048] 3. The present invention improves the comprehensive performance of route planning by making real-time local adjustments to the motion strategy based on the third state feature constraints, achieves better route planning, and reduces unnecessary resource consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a diagram of the method steps of the present invention.

[0050] Figure 2 The figure is a flow chart of the method of the present invention.

[0051] Figure 3 It is a schematic diagram of the structure of the method of the present invention. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification of the present application, the singular expressions "one", "a kind of", "said", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to and includes any or all possible combinations of one or more listed items.

[0054] In the following, the terms "first", "second", and "third" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", and "third" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.

[0055] As attached Figure 1A ship route planning auxiliary method based on intelligent algorithm and optimization theory is shown, including S1: real-time perception of meteorological and ocean information, S2: analysis of ship characteristics, S3: intelligent fusion algorithm, S4: generation of dynamic ship route planning, S5: evaluation of multi-dimensional risks, S6: optimization of ship path planning, and S7: human-computer interaction.

[0056] S1: Real-time perception of meteorological and oceanographic information: Obtain environmental information of the target ship's working conditions through a group of meteorological and oceanographic sensors to generate the first state characteristics.

[0057] Specifically, the environmental information of the working conditions of the target ship is obtained through the meteorological and ocean sensor group, which includes but is not limited to wind speed sensor, wind direction sensor, temperature sensor, humidity sensor, air pressure sensor, geomagnetic sensor, radar sensor, wave sensor, ocean current sensor and camera, etc.; among them, the wind speed sensor and wind direction sensor are installed on the top of the ship's mast to obtain unobstructed wind environment data; the temperature sensor, humidity sensor and air pressure sensor are installed on the bridge and the meteorological observation room to monitor the temperature, humidity and air pressure of the atmosphere around the ship; the geomagnetic sensor is used to measure the geomagnetic information of the ship's location; the radar sensor is installed at a high place on the ship to detect surrounding targets; the camera is installed on both sides and the front of the ship to obtain visual images in real time; the wave sensor is used to measure the wave information of the ship's location; the current sensor is used to measure the ocean current information of the ship's location; the data from different sensors are synchronized in time and spatially aligned, and fused to generate the first state feature.

[0058] It should be noted that the first state characteristics include real-time meteorological and oceanographic data and dynamic environmental information, wherein the real-time meteorological and oceanographic data include the wind speed, wind direction, air pressure, temperature, humidity, waves and currents of the target ship; the dynamic environmental information includes the types of dangerous areas and obstacles and their corresponding locations.

[0059] S2: Analyze ship characteristics: synchronize and collect the actual operating data of the target ship's working conditions to generate the second state characteristics.

[0060] Specifically, the performance of the corresponding power system of the target ship under different working conditions is obtained, including but not limited to rated power, torque curve, etc. The rated power of the power system and the torque value at different speeds are recorded, the torque curve is drawn, and during the navigation of the target ship, the engine output power, speed and torque are recorded in real time by using an engine monitor, and the power performance data is obtained at different cargo capacities and different speeds; the maximum cargo capacity and minimum weight requirements during ballast navigation are obtained from the load line certificate and the stability report of the target ship; when the ship is sailing in a straight line at a preset speed, a turning command is issued instantaneously, and a gyroscope is used to record the time from issuing the command to the actual start of the ship to change its course, and multiple tests are carried out at different speeds and loading conditions to obtain the steering response time under different working conditions; In open seas, steering tests are carried out at different speeds and rudder angles. The AIS system is used to record the ship's steering trajectory and calculate the ship's minimum turning radius under different working conditions. Accelerometers and gyroscopes are installed at the bow, stern, and midship of the target ship. The accelerometer is used to measure the acceleration of the hull in the longitudinal and horizontal directions, so as to calculate the swing amplitude and vibration frequency. The gyroscope is used to measure the attitude changes of the hull, including roll, pitch and bow roll angles. A Doppler log and AIS system are used to record the speed in real time during the ship's navigation, and multiple navigation tests are carried out at different cargo capacities to record the speed changes at the same engine speed. In open seas, a speed sensor is used to record the braking distance and time of the ship decelerating from the preset speed to a standstill, so as to evaluate the acceleration and deceleration performance.

