A ship collision avoidance warning system and method based on multi-source information

Abstract

The present invention discloses a ship collision avoidance warning system and method based on multi-source information. The system comprises: a multi-source data association module, a collision warning module, and a warning feedback module. The multi-source data association module is configured to correlate ship information, environmental information, and bridge information to obtain correlation information; the collision warning module is configured to use the correlation information to predict ship collision avoidance and obtain collision warning information; and the warning feedback module is configured to issue collision warnings based on the collision warning information. The present invention rapidly integrates multiple real-time data, avoiding the impact on the effectiveness and timeliness of warnings caused by data processing delays or inaccuracies.

Description

Technical Field

[0001] The present invention relates to the technical field of ship collision avoidance, and in particular to a ship collision avoidance warning system and method based on multi-source information. Background Art

[0002] With the development of the shipping industry, both the size and frequency of vessels have increased significantly. At the same time, natural conditions and human error have also led to ship collisions. These collisions primarily include ship-to-ship collisions, ship-to-bridge pier collisions, ship-to-pier collisions, and ship-to-lock collisions. These collisions not only cause economic losses but also threaten the lifespan of buildings and even human life. Ship collisions are short-lived processes that incorporate complex nonlinear characteristics. Summary of the Invention

[0003] In view of the above problems, the present invention provides a ship anti-collision warning system based on multi-source information, the system comprising: a multi-source data association module, a collision warning module and a warning feedback module;

[0004] The multi-source data association module is used to associate the ship information, environmental information and bridge information to obtain associated information;

[0005] The collision warning module is used to use the associated information to perform ship collision avoidance prediction and obtain collision warning information;

[0006] The warning feedback module is used to issue a collision warning based on the collision warning information.

[0007] Optionally, the multi-source data association module includes: a ship information monitoring submodule, an environment monitoring submodule and an associated information submodule;

[0008] The ship information monitoring submodule is used to obtain basic information of the ship and real-time information of the ship during navigation;

[0009] The environmental monitoring submodule is used to obtain information about the vessel's surrounding environment;

[0010] The associated information submodule is used to associate the real-time information of the ship, the basic information of the ship and the information of the environment around the ship.

[0011] Optionally, the workflow of the association information submodule includes:

[0012] Preprocessing the real-time information of the ship, the basic information of the ship, and the information of the surrounding environment of the ship to obtain preprocessed data in a unified format;

[0013] Calculate the credibility of each pre-processed data and obtain the correlation with the processed data.

[0014] Optionally, the collision warning module includes a ship route planning submodule, a bridge information monitoring submodule, an obstruction information monitoring submodule, a channel deviation monitoring submodule, a ship collision monitoring submodule and a ship collision warning submodule;

[0015] The ship route planning submodule is used to plan the ship route using a chaotic sparrow optimization algorithm based on the association information of the multi-source data association module;

[0016] The bridge information monitoring submodule is used to collect early warning signals sent by the bridge;

[0017] The obstruction information monitoring submodule is used to monitor obstructions in the water, obtain obstruction information and send warning signals;

[0018] The channel deviation monitoring submodule is used to calculate the deviation between the real-time positioning of the ship and the planned route to obtain the channel deviation value;

[0019] The ship collision monitoring submodule is used to monitor whether the current ship has collided;

[0020] The ship collision warning submodule is used to send a ship collision signal when a ship collision occurs.

[0021] Optionally, the workflow of the ship route planning submodule specifically includes:

[0022] Initializing the association information using an improved Circle chaotic map;

[0023] Divide a preset number of nodes from the ship's departure point to the destination as sparrow individuals, calculate the fitness values ​​of the sparrow individuals and sort them in descending order to find the optimal fitness value, the worst fitness value and the corresponding optimal and worst positions;

[0024] Select 20% from the top according to the ranking as the discoverers in the Chaos Sparrow Optimization Algorithm;

[0025] The route position is updated according to the optimal fitness value found by the discoverer to obtain the optimal planned route.

[0026] The present invention also discloses a ship collision avoidance warning method based on multi-source information, the method comprising:

[0027] S1. Correlate the ship information, environmental information, and bridge information to obtain correlation information;

[0028] S2. Using the associated information to perform ship collision avoidance prediction and obtain collision warning information;

[0029] S3. Issue a collision warning based on the collision warning information.

