Artificial intelligence-based ship behavior identification and dynamic route planning method and system

Abstract

The invention discloses a ship behavior identification and dynamic route planning method and system based on artificial intelligence. The method comprises the following steps: 1) collecting ship data and environment data; 2) preprocessing the collected ship data and environment data; 3) identifying the ship data to obtain ship behavior data; 4) analyzing the ship behavior data, calculating a ship behavior risk index, and evaluating the ship behavior risk; 5) integrating the environment data, the ship behavior data and the risk index to generate a route; and 6) controlling the ship to sail according to the route, monitoring ship data and environment data in the ship sailing process in real time, and dynamically adjusting the route based on the real-time data and the risk index. According to the method, ship data and environment data are collected in real time, ship behavior data analysis and environment risk assessment are combined, the ship route is dynamically optimized, the adaptability and response capacity of the ship are improved, and the safety and economical efficiency of navigation are ensured.

Description

Technical Field

[0001] The present invention relates to a ship path dynamic planning technology, and in particular to a ship behavior recognition and dynamic route planning method and system based on artificial intelligence. Background Art

[0002] During maritime navigation, the automatic identification system (AIS) and other ship monitoring systems have become popular. These systems can provide real-time ship dynamic information, such as a real-time virtual enhanced simulation system and method for ship driving based on AIS disclosed in the Chinese patent application publication number CN103631148A. The system consists of director monitoring software, AIS comprehensive processing software, ship driving display and control software, ship power model solving software, electronic chart software, navigation radar software and ship driving vision software. At the beginning of the simulation, the director monitoring software releases the simulation task, and other simulation software simulates according to the simulation task. The director monitoring software records the entire simulation process, and finally replays, evaluates, and analyzes the entire simulation process, and gives the evaluation results. However, there is still a lot of room for improvement in the use of ship dynamic information data in the prior art. For example, how to accurately identify the normal behavior and potential abnormal behavior of the ship from massive data, and how to provide the ship with the optimal route planning based on complex environmental factors, these problems are challenges that need to be solved in the current navigation field. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a method and system for ship behavior recognition and dynamic route planning based on artificial intelligence in view of the defects in the prior art.

[0004] The technical solution adopted by the present invention to solve the technical problem is: a method for ship behavior recognition and dynamic route planning based on artificial intelligence, comprising the following steps:

[0005] 1) Collect vessel data and environmental data;

[0006] 2) Preprocessing the collected ship data and environmental data;

[0007] 3) Identify ship data and obtain ship behavior data;

[0008] 4) Analyze ship behavior data, calculate ship behavior risk index, and assess ship behavior risk;

[0009] 5) Generate routes by integrating environmental data, vessel behavior data and risk index;

[0010] 6) Control the ship to sail along the route, monitor the ship data and environmental data during the navigation process in real time, and dynamically adjust the route based on real-time data and risk index.

[0011] According to the above scheme, in step 3), the ship data is identified to obtain the ship behavior data by: classifying the identified ship behavior using machine learning and pattern recognition models based on the ship data to form ship behavior data; the ship behavior is divided into normal behavior and abnormal behavior.

[0012] According to the above scheme, in step 3), the machine learning and pattern recognition model are used to classify the identified ship behaviors as follows:

[0013] 3.1) A single-layer LSTM network with 128 neurons is used as the feature learning network, and the historical data of ships is brought into the feature learning network for training;

[0014] 3.2) Input the collected ship data into the trained feature learning network, use the LSTM network to predict future behavior, and output the ship behavior data.

[0015] According to the above scheme, in step 4), the ship behavior data is analyzed by combining the Bayesian network and the anomaly detection algorithm to calculate the risk index;

[0016] The details are as follows:

[0017] 4.1) Using the Gaussian mixture model as the anomaly detection model and the historical ship behavior data as the training set, the anomaly detection model is trained with the training set to obtain the anomaly threshold of each behavior data;

[0018] 4.2) Input the ship behavior data into the trained anomaly detection model, combine the isolation forest and DBSCAN algorithms to detect abnormal behavior and distinguish normal behavior from abnormal behavior;

[0019] 4.3) Based on normal and abnormal behaviors, the random forest algorithm is used to score the ship behavior to form a ship behavior risk index, and whether the ship behavior is risky is determined based on the risk index;

[0020] 4.4) The risk index is normalized into a dimensionless value (ranging from 0 to 1) as input for route generation in step 5) and route adjustment in step 6).