[0061] It should be noted that the second state characteristics include hull motion data, speed and acceleration / deceleration performance data, steering performance data, power performance data of the target ship, and load capacity data, wherein the hull motion data include the heading angle, longitude position and latitude position of the target ship, the speed and acceleration / deceleration performance data include the bow angular velocity, longitudinal speed, swing amplitude and vibration frequency of the target ship, the steering performance data include the steering response time and minimum turning radius under different working conditions, the power performance data of the target ship include the rated power of the target ship and the corresponding torque curve of the target ship under different working conditions, and the load capacity data include the maximum cargo capacity and minimum weight requirement during ballasted navigation.

[0062] S3: Intelligent fusion algorithm: Based on the historical status characteristics corresponding to the target ship, an intelligent algorithm fusion framework is constructed to obtain a route planning auxiliary model. The intelligent algorithm fusion framework integrates the improved genetic algorithm and particle swarm optimization algorithm, and introduces dynamic inertia weight and adaptive mutation strategy.

[0063] Specifically, after obtaining the historical state characteristics corresponding to the target ship, the route planning auxiliary model corresponding to the optimal parameters can be obtained through the intelligent algorithm fusion framework. The construction steps of the intelligent algorithm fusion framework are as follows: Initialize the population: randomly generate the initial population through the genetic algorithm, and perform particle swarm optimization initialization on the generated initial population. Each particle individual is represented as a route planning scheme, and the particle individual is a route planning candidate scheme in the initial population that has been through the genetic algorithm and the particle swarm optimization algorithm; Fitness evaluation: based on the optimal position, speed and global optimal position and speed of each particle individual, calculate the fitness value of each particle individual; Select particle individuals: dynamically adjust the inertia weight, and use the tournament selection method to select excellent particle individuals to enter the next generation; in this embodiment, in Among the particle individuals, the particle individual with the largest fitness value is selected as the excellent particle individual, among which, It only indicates the number of individual particles in the selection stage; Crossover operation: perform crossover operation on the selected excellent individual particles to generate new individual particles; Mutation: adopt an adaptive mutation strategy, adaptively adjust the mutation probability based on the fitness value, and mutate the individual particles after crossover.

[0064] Furthermore, the steps of obtaining the route planning auxiliary model are as follows:

[0065] S301: Collect historical meteorological and oceanographic data from multiple data sources, including but not limited to global meteorological and oceanographic observation data for many years, satellite remote sensing data, etc.; collect actual track and status information of ships during historical voyages;

[0066] S302: Divide the data into a training set, a validation set, and a test set in a ratio of 70%, 15%, and 15%;

[0067] S303: Use a convolutional neural network to build a route planning auxiliary model;

[0068] Furthermore, a network structure with multiple convolutional layers, pooling layers, and fully connected layers is constructed; the model parameters are adjusted using the back propagation algorithm through multiple iterative training of the training set data; regularization technology is used to prevent overfitting, and the performance of the model is tested in real time through the validation set to adjust the hyperparameters;

[0069] S304: defining a reinforcement learning environment, clarifying the ship state space and action space, and setting a reward function; the ship state space includes the position, speed, and heading of the ship; the action space includes adjusting the speed, changing the heading, and adjusting the power output;

[0070] S305: The strategy generated by reinforcement learning is used as the initial solution, input into the intelligent algorithm fusion framework, and the route planning auxiliary model corresponding to the optimal parameters is obtained.

[0071] S4: Generate dynamic ship route planning: Based on the first state characteristics and the second state characteristics corresponding to the target ship, the probability distribution of the target ship's movement direction is analyzed through the route planning auxiliary model to obtain the target ship's movement strategy.