[0030] Optionally, in S2, using the associated information to perform ship collision avoidance prediction, and obtaining collision warning information specifically includes:

[0031] Based on the association information of the multi-source data association module, a chaotic sparrow optimization algorithm is used to plan the ship's route;

[0032] Collect early warning signals sent by bridges;

[0033] Calculate the deviation between the ship's real-time positioning and the planned route to obtain the channel deviation value;

[0034] Calculate the probability of collision with dangerous objects based on the current ship lane deviation value;

[0035] When a ship collision occurs, a ship collision signal is sent.

[0036] Optionally, the content of performing ship route planning using a chaotic sparrow optimization algorithm based on the association information of the multi-source data association module specifically includes:

[0037] Initializing the association information using an improved Circle chaotic map;

[0038] Divide a preset number of nodes from the ship's departure point to the destination as sparrow individuals, calculate the fitness values ​​of the sparrow individuals and sort them in descending order to find the optimal fitness value, the worst fitness value and the corresponding optimal and worst positions;

[0039] Select 20% from the top according to the ranking as the discoverers in the Chaos Sparrow Optimization Algorithm;

[0040] The route position is updated according to the optimal fitness value found by the discoverer to obtain the optimal planned route.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] The present invention rapidly integrates multiple real-time data sources, avoiding the impact of data processing delays or inaccuracies that can affect the effectiveness and timeliness of early warnings. Furthermore, the present invention accounts for situations where sensor performance errors in harsh maritime environments can lead to false alarms or missed warnings, ensuring the timely and secure transmission of early warning information. Furthermore, the present invention uses a chaotic sparrow search optimization algorithm to plan routes, enabling rapid identification of the globally optimal route, real-time monitoring of ship yaw information, and prediction of the likelihood of collision following yaw. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0044] Figure 1 This is a system structure diagram of an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0046] Example 1

[0047] The present invention provides a ship collision warning system based on multi-source information, such as Figure 1 As shown, the system includes: a multi-source data association module, a collision warning module and a warning feedback module;

[0048] The multi-source data association module is used to associate ship information, environmental information, and bridge information to obtain associated information. The multi-source data association module includes a ship information monitoring submodule, an environmental monitoring submodule, and an associated information submodule. The ship information monitoring submodule is used to obtain basic ship information and real-time information about the ship during navigation; the environmental monitoring submodule is used to obtain information about the ship's surrounding environment; and the associated information submodule is used to associate the ship's real-time information, basic ship information, and information about the ship's surrounding environment.

[0049] The vessel information monitoring submodule's workflow includes recording basic vessel information, including length, width, tonnage, waterline, and hull integrity. Real-time vessel information during navigation is obtained through sensor monitoring. The environmental monitoring submodule's workflow includes acquiring real-time marine environmental information, including wind direction and force, current velocity, wave height, rainfall, temperature, and the location and size of reefs. The correlation information submodule's workflow includes preprocessing the vessel's real-time information, basic vessel information, and information about the vessel's surrounding environment to generate preprocessed data in a unified format; calculating the credibility of each piece of preprocessed data and determining its correlation with the processed data.

[0050] The real-time sensor data is correlated with historical data, and the features of the historical data are extracted and statistically analyzed. A credibility calculation model is established based on the features to obtain the credibility coefficient of the real-time measurement data. The credibility calculation model is:

[0051] ;

[0052] Among them, sum is the total number of historical data points, and sumcount is the sum of the number of occurrences of all historical data points within the statistical interval of real-time monitoring data.

[0053] The calculation method of the credibility coefficient is:

[0054] ;

[0055] Among them, C is the credibility critical value of the segmentation point.

[0056] The collision warning module is used to use the associated information to predict ship collisions and obtain collision warning information. The collision warning module includes a ship route planning submodule, a bridge information monitoring submodule, an obstruction information monitoring submodule, a channel deviation monitoring submodule, a ship collision monitoring submodule, and a ship collision warning submodule. The ship route planning submodule is used to plan the ship's route based on the associated information from the multi-source data association module using a chaotic sparrow optimization algorithm. The bridge information monitoring submodule is used to collect warning signals sent by bridges. The obstruction information monitoring submodule is used to monitor obstructions in the water, obtain obstruction information, and send warning signals. The channel deviation monitoring submodule is used to calculate the deviation between the ship's real-time positioning and the planned route to obtain a channel deviation value. The ship collision monitoring submodule is used to monitor whether the current ship has collided. The ship collision warning submodule is used to send a ship collision signal when a collision occurs.