[0021] According to the above scheme, in step 5), the route is generated by integrating environmental data, ship behavior data and risk index. The specific steps are as follows:

[0022] 5.1) Plan the navigation route based on the ship’s starting point, necessary waypoints, and obstacle avoidance areas;

[0023] 5.2) Construct a multi-objective cost function with navigation time, fuel consumption and risk index as targets;

[0024] 5.3) Adopt hybrid A* algorithm or genetic algorithm to select several routes with lower costs as alternative routes through iterative calculation, and use risk index weight factor as additional constraint to ensure that the generated alternative routes have the lowest cost and the risk level is within an acceptable range;

[0025] 5.4) Based on environmental data, numerical weather forecast models and ocean dynamics models are used to evaluate the safety and environmental impact of alternative routes. Based on the cost, safety and environmental impact of the alternative routes, the alternative routes are scored respectively, and the preferred route is selected from the alternative routes as the route for the ship's current voyage based on the scores.

[0026] According to the above scheme, in step 6), the route is dynamically adjusted based on real-time vessel data, environmental data and risk index, specifically including:

[0027] Obtain the ship's position data and calculate the distance the ship has currently deviated from the route. When the deviation of the ship from the route is greater than the set route deviation threshold, regenerate the route by combining the current environmental data and the ship's behavior data.

[0028] When the deviation of the ship from the route is greater than the set route deviation threshold, it is determined whether to adjust the route based on the real-time environmental data and real-time risk index, as follows:

[0029] 6.1) Apply time series analysis and pattern recognition techniques to identify trends in environmental parameters;

[0030] 6.2) Dynamically assess the potential risks of the route by referring to the changing trends of environmental parameters and combining the real-time risk index:

[0031] Use geographic information systems and risk maps to spatially interpolate current environmental data to generate high-resolution dynamic risk scenario maps;

[0032] Based on the real-time risk index, dynamic weights are updated and the comprehensive risk coefficient of the route is calculated using an improved fast-convergence optimization algorithm;

[0033] The comprehensive risk factor formula is:

[0034] R c =w 1 ·R t +w 2 ·E c

[0035] Among them, R c is the comprehensive risk factor of the current route, R t is the real-time risk index, E c is the environmental risk parameter, w 1 and w 2is the weight factor, which is adjusted dynamically according to the route priority;

[0036] 6.3) When the risk factor exceeds the threshold, the route is recalculated according to step 5).

[0037] On the other hand, the present invention also provides a ship behavior recognition and dynamic route planning system based on artificial intelligence, comprising:

[0038] Data acquisition module, used to collect vessel data and environmental data;

[0039] A preprocessing module, used for preprocessing the vessel data and the environmental data;

[0040] The behavior recognition module is used to identify the ship data and obtain the ship behavior data; the acquisition process is: according to the ship data, the identified ship behavior is classified by using machine learning and pattern recognition models to form ship behavior data; the ship behavior is divided into normal behavior and abnormal behavior;

[0041] Anomaly detection module, used to analyze ship behavior data, calculate ship behavior risk index, and assess ship behavior risk;

[0042] The route planning module is used to generate routes by integrating environmental data, vessel behavior data and risk index;

[0043] The route dynamic update module is used to summarize the ship's navigation process, monitor the ship data and environmental data in real time during the navigation process, and adjust the route based on the real-time ship data and environmental data.