[0072] Specifically, the motion strategy of the target ship is based on the operating state equation of the target ship in the time dimension and the artificial potential field corresponding to the target ship in the space dimension. The target motion strategy is formed through the state-potential field coupling relationship. The target motion strategy consists of the operating state and the artificial potential field. The operating state is the operating state equation of the target ship in unit time, and the artificial potential field includes the gravitational potential field. and repulsive potential field , the gravitational potential field is used to guide the target ship to move towards the target point, and the repulsive potential field is used to guide the ship to avoid obstacles.

[0073] Furthermore, the route planning auxiliary model is used to predict the operating state of the target ship within a unit time based on the second state characteristics; the artificial potential field of the target ship per unit range is calculated based on the predicted operating state of the target ship within the unit time and the first state characteristics; the motion strategies of multiple target ships are obtained, and the motion strategies of the multiple target ships include a target motion strategy, which is composed of any one operating state among the operating states corresponding to multiple unit times and an artificial potential field corresponding to the operating state in the artificial potential fields corresponding to multiple unit ranges.

[0074] In a possible implementation, obtaining the motion strategy of the target ship includes: based on real-time monitoring of the second state feature of the target ship, obtaining the operating state equation of the target ship in unit time, the operating state equation of the target ship includes the longitudinal motion state equation , lateral motion state equation And the vertical motion equation of state .

[0075] Furthermore, the longitudinal motion state equation corresponding to the running state equation of the target ship is specifically expressed as:

[0076]

[0077] in, It is expressed as the motion state equation of the target ship's longitudinal position in the geographic coordinate system changing with time, is represented as the initial longitude position of the target ship, Expressed as the longitudinal velocity of the target ship, is expressed as the yaw angular velocity of the target ship, Expressed as a unit of time, Expressed as the lateral velocity of the target ship, It is represented as the initial heading angle of the target ship. It should be noted that the initial longitude position of the target ship is the real-time position at time t.

[0078] The lateral motion state equation corresponding to the target ship's operating state equation is specifically expressed as:

[0079]

[0080] in, It is expressed as the motion state equation of the target ship's lateral position in the geographic coordinate system changing with time, is represented as the initial latitude position of the target ship, Expressed as the longitudinal velocity of the target ship, is expressed as the yaw angular velocity of the target ship, Expressed as a unit of time, It is represented as the initial heading angle of the target ship;

[0081] The vertical motion state equation corresponding to the target ship's operating state equation is specifically expressed as:

[0082]

[0083] in, It is expressed as the motion state equation of the target ship's vertical position in the geographic coordinate system changing with time, that is, the change of the target ship's heading angle with time, It is represented as the initial heading angle of the target ship, is expressed as the yaw angular velocity of the target ship, Expressed as a unit of time.

[0084] It should be noted that the longitudinal motion state equation describes the change of the longitudinal position of the target ship in the geographic coordinate system over time. The longitudinal motion state equation can be used to understand the change of the position of the ship in the longitude direction in real time, so as to accurately determine whether the ship is moving along the predetermined route;

[0085] In this embodiment, when the target ship is sailing across the ocean, the longitudinal position is mastered to help the target ship avoid dangerous waters; when the ship is berthing, a reference basis is provided for the precise docking of the ship.

[0086] It should be noted that the lateral motion state equation reflects the change of the lateral position of the target ship in the geographic coordinate system over time, providing a basis for the ship to avoid obstacles and maintain navigation safety;

[0087] In this embodiment, when the target ship is in a narrow waterway, it avoids collision with other ships, obstacles and the shore by monitoring the lateral position, and adjusts the course and speed in real time to ensure that a safe distance is maintained in the lateral direction.

[0088] In this embodiment, the change of the heading angle is grasped by the vertical motion state equation to adjust the rudder angle and power output of the ship in real time to maintain the predetermined heading and avoid dangerous situations caused by heading deviation.