[0057] The workflow of the ship route planning submodule specifically includes: initializing the associated information using an improved Circle chaotic map; dividing a preset number of nodes from the ship's departure point to the destination as sparrow individuals, calculating the fitness values ​​of the sparrow individuals and sorting them in descending order to find the optimal fitness value, the worst fitness value and the corresponding optimal and worst positions; selecting 20% ​​from the top according to the sorting as finders in the chaotic sparrow optimization algorithm; and updating the route position according to the optimal fitness value found by the finder to obtain the optimal planned route.

[0058] Specifically, the improved Circle chaos mapping formula is:

[0059] ;

[0060] Wherein, n is the dimension, and in this embodiment, n=3000.

[0061] In the sparrow search algorithm, when the search range is smaller than the preset range, as the number of iterations increases, the search range of the individual sparrows will gradually decrease, and the discoverers will easily gather, resulting in a local optimum. In this embodiment, adaptive weights are introduced to perform global search to avoid global optimality.

[0062] ;

[0063] in, A random number in [0, 1].

[0064] When the search range is smaller than the preset range, the finder's front position is updated to:

[0065] ;

[0066] When the search range is greater than or equal to the preset range, nonlinear adaptive weights are introduced to increase the search space and improve efficiency.

[0067] ;

[0068] At this point, the finder's rear position is updated to:

[0069] ;

[0070] During the discovery process of the discoverer, the remaining 80% will be used as followers to compete with the discoverer. However, this method will also lead to local optimality. The present invention introduces the whale optimization algorithm in the follower update position to ensure the optimization capability. After improvement, the follower position update is:

[0071] ;

[0072] in, represents the optimal sparrow position of the t+1th discoverer. Finally, the optimal sparrow position is reversely learned to obtain the planned route.

[0073] The obstruction depth metadata collected by the multibeam system was processed and imported into Matlab. The scatter function in Matlab was then used to create a 3D depth point cloud map of the obstruction. During the construction process, the obstruction depth data collected by the multibeam system was generally irregular and discrete, and the distribution did not meet the modeling requirements. To ensure a more accurate 3D model of the obstruction, Matlab interpolated the discrete points before plotting to generate a new, regularly distributed grid of depth points. Interpolation methods used included linear, cubic, spline, and nearest, with linear interpolation being the default. A new regular grid was generated using the meshgrid function in Matlab. Values ​​were assigned to this new grid using the griddata function. Two-dimensional depth contours were plotted based on the resulting values, and finally, a 3D grid map with contour lines was created. Matlab interpolation generates a regular grid of water depth data, containing all the data on the water depths of the obstruction area. Using Matlab's Surf function, this data can be plotted as a three-dimensional surface plot of the obstruction. The characteristics of a surface plot are the opposite of a grid plot: the lines are black, and the patches between the lines are colored; in a grid plot, the patches are black and the lines are colored. Matlab also provides smoothing and interpolation coloring functions by calling the shading function. Spatial analysis of obstacles primarily focuses on their spatial form and their spatial relationship to the surrounding navigation environment. Obstacles are monitored based on their spatial relationships.

[0074] The warning feedback module is used to issue a collision warning based on the collision warning information. The module receives the collision signal and packages the collision signal to send it to relevant personnel to implement collision avoidance warning.

[0075] The early warning feedback module includes a position recording submodule, a collision information submodule and a ship situation submodule.

[0076] The position recording submodule is used to record the specific location of the ship collision. The collision information submodule is used to record the object of the collision, the intensity of the collision, etc. The ship condition submodule is used to record the monitoring values ​​of various sensors of the ship after the collision.

[0077] Example 2

[0078] A ship collision avoidance warning method based on multi-source information, the method comprising:

[0079] S1. Correlate the ship information, environmental information, and bridge information to obtain correlation information.

[0080] Obtain basic information of the ship and real-time information of the ship during navigation; obtain information about the environment surrounding the ship; and associate the real-time information of the ship, basic information of the ship, and information about the environment surrounding the ship.