[0044] According to the above scheme, in the route planning module, the environmental data, the ship behavior data and the risk index are integrated to generate the route; the details are as follows:

[0045] 1) Plan the navigation route based on the ship's starting point, necessary waypoints, and obstacle avoidance areas;

[0046] 2) Construct a multi-objective cost function with navigation time, fuel consumption and risk index as targets;

[0047] 3) Using a hybrid A* algorithm or genetic algorithm, several routes with lower costs are selected as alternative routes through iterative calculation, and the ship behavior risk index is used as an additional constraint to ensure that the generated alternative routes have the lowest cost and an acceptable risk level;

[0048] 4) Based on environmental data, numerical weather forecast models and ocean dynamics models are used to evaluate the safety and environmental impact of alternative routes. Based on the cost, safety and environmental impact of the alternative routes, the alternative routes are scored respectively, and the preferred route is selected from the alternative routes as the route for the ship's current voyage based on the scores.

[0049] According to the above scheme, in the route dynamic update module, the route is dynamically adjusted based on real-time ship data, environmental data and risk index, specifically including:

[0050] 1) Obtain the ship's position data and calculate the distance the ship has deviated from the route. When the deviation of the ship from the route is greater than the set route deviation threshold, the route is regenerated by combining the current environmental data and the ship's behavior data;

[0051] 2) When the deviation of the ship from the route is greater than the set route deviation threshold, it is determined whether to adjust the route based on the real-time environmental data and real-time risk index, as follows:

[0052] 2.1) Apply time series analysis and pattern recognition techniques to identify the changing trends of environmental parameters;

[0053] 2.2) Dynamically assess the potential risks of the route by referring to the changing trends of environmental parameters and combining the real-time risk index:

[0054] Use geographic information systems and risk maps to spatially interpolate current environmental data to generate high-resolution dynamic risk scenario maps;

[0055] Based on the real-time risk index, dynamic weights are updated and the comprehensive risk factor of the route is calculated using a fast-converging optimization algorithm;

[0056] The comprehensive risk factor formula is:

[0057] R c =w 1 ·R t +w 2 ·E c

[0058] Among them, R c is the comprehensive risk factor of the current route, R t is the real-time risk index, E c is the environmental risk parameter, w 1 and w 2 is the weight factor, which is adjusted dynamically according to the route priority;

[0059] 3) When the risk factor exceeds the threshold, the route planning module is called to regenerate the route.

[0060] The beneficial effects produced by the present invention are:

[0061] 1. The present invention realizes dynamic optimization of routes by real-time collection of ship data and environmental data, combined with ship behavior data analysis and environmental risk assessment, improves the adaptability and responsiveness of ships in complex sea conditions and emergencies, and ensures the safety and economy of navigation.

[0062] 2. The method of the present invention uses the risk index as the core indicator throughout the entire process of route generation and dynamic adjustment, and combines machine learning and optimization algorithms to make route planning more intelligent and efficient, which can effectively avoid potential risks and reduce fuel consumption and time costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:

[0064] Figure 1 It is a method flow chart of an embodiment of the present invention. DETAILED DESCRIPTION

[0065] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0066] like Figure 1 As shown, the present invention relates to a method for ship behavior recognition and dynamic route planning based on artificial intelligence, which comprises the following steps:

[0067] 1) Collecting vessel data and environmental data:

[0068] The received ship data includes: AIS signals from ships, extracting basic ship information, MMSI number, ship name, location, heading, speed, etc.;

[0069] In addition to AIS data, various sensors deployed in the marine environment are important sources of environmental data. These sensors include anemometers, wave meters, tide gauges, water temperature meters, etc., which can provide real-time data about the marine environment.

[0070] In order to have a more comprehensive understanding of the behavior of vessels and the state of the marine environment, additional information such as weather forecasts, ocean current charts, fishing activity areas, etc. can be obtained from third-party organizations such as meteorological stations, satellite data providers, and IMO through the third-party data integration submodule.