[0089] In a possible implementation, obtaining the motion strategy of the target ship further includes: performing collision avoidance planning by an artificial potential field method, treating the environment of the target ship within a unit range as an artificial potential field, and the target ship is affected by the gravitational potential field and the repulsive potential field; calculating the artificial potential field corresponding to the target ship within a unit range according to the running state and the first state characteristics corresponding to the running state equation of the target ship within a unit time;

[0090] It should be noted that the artificial potential field includes the gravitational potential field and the repulsive potential field. The gravitational potential field is used to guide the target ship to move toward the target point, while the repulsive potential field is used to guide the ship to avoid obstacles.

[0091] Furthermore, the gravitational potential field that guides the target ship to move toward the target point , specifically expressed as:

[0092]

[0093] in, Expressed as the position vector of the target ship in the geographic coordinate system, Represents the current real-time speed of the target ship. and is expressed as the gravitational coefficient, Represented as the target ship position To the destination The distance It is expressed as the limit of the gravitational range, It is expressed as the target speed of the target ship;

[0094] Furthermore, the repulsive potential field that guides the ship to avoid obstacles is specifically expressed as:

[0095]

[0096] in, Expressed as the position vector of the target ship in the geographic coordinate system, Expressed as the repulsion coefficient, Represents the influence range of the obstacle. Represented as ship position To obstacles distance.

[0097] In this embodiment, the total artificial potential field of the target ship within a unit range is calculated by combining the gravitational potential field and the repulsive potential field. , guide the target ship along the total artificial potential field gradient Move in the descending direction, thereby realizing the collision avoidance planning of the target ship's operating state within unit time.

[0098] S5: Evaluate multi-dimensional risk: obtain a multi-objective reward function, perform risk assessment on the motion strategy of the target ship according to the multi-objective reward function, and obtain the third state feature.

[0099] Specifically, an optimization target set corresponding to the target motion strategy is obtained, and the optimization target set includes a navigation safety target, a navigation efficiency target, and a navigation economic target. According to the optimization target set, a multi-objective reward function corresponding to the target motion strategy is obtained. According to the multi-objective reward function, a risk assessment is performed on the target motion strategy to obtain the optimization target set and the reward value of each target in the optimization target set, that is, the third state feature.

[0100] Furthermore, the achievement degree of each goal in the optimization goal set is subdivided;

[0101] Specifically, comprehensive consideration is given to the collision risk, the impact of severe weather, and the proximity to dangerous areas in the navigation safety goals; attention is paid to the navigation time and speed stability in the navigation efficiency goals; and energy consumption cost and maintenance cost are considered in the navigation economy goals.

[0102] In this embodiment, the collision risk in the navigation safety goal is quantified by calculating the collision risk assessment index based on the distance between the target ship and the obstacle in real time. and the location of obstacles , calculate the relative position vector of the target ship and the obstacle , specifically expressed as: ; Based on the target ship's heading , the direction of the obstacle , the speed of the target ship , and the speed of the obstacle , calculate the relative speed between the target ship and the obstacle , specifically expressed as: ; Based on the relative position vector of the target ship and the obstacle and the relative speed of the target ship and the obstacle , calculate the collision risk assessment index , specifically expressed as:

[0103] ;

[0104] The potential threat of severe weather to ship navigation is assessed based on the real-time meteorological and oceanographic data corresponding to the first state characteristics; the proximity of the ship to the known dangerous areas is determined using geographic information system data. The known dangerous areas are automatically marked using geographic information system technology. The known dangerous areas include but are not limited to reefs, strong wind areas, abnormal water flow areas, etc.

[0105] In this embodiment, the estimated sailing time of the target ship according to the current motion strategy is calculated and compared with the theoretical shortest sailing time; by monitoring the fluctuation of the speed, the influence of the speed stability on the sailing efficiency is evaluated.

[0106] In this embodiment, the energy consumption cost according to the current motion strategy is calculated, and the maintenance cost is estimated in combination with the operating status of the target ship.

[0107] Furthermore, by substituting the target motion strategy into the multi-objective reward function, the corresponding reward value of the target motion strategy is calculated. The higher the reward value, the lower the risk of the target motion strategy.