[0081] Basic vessel information, including length, width, tonnage, waterline, and hull integrity, is recorded. Real-time vessel information during navigation is obtained through sensor monitoring. Real-time marine environmental information, including wind direction and force, current velocity, wave height, rainfall, temperature, and reef location and size, is acquired. The real-time vessel information, basic vessel information, and information about the vessel's surroundings are preprocessed to generate preprocessed data in a unified format. The credibility of each preprocessed data set is calculated, and its correlation with the processed data is determined.

[0082] The real-time sensor data is correlated with historical data, and the features of the historical data are extracted and statistically analyzed. A credibility calculation model is established based on the features to obtain the credibility coefficient of the real-time measurement data. The credibility calculation model is:

[0083] ;

[0084] Among them, sum is the total number of historical data points, and sumcount is the sum of the number of occurrences of all historical data points within the statistical interval of real-time monitoring data.

[0085] The calculation method of the credibility coefficient is:

[0086] ;

[0087] Among them, C is the credibility critical value of the segmentation point.

[0088] S2. Use the associated information to perform ship collision avoidance prediction and obtain collision warning information.

[0089] In S2, the association information is used to perform ship collision avoidance prediction, and the content of the collision warning information obtained specifically includes: based on the association information of the multi-source data association module, the chaotic sparrow optimization algorithm is used to plan the ship's route; the warning signal sent by the bridge is collected; the deviation between the real-time positioning of the ship and the planned route is calculated to obtain the channel deviation value; the probability of collision with dangerous objects based on the current ship channel deviation value is calculated; when the ship collides, a ship collision signal is sent.

[0090] Based on the association information of the multi-source data association module, the content of using the chaotic sparrow optimization algorithm to plan the ship's route specifically includes: using an improved Circle chaotic map to initialize the association information; dividing a preset number of nodes from the ship's departure point to the destination end point as sparrow individuals, calculating the fitness values ​​of the sparrow individuals and sorting them in descending order to find the optimal fitness value, the worst fitness value and the corresponding optimal position and the worst position; selecting 20% ​​from the top according to the sorting as finders in the chaotic sparrow optimization algorithm; updating the route position according to the optimal fitness value found by the finder to obtain the optimal planned route.

[0091] The improved Circle chaos mapping formula is:

[0092] ;

[0093] Wherein, n is the dimension, and in this embodiment, n=3000.

[0094] In the sparrow search algorithm, when the search range is smaller than the preset range, as the number of iterations increases, the search range of the individual sparrows will gradually decrease, and the discoverers will easily gather, resulting in a local optimum. In this embodiment, adaptive weights are introduced to perform global search to avoid global optimality.

[0095] ;

[0096] in, A random number in [0, 1].

[0097] When the search range is smaller than the preset range, the finder's front position is updated to:

[0098] ;

[0099] When the search range is greater than or equal to the preset range, nonlinear adaptive weights are introduced to increase the search space and improve efficiency.

[0100] ;

[0101] At this point, the finder's rear position is updated to:

[0102] ;

[0103] During the discovery process of the discoverer, the remaining 80% will be used as followers to compete with the discoverer. However, this method will also lead to local optimality. The present invention introduces the whale optimization algorithm in the follower update position to ensure the optimization capability. After improvement, the follower position update is:

[0104] ;

[0105] in, Represents the optimal sparrow position of the t+1th discoverer. Finally, the optimal sparrow position is reversely learned to obtain the planned route.

[0106] S3. Issue a collision warning based on the collision warning information.

[0107] The specific location of the ship collision is recorded. The collision target and severity are also recorded. After the collision is recorded, the ship's sensors monitor the values. This information is packaged and sent to relevant personnel for collision warning.

[0108] The embodiments described above are merely descriptions of preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by persons skilled in the art should fall within the scope of protection defined by the claims of the present invention.