[0071] 2) Preprocessing of vessel data and environmental data, including:

[0072] 2.1) Use checksum or hash function (such as MD5 or SHA-256) to detect errors in data transmission or entry, and delete erroneous data;

[0073] 2.2) Use the Z-score model to identify outliers, where the Z-score threshold is set to 3, indicating that data points exceeding 3 standard deviations are considered outliers. For outliers, the average value of adjacent data points is used to replace them;

[0074] 2.3) Identify duplicate data by comparing key values ​​or record tuples, use matching algorithms such as fuzzy matching to handle nearly duplicate data, and eliminate duplicate and nearly duplicate data;

[0075] 2.4) Identify missing values ​​in the data. Any value containing NaN (non-number) or infinity is considered a missing value. When the missing value is in the middle of continuous data, use linear interpolation to fill in the missing value for a short period of time; when the missing value is at the beginning or end of continuous data, use linear extrapolation based on previous and subsequent trends to fill in the missing value;

[0076] 2.5) Integrate the vessel data and environmental data. The integration includes time synchronization and unit unification. All sensor data are synchronized to UTC time, allowing a maximum deviation of 1 second. All speed-related measurements are converted to knots, and wind speed is converted to meters per second (m / s).

[0077] 2.6) Normalize and standardize the ship data and environmental data, scale all numerical data to between 0 and 1, and make the data have zero mean and unit variance;

[0078] 2.7) Extract feature values ​​from vessel data and environmental data, including time domain features and frequency domain features. Time domain features refer to extracting statistical features such as mean, variance, maximum, minimum and volatility from time series data. Frequency domain features refer to applying fast Fourier transform (FFT) to analyze periodic features and energy distribution.

[0079] 2.8) Use Kalman filter to fuse the position data from different sensors to reduce noise and improve the accuracy of position estimation;

[0080] After data processing, use relational databases such as MySQL to store structured data, and non-relational databases such as MongoDB to store semi-structured data. Back up data regularly, adopt a strategy that combines incremental backup and full backup, and set up a disaster recovery plan to ensure that data can be quickly restored in the event of a system failure.

[0081] 3) Based on the ship data, machine learning and pattern recognition technology are used to obtain ship behavior data. The specific steps include:

[0082] 3.1) A single-layer LSTM network with 128 neurons is used as the feature learning network, and the historical data of ships is brought into the feature learning network for training;

[0083] 3.2) Receive pre-processed JSON format data through REST API, and then input the collected ship data into the trained feature learning network. Each data contains timestamp (UNIX time), GPS coordinates (latitude and longitude), speed (knots), heading (degrees), ship identification code (IMO number), etc. The data is updated once a second, and the LSTM network is used to predict future behavior and output ship behavior data;

[0084] The ship behavior data includes navigation speed, turning angle, acceleration, route stability (standard deviation) and corresponding ship type and ship size.

[0085] 4) Combine Bayesian network and anomaly detection algorithm to analyze ship behavior data, calculate risk index, and then evaluate ship behavior risk. The specific steps are as follows:

[0086] 4.1) Using the Gaussian mixture model as the anomaly detection model, the historical ship behavior data in the past 30 days was selected as the training set, and the Gaussian mixture model was used to model the distribution of the behavior data and determine the anomaly threshold; the model was retrained every week to reflect the latest changes in the ship behavior pattern and improve the accuracy of the model in anomaly detection;

[0087] 4.2) Input the real-time ship behavior data into the trained anomaly detection model, and combine the isolation forest and DBSCAN algorithms to detect abnormal behavior. The number of samples of the isolation forest is set to 2000 to detect isolated point behavior; the neighborhood radius (Eps) of the DBSCAN algorithm is set to 0.5, and the minimum number of points (MinPts) is set to 5 for cluster analysis and distinguishing normal behavior from abnormal behavior. Whenever at least 100 new abnormal behavior instances are detected, these instances are added to the training set and the anomaly detection model is retrained to improve the real-time detection capability.

[0088] 4.3) Risk scoring is performed on the detected behaviors. Based on normal and abnormal behaviors, the random forest algorithm is used to score the ship behaviors and generate a risk index. The risk index uses a standardized value from 0 to 1, and the higher the value, the greater the risk of the behavior (such as exceeding 0.75, it is considered high risk).

[0089] 4.4) Output the risk index, and use the standardized risk index as an important input for route generation in step 5) and dynamic adjustment in step 6) to comprehensively evaluate the safety and economy of the route.