[0108] In this embodiment, the weight coefficients corresponding to each target in the optimization target set are respectively expressed as the navigation safety target weight , navigation efficiency target weight , navigation economic target weight ,and , multi-objective reward function , specifically expressed as:

[0109]

[0110] in, It is represented by the reward value corresponding to the navigation safety goal, It is expressed as the reward value corresponding to the navigation efficiency target, It is expressed as the reward value corresponding to the navigation economic target.

[0111] S6: Optimize ship path planning: Based on the constraints of the third state characteristics, the movement strategy of the target ship is locally adjusted in real time to obtain the route optimization result corresponding to the movement strategy of the target ship.

[0112] Specifically, the third state characteristics are used as constraints, including navigation safety constraints, navigation efficiency constraints, and navigation economic constraints. The constraints are used to locally adjust the target motion strategy through an optimization algorithm to obtain the optimal route of the target ship.

[0113] In a possible implementation manner, the constraint of the third state feature is based on the motion trajectory of the target ship. and control trajectory , Expressed as planning time horizon, based on navigation safety objective weights , navigation efficiency target weight and weight of navigation economic objectives ,and , the multi-objective reward values ​​corresponding to the third state feature are accumulated to the maximum extent, which can be specifically expressed as:

[0114]

[0115] in, Expressed as the number of time steps, is represented as an index of time steps, The value ranges from 0 to , Expressed as time Reward for safe navigation Expressed as time The navigation efficiency bonus is Expressed as time The economic reward for sailing at that time, Expressed as time The target ship's motion trajectory, including longitudinal position, lateral position and heading angle, Expressed as time The control trajectory of the target ship includes the gravitational potential field and the repulsive potential field.

[0116] It should be noted that the multi-objective genetic algorithm is used at each time step. , calculate the value of the multi-objective reward function, and then update the control input To maximize the cumulative reward, the iteration stops when the planning time horizon T is reached.

[0117] S7: Human-computer interaction: Output the route optimization results corresponding to the target ship's motion strategy in the form of a data structure to generate a visualization result.

[0118] In this embodiment, the optimized route is drawn on the electronic nautical chart by using a drawing tool, and different colors and line shapes are set to represent different parts of the route, with the green solid line representing the optimized safe route, the red dotted line representing the original planned route, and the blue dots representing the checkpoints or safe havens along the way; a curve of changes in navigation safety, efficiency, and economic rewards is drawn on the timeline to show the achievement of each goal during the entire navigation process, with the horizontal axis representing time and the vertical axis representing the reward value.

[0119] As attached Figure 2A route planning assistance system based on intelligent algorithms and optimization theory is shown, including a system operation database, a system central processing module and a user information terminal, and also includes: a real-time meteorological and ocean perception module, a ship holographic feature analysis module, an intelligent algorithm fusion module, a dynamic route generation module, a multi-dimensional risk assessment module, a dual-optimal path planning module, and a human-computer interaction module.

[0120] Real-time meteorological and oceanographic perception module: used to obtain environmental information of the working conditions of the target ship through the meteorological and oceanographic sensor group and generate the first state characteristics;

[0121] Ship holographic feature analysis module: used to synchronize and collect the actual operating data of the target ship's working conditions and generate the second state feature;

[0122] Intelligent algorithm fusion module: used to build an intelligent algorithm fusion framework based on the historical status characteristics of the target ship and obtain a route planning auxiliary model;

[0123] Dynamic route generation module: used to analyze the probability distribution of the target ship's movement direction based on the first state characteristics transmitted by the real-time meteorological and ocean perception module and the second state characteristics transmitted by the ship holographic feature analysis module, through the route planning auxiliary model transmitted by the intelligent algorithm fusion module, and obtain the movement strategy of the target ship;

[0124] Multi-dimensional risk assessment module: used to make real-time local adjustments to the target ship's motion strategy based on the constraints of the third state characteristics transmitted by the dual-optimal path planning module, and obtain the route optimization result corresponding to the target ship's motion strategy;

[0125] Dual-optimal path planning module: used to obtain a multi-objective reward function, conduct risk assessment on the motion strategy of the target ship transmitted by the dynamic route generation module according to the multi-objective reward function, and obtain the third state characteristics;

[0126] Human-computer interaction module: It is used to output the route optimization results corresponding to the movement strategy of the target ship transmitted by the multi-dimensional risk assessment module in the form of a data structure to generate visualization results.