Claims

1. A ship collision warning system based on multi-source information, characterized in that: The system includes: a multi-source data association module, a collision warning module and a warning feedback module; The multi-source data association module is used to associate the ship information, environmental information and bridge information to obtain associated information; The collision warning module is used to use the associated information to perform ship collision avoidance prediction and obtain collision warning information; The warning feedback module is used to issue a collision warning based on the collision warning information; The multi-source data association module includes: a ship information monitoring submodule, an environment monitoring submodule and an associated information submodule; The ship information monitoring submodule is used to obtain basic information of the ship and real-time information of the ship during navigation; The environmental monitoring submodule is used to obtain information about the vessel's surrounding environment; The associated information submodule is used to associate the real-time information of the ship, the basic information of the ship and the information of the surrounding environment of the ship; The workflow of the association information submodule includes: Preprocessing the real-time information of the ship, the basic information of the ship, and the information of the surrounding environment of the ship to obtain preprocessed data in a unified format; Calculate the credibility of each pre-processed data and obtain the correlation with the processed data; The collision warning module includes a ship route planning submodule, a bridge information monitoring submodule, an obstruction information monitoring submodule, a channel deviation monitoring submodule, a ship collision monitoring submodule and a ship collision warning submodule; The ship route planning submodule is used to plan the ship route using a chaotic sparrow optimization algorithm based on the association information of the multi-source data association module; The bridge information monitoring submodule is used to collect early warning signals sent by the bridge; The obstruction information monitoring submodule is used to monitor obstructions in the water, obtain obstruction information and send warning signals; The channel deviation monitoring submodule is used to calculate the deviation between the real-time positioning of the ship and the planned route, obtain the channel deviation value and perform correction; The ship collision monitoring submodule is used to monitor whether the current ship has collided; The ship collision warning submodule is used to send a ship collision signal when a ship collision occurs; The workflow of the ship route planning submodule specifically includes: initializing the association information using an improved Circle chaotic map; dividing a preset number of nodes from the ship's departure point to the destination as sparrow individuals, calculating the fitness values ​​of the sparrow individuals and sorting them in descending order to find the optimal fitness value, the worst fitness value, and the corresponding optimal and worst positions; selecting 20% ​​of the nodes from the top of the list as finders in the chaotic sparrow optimization algorithm; and updating the route position based on the optimal fitness value found by the finders to obtain the optimal planned route. Specifically, the improved Circle chaos mapping formula is: ； Wherein, n is the dimension, and in this embodiment, n=3000; In the sparrow search algorithm, when the search range is smaller than the preset range, as the number of iterations increases, the search range of the individual sparrow will decrease dimension by dimension, and the discoverers will easily gather, leading to a local optimum. Adaptive weights are introduced to perform global search to avoid the global optimum. ； in, A random number in the range [0, 1]; When the search range is smaller than the preset range, the finder's front position is updated to: ； When the search range is greater than or equal to the preset range, nonlinear adaptive weights are introduced to increase the search space and increase efficiency; ； At this point, the finder's rear position is updated to: ； During the discovery process of the discoverer, the remaining 80% will be used as followers to compete with the discoverer. However, this method will also lead to local optimality. The whale optimization algorithm is introduced in the follower update position to ensure the optimization ability. After improvement, the follower position update is: ； in, Represents the optimal sparrow position of the t+1th discoverer; finally, the optimal sparrow position is reversely learned to obtain the planned route.

2. A ship collision avoidance warning method based on multi-source information, the method is implemented based on the warning system according to claim 1, characterized in that: Methods include: S1. Correlate the ship information, environmental information, and bridge information to obtain correlation information; S2. Using the associated information to perform ship collision avoidance prediction and obtain collision warning information; S3. Issue a collision warning based on the collision warning information.

3. The ship collision avoidance warning method based on multi-source information according to claim 2, characterized in that: In S2, the ship collision avoidance prediction is performed using the associated information, and the content of the collision warning information obtained specifically includes: Based on the association information of the multi-source data association module, a chaotic sparrow optimization algorithm is used to plan the ship's route; Collect early warning signals sent by bridges; Calculate the deviation between the ship's real-time positioning and the planned route to obtain the channel deviation value; Calculate the probability of collision with dangerous objects based on the current ship lane deviation value; When a ship collision occurs, a ship collision signal is sent.

4. The ship collision avoidance warning method based on multi-source information according to claim 3 is characterized in that: The content of performing ship route planning using the chaotic sparrow optimization algorithm based on the association information of the multi-source data association module specifically includes: Initializing the association information using an improved Circle chaotic map; Divide a preset number of nodes from the ship's departure point to the destination as sparrow individuals, calculate the fitness values ​​of the sparrow individuals and sort them in descending order to find the optimal fitness value, the worst fitness value and the corresponding optimal and worst positions; Select 20% from the top according to the ranking as the discoverers in the Chaos Sparrow Optimization Algorithm; The route position is updated according to the optimal fitness value found by the discoverer to obtain the optimal planned route.

Images