[0090] 5) Using artificial intelligence optimization algorithms, we can generate routes by integrating environmental data, vessel behavior data, and risk index. The specific steps are as follows:

[0091] 5.1) Plan the initial route based on the ship's starting point (latitude and longitude), necessary waypoints and obstacle avoidance areas (such as traffic separation zones and reef areas); use the geographic information system to generate a preliminary route map and verify the feasibility of the route.

[0092] 5.2) Based on the voyage time, fuel consumption and risk index, a cost function is constructed using a multi-objective optimization method;

[0093] J=r 1 ·T+r 2 ·F+r 3 ·R

[0094] Among them, T is the voyage time, F is the fuel consumption, R is the risk index, and r 1 、r 2 、r 3 It is a weight factor, which is adjusted dynamically according to specific navigation requirements.

[0095] 5.3) Use the hybrid A* algorithm to screen several routes with lower costs through iterative calculation; in the hybrid A* algorithm, set the heuristic factors α=1.0 and β=2.0 to give priority to avoiding high-risk areas; or use the genetic algorithm (GA) with a population size of 100, a crossover rate of 0.8, and a mutation rate of 0.02 to optimize the route planning process; ensure that each alternative route meets both the economic requirements and the risk control constraints.

[0096] 5.4) Based on environmental data (such as weather forecasts and ocean dynamics models), safety and environmental impact assessments are conducted on alternative routes; alternative routes are scored, and the routes with the highest scores are selected as the preferred routes based on comprehensive considerations of cost, safety and environmental impact, and the final navigation plan is output.

[0097] 6) Control the ship to sail according to the route, monitor the ship data and environmental data during the navigation process in real time, and dynamically adjust the route based on real-time data and risk index, including:

[0098] Obtain the ship's location data (through AIS and GPS), calculate the distance the ship currently deviates from the route, and when the deviation of the ship from the route is greater than the set route deviation threshold, combine the current environmental data and the ship's behavior data to regenerate the route according to the method in step 5);

[0099] When the deviation of the ship from the route is less than the set route deviation threshold, it is determined whether to adjust the route based on the real-time environmental data and real-time risk index, as follows:

[0100] The ship sails according to the route, monitors the ship data and environmental data in real time during the voyage, and dynamically adjusts the route based on the real-time ship data, environmental data and risk index, including:

[0101] 6.1) Apply time series analysis and pattern recognition techniques to identify trends in environmental parameters;

[0102] 6.2) Dynamically assess the potential risks of the route by referring to the changing trends of environmental parameters and combining the real-time risk index:

[0103] Use geographic information systems and risk maps to spatially interpolate current environmental data (such as wind speed, wave height, and tidal currents) to generate high-resolution dynamic risk scenario maps;

[0104] Based on the real-time risk index, dynamic weights are updated and the comprehensive risk factor of the route is calculated using a fast-converging optimization algorithm (such as gradient descent or particle swarm optimization algorithm);

[0105] The comprehensive risk factor formula is:

[0106] R c =w 1 ·R t +w 2 ·E c

[0107] Among them, R c is the comprehensive risk factor of the current route, R t is the real-time risk index, E c is the environmental risk parameter (such as wave height and tidal strength), w 1 and w 2 is the weight factor, which is adjusted dynamically according to the route priority;

[0108] 6.3) When the risk factor exceeds the threshold, the route is recalculated according to step 5).

[0109] According to the above method, the present invention also provides a ship behavior recognition and dynamic route planning system based on artificial intelligence, comprising:

[0110] Data acquisition module, used to collect vessel data and environmental data;

[0111] A preprocessing module, used for preprocessing the vessel data and the environmental data;

[0112] The behavior recognition module is used to identify the ship data and obtain the ship behavior data; the acquisition process is: according to the ship data, the identified ship behavior is classified by using machine learning and pattern recognition models to form ship behavior data; the ship behavior is divided into normal behavior and abnormal behavior;

[0113] Anomaly detection module, used to analyze ship behavior data, calculate ship behavior risk index, and assess ship behavior risk;

[0114] The route planning module is used to integrate environmental data, vessel behavior data and risk index to generate routes;

[0115] The route dynamic update module is used to summarize the ship's navigation process, monitor the ship data and environmental data in real time during the navigation process, and adjust the route based on the real-time ship data and environmental data.