[0127] The system operation database includes all data texts of the ship route planning auxiliary system, and collects the information text output by each module in real time. The system central processing module is used for the information text instructions output by each module in the central control method, and the user information terminal is an information output device for receiving the ship route planning auxiliary system.

[0128] Secondly: In the drawings of the embodiments disclosed in the present invention, only the structures related to the embodiments disclosed in the present invention are involved, and other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of the present invention can be combined with each other;

[0129] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A ship route planning auxiliary method based on intelligent algorithm and optimization theory, characterized in that: include: S1: Real-time perception of meteorological and oceanographic information: Obtain environmental information of the target ship's working conditions through a group of meteorological and oceanographic sensors to generate the first state feature; S2: Analyze ship characteristics: synchronize and collect the actual operating data of the target ship's working conditions to generate the second state characteristics; S3: Intelligent fusion algorithm: Based on the historical status characteristics of the target ship, an intelligent algorithm fusion framework is constructed to obtain a route planning auxiliary model. The intelligent algorithm fusion framework integrates the improved genetic algorithm and particle swarm optimization algorithm, and introduces dynamic inertia weight and adaptive mutation strategy; S4: Generate dynamic ship route planning: Based on the first state characteristics and the second state characteristics corresponding to the target ship, the probability distribution of the target ship's movement direction is analyzed through the route planning auxiliary model to obtain the target ship's movement strategy; The motion strategy of the target ship is based on the operating state equation of the target ship in the time dimension and the artificial potential field corresponding to the target ship in the space dimension. The target motion strategy is formed through the state-potential field coupling relationship. The target motion strategy consists of the operating state and the artificial potential field. The operating state is the operating state equation of the target ship in unit time. The operating state equation of the target ship includes the longitudinal motion state equation , lateral motion state equation And the vertical motion equation of state , the vertical motion state equation, i.e., the change of the heading angle of the target ship with time; The collision avoidance planning is performed by the artificial potential field method, the environment of the target ship per unit range is regarded as the artificial potential field, and the artificial potential field corresponding to the target ship per unit range is calculated according to the running state and the first state characteristics corresponding to the running state equation of the target ship in unit time; the artificial potential field includes the gravitational potential field and repulsive potential field , the gravitational potential field is used to guide the target ship to move towards the target point, and the repulsive potential field is used to guide the ship to avoid obstacles; S5: Evaluate multi-dimensional risk: obtain a multi-objective reward function, perform risk assessment on the target ship's motion strategy based on the multi-objective reward function, and obtain the third state feature; S6: Optimize ship path planning: Based on the constraints of the third state characteristics, the movement strategy of the target ship is locally adjusted in real time to obtain the route optimization result corresponding to the movement strategy of the target ship; The constraint of the third state feature is based on the motion trajectory of the target ship. and control trajectory , Expressed as planning time horizon, based on navigation safety objective weights , navigation efficiency target weight and weight of navigation economic objectives ,and , the multi-objective reward values ​​corresponding to the third state features are accumulated to the maximum extent, which can be specifically expressed as: in, Expressed as the number of time steps, is represented as an index of time steps, The value ranges from 0 to , Expressed as time Reward for safe navigation Expressed as time The navigation efficiency bonus is Expressed as time The economic reward for sailing at that time, Expressed as time The target ship's motion trajectory, including longitudinal position, lateral position and heading angle, Expressed as time The control trajectory of the target ship, including the gravitational potential field and the repulsive potential field; S7: Human-computer interaction: Output the route optimization results corresponding to the target ship's motion strategy in the form of a data structure to generate a visualization result.