[0116] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all these improvements and changes should fall within the scope of protection of the appended claims of the present invention.

Claims

1. A method for ship behavior recognition and dynamic route planning based on artificial intelligence, characterized in that: The following steps are involved: 1) Collect vessel data and environmental data; 2) Preprocessing the collected ship data and environmental data; 3) Identify ship data and obtain ship behavior data; 4) Analyze ship behavior data, calculate ship behavior risk index, and assess ship behavior risk; 5) Generate routes by integrating environmental data, vessel behavior data and risk index; 6) Control the ship to sail along the route, monitor the ship data and environmental data during the navigation process in real time, and dynamically adjust the route based on real-time data and risk index.

2. The method for ship behavior recognition and dynamic route planning based on artificial intelligence according to claim 1 is characterized in that: In the step 3), the ship data is identified to obtain the ship behavior data by: classifying the identified ship behavior using machine learning and pattern recognition models based on the ship data to form the ship behavior data; the ship behavior is divided into normal behavior and abnormal behavior.

3. The method for ship behavior recognition and dynamic route planning based on artificial intelligence according to claim 1 is characterized in that: In step 3), machine learning and pattern recognition models are used to classify the identified ship behaviors, as follows: 3.1) A single-layer LSTM network with 128 neurons is used as the feature learning network, and the historical data of ships is brought into the feature learning network for training; 3.2) Input the collected ship data into the trained feature learning network, use the LSTM network to predict future behavior, and output the ship behavior data.

4. The method for ship behavior recognition and dynamic route planning based on artificial intelligence according to claim 1 is characterized in that: In the step 4), the vessel behavior data is analyzed by combining the Bayesian network and the anomaly detection algorithm to calculate the risk index; The details are as follows: 4.1) Using the Gaussian mixture model as the anomaly detection model and the historical ship behavior data as the training set, the anomaly detection model is trained with the training set to obtain the anomaly threshold of each behavior data; 4.2) Input the ship behavior data into the trained anomaly detection model, combine the isolation forest and DBSCAN algorithms to detect abnormal behavior and distinguish normal behavior from abnormal behavior; 4.3) Based on normal and abnormal behaviors, the random forest algorithm is used to score the ship behavior to form a ship behavior risk index, and whether the ship behavior is risky is determined based on the risk index.

5. The method for ship behavior recognition and dynamic route planning based on artificial intelligence according to claim 1 is characterized in that: In step 5), the route is generated by integrating environmental data, ship behavior data and risk index. The specific steps are as follows: 5.1) Plan the navigation route based on the ship’s starting point, necessary waypoints, and obstacle avoidance areas; 5.2) Construct a multi-objective cost function with navigation time, fuel consumption and risk index as targets; 5.3) Using a hybrid A* algorithm or genetic algorithm, several routes with lower costs are selected as alternative routes through iterative calculation, and the ship behavior risk index is used as an additional constraint to ensure that the generated alternative routes have the lowest cost and an acceptable risk level; 5.4) Based on environmental data, numerical weather forecast models and ocean dynamics models are used to evaluate the safety and environmental impact of alternative routes. Based on the cost, safety and environmental impact of the alternative routes, the alternative routes are scored respectively, and the preferred route is selected from the alternative routes as the route for the ship's current voyage based on the scores.

6. The method for ship behavior recognition and dynamic route planning based on artificial intelligence according to claim 1 is characterized in that: In step 6), the route is dynamically adjusted based on real-time vessel data, environmental data and risk index, specifically including: 6.1) Obtain the ship's position data and calculate the distance the ship has deviated from the route. When the deviation of the ship from the route is greater than the set route deviation threshold, the route is regenerated by combining the current environmental data and the ship's behavior data; 6.2) When the deviation of the ship from the route is greater than the set route deviation threshold, it is determined whether to adjust the route based on the real-time environmental data and real-time risk index, as follows: 6.2.1) Apply time series analysis and pattern recognition techniques to identify trends in environmental parameters; 6.2.2) Dynamically assess the potential risks of the route by referring to the changing trends of environmental parameters and combining the real-time risk index: Use geographic information systems and risk maps to spatially interpolate current environmental data to generate high-resolution dynamic risk scenario maps; Based on the real-time risk index, dynamic weights are updated and the comprehensive risk factor of the route is calculated using a fast-converging optimization algorithm; The comprehensive risk factor formula is: R c =w1·R t +w2·E c Among them, R c is the comprehensive risk factor of the current route, R t is the real-time risk index, E c is the environmental risk parameter, w1 and w2 are weight factors, which are dynamically adjusted according to the route priority; 6.3) When the risk factor exceeds the threshold, the route is regenerated according to step 5).

7. A ship behavior recognition and dynamic route planning system based on artificial intelligence, characterized in that: include: Data acquisition module, used to collect vessel data and environmental data; A preprocessing module, used for preprocessing the vessel data and the environmental data; A behavior recognition module is used to recognize ship data and obtain ship behavior data; The acquisition process is as follows: according to the ship data, the identified ship behaviors are classified using machine learning and pattern recognition models to form ship behavior data; the ship behaviors are divided into normal behaviors and abnormal behaviors; Anomaly detection module, used to analyze ship behavior data, calculate ship behavior risk index, and assess ship behavior risk; The route planning module is used to generate routes by integrating environmental data, vessel behavior data and risk index; The route dynamic update module is used to summarize the ship's navigation process, monitor the ship data and environmental data in real time during the navigation process, and adjust the route based on the real-time ship data and environmental data.

8. The artificial intelligence-based ship behavior recognition and dynamic route planning system according to claim 7 is characterized in that: In the route planning module, the environmental data, the vessel behavior data and the risk index are integrated to generate the route; the details are as follows: 1) Plan the navigation route based on the ship's starting point, necessary waypoints, and obstacle avoidance areas; 2) Construct a multi-objective cost function with navigation time, fuel consumption and risk index as targets; 3) Using a hybrid A* algorithm or genetic algorithm, several routes with lower costs are selected as alternative routes through iterative calculation, and the ship behavior risk index is used as an additional constraint to ensure that the generated alternative routes have the lowest cost and an acceptable risk level; 4) Based on environmental data, numerical weather forecast models and ocean dynamics models are used to evaluate the safety and environmental impact of alternative routes. Based on the cost, safety and environmental impact of the alternative routes, the alternative routes are scored respectively, and the preferred route is selected from the alternative routes as the route for the ship's current voyage based on the scores.

9. The artificial intelligence-based ship behavior recognition and dynamic route planning system according to claim 7, characterized in that: In the route dynamic update module, the route is dynamically adjusted based on real-time vessel data, environmental data and risk index, specifically including: 1) Obtain the ship's position data and calculate the distance the ship has deviated from the route. When the deviation of the ship from the route is greater than the set route deviation threshold, the route is regenerated by combining the current environmental data and the ship's behavior data; 2) When the deviation of the ship from the route is greater than the set route deviation threshold, it is determined whether to adjust the route based on the real-time environmental data and real-time risk index, as follows: 2.1) Apply time series analysis and pattern recognition techniques to identify the changing trends of environmental parameters; 2.2) Dynamically assess the potential risks of the route by referring to the changing trends of environmental parameters and combining the real-time risk index: Use geographic information systems and risk maps to spatially interpolate current environmental data to generate high-resolution dynamic risk scenario maps; Based on the real-time risk index, dynamic weights are updated and the comprehensive risk factor of the route is calculated using a fast-converging optimization algorithm; The comprehensive risk factor formula is: R c =w1·R t +w2·E c Among them, R c is the comprehensive risk factor of the current route, R t is the real-time risk index, E c is the environmental risk parameter, w1 and w2 are weight factors, which are dynamically adjusted according to the route priority; 3) When the risk factor exceeds the threshold, the route planning module is called to regenerate the route.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images