2. The ship route planning auxiliary method based on intelligent algorithm and optimization theory according to claim 1 is characterized by: The steps for constructing the S3 intelligent algorithm fusion framework are as follows: Initialize the population: Generate an initial population randomly through a genetic algorithm, and perform particle swarm optimization on the generated initial population. Each particle individual is represented as a route planning scheme, and the particle individual is a route planning candidate scheme in the initial population after the genetic algorithm and the particle swarm optimization algorithm. Fitness evaluation: Calculate the fitness value of each individual particle based on the optimal position, speed of each individual particle and the global optimal position and speed; Select individual particles: dynamically adjust the inertia weight and use the tournament selection method to select excellent individual particles to enter the next generation; Crossover operation: Perform crossover operation on the selected excellent particle individuals to generate new particle individuals; Mutation: Adopt an adaptive mutation strategy, adaptively adjust the mutation probability based on the fitness value, and mutate the individual particles after crossover.

3. The ship route planning auxiliary method based on intelligent algorithm and optimization theory according to claim 1 is characterized in that: The step S4, obtaining the movement strategy of the target ship, specifically includes: The motion strategies of multiple target ships are obtained, wherein the motion strategies of the multiple target ships include a target motion strategy, and the target motion strategy is composed of any one of the operating states corresponding to multiple unit times and an artificial potential field corresponding to the operating state in the artificial potential fields corresponding to multiple unit ranges.

4. The ship route planning auxiliary method based on intelligent algorithm and optimization theory according to claim 2 is characterized by: The longitudinal motion state equation corresponding to the running state equation of the target ship in S4 is specifically expressed as: in, It is expressed as the motion state equation of the target ship's longitudinal position in the geographic coordinate system changing with time, Represented as the initial longitude position of the target ship, Expressed as the longitudinal velocity of the target ship, is expressed as the yaw angular velocity of the target ship, Expressed as a unit of time, Expressed as the lateral velocity of the target ship, Represented as the initial heading angle of the target ship.

5. The ship route planning auxiliary method based on intelligent algorithm and optimization theory according to claim 2 is characterized by: The S4, the lateral motion state equation corresponding to the running state equation of the target ship, is specifically expressed as: in, It is expressed as the motion state equation of the target ship's lateral position in the geographic coordinate system changing with time, is represented as the initial latitude position of the target ship, Expressed as the longitudinal velocity of the target ship, is expressed as the yaw angular velocity of the target ship, Expressed as a unit of time, Represented as the initial heading angle of the target ship.

6. The ship route planning auxiliary method based on intelligent algorithm and optimization theory according to claim 2 is characterized by: The vertical motion state equation corresponding to the running state equation of the target ship in S4 is specifically expressed as: in, It is expressed as the motion state equation of the target ship's vertical position in the geographic coordinate system changing with time, that is, the change of the target ship's heading angle with time, It is represented as the initial heading angle of the target ship, is expressed as the yaw angular velocity of the target ship, Expressed as a unit of time.

7. The ship route planning auxiliary method based on intelligent algorithm and optimization theory according to claim 2 is characterized by: S4, a gravitational potential field that guides the target ship to move toward the target point , specifically expressed as: in, Expressed as the position vector of the target ship in the geographic coordinate system, Represents the current real-time speed of the target ship. and is expressed as the gravitational coefficient, Represented as the target ship position To the destination The distance It is expressed as the limit of the gravitational range, It is expressed as the target speed of the target ship; Repulsive potential field that guides ships to avoid obstacles , specifically expressed as: in, Expressed as the position vector of the target ship in the geographic coordinate system, Expressed as the repulsion coefficient, Represents the influence range of the obstacle. Represented as ship position To obstacles distance.

8. The ship route planning auxiliary method based on intelligent algorithm and optimization theory according to claim 1 is characterized by: The step S5, obtaining the third state feature, specifically includes: The optimization target set corresponding to the target motion strategy is obtained, and the optimization target set includes navigation safety target, navigation efficiency target and navigation economic target.

Images