Water traffic safety early warning system

Abstract

The invention discloses a water traffic safety early warning system. Comprising a monitor, a ship end and a calculation end, the monitor is arranged in a monitored water area, the ship end comprises a signal sending module, a data acquisition module, a data preprocessing module and a response module, the calculation end comprises a data calculation module, and the ship end and the calculation end are connected through a data transmission module. By integrating a sensing technology, a data processing technology and a machine learning method, the safety and reliability of water traffic are remarkably improved, and more intelligent and reliable navigation support is provided for ships.

Description

Technical Field

[0001] The invention belongs to the technical field of traffic safety and relates to a water traffic safety early warning system. Background Art

[0002] With the unstoppable trend of economic globalization and the continuous development of the water transportation industry, the status and role of water transportation have gradually increased, and more and more ships have been put into water transportation to meet the transportation needs.

[0003] In the process of water transportation, ships always face many potential risks, one of which is the possibility of collision with obstacles. Whether in narrow rivers, wide lakes, or vast oceans, there are obstacles that threaten the safety of ships.

[0004] Static obstacles include water bridges, water culverts, tunnels, water windmills, water gas stations, docks, harbors and other facilities. Dynamic obstacles include ships, floating objects, fish and so on.

[0005] Usually, there are many factors to consider when predicting dynamic obstacles, and the prediction results are difficult to achieve the expected accuracy. At present, the traditional real-time navigation risk warning method is mainly based on the real-time dynamic information of other ships obtained by ship-borne radar or sensors and monitors in the water. When predicting dynamic obstacles in the water, it is impossible to consider the interference factors of the water flow in the surrounding environment to calculate the future position of the dynamic obstacle, and the collision prediction of the dynamic obstacle is not accurate enough. Summary of the invention

[0006] In order to solve the problems existing in the background technology, the present invention proposes a water traffic safety warning system.

[0007] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0008] A water traffic safety warning system includes a monitor, a ship end and a computing end. The monitor is arranged in the monitored waters. The ship end includes a signal sending module, a data acquisition module, a data preprocessing module and a response module. The computing end includes a data computing module. The ship end and the computing end are connected via a data transmission module. The modules are specifically:

[0009] A signal sending module receives a computing end command including the location of the monitor, and sends a request signal to the monitor within the target direction through a signal sending device;

[0010] The data acquisition module receives the response signal sent by the monitor, and the response signal includes the water area condition information acquired by the monitor, and the water area condition information includes the flow velocity, water depth, and the location and volume of obstacles within the monitoring range;

[0011] The data preprocessing module determines the difference between the actual position and the expected position of the monitor according to the time difference between the time point when the signal sending module sends the request signal and the time point when the data acquisition module receives the response signal;

[0012] The data transmission module sends the distance between the current ship and the target direction monitor to the data calculation module, and receives the processing result sent by the data calculation module;

[0013] The data calculation module calculates the position of the monitor and the current route of the ship based on the information obtained by the monitor;

[0014] The response module prompts the current navigation of the ship based on the calculation results of the data calculation module.

[0015] Furthermore, the specific method for the data preprocessing module to determine the difference between the actual calculated distance of the monitor and the expected distance is:

[0016] According to the type of request signal sent by the signal sending module, and the time difference between the time when the signal sending module sends the request signal and the time when the data acquisition module receives the response signal ;

[0017] pass Calculate the actual distance between the monitor and the current ship in the current waters, where is the propagation speed of the current signal type in the air;

[0018] Calculate the ratio of the actual distance to the expected distance and set the user preset error to , if the ratio of the actual distance to the expected distance is If the actual distance is within the interval, the content in the computing end command is marked as trusted data. If the ratio of the actual distance to the expected distance is not within this interval, the data in the computing end command is marked as untrusted and transmitted to the computing end through the data transmission module.

[0019] Furthermore, the specific method for the data calculation module to calculate the position of the monitor is:

[0020] The data calculation module sets the maximum limit distance m of the monitor, and determines the current position of the monitor through the water flow mechanics model within the maximum limit distance;

[0021] Based on the continuity equation, momentum equation and energy equation, the position of the detector under the influence of water flow and water depth in the current water area is described , introduce the constraint of maximum limiting distance;

[0022] Define the objective function to find out the influence of water velocity and water depth on the position of the current monitor in the current water area. The objective function is:

[0023] ;

[0024] in Represents the combined factors of water velocity and water depth, ;

[0025] is the water velocity function, is the square of the water velocity;

[0026] Water flow velocity function The specific formula is:

[0027] ;

[0028] in is the Manning roughness coefficient, is the hydraulic radius, is the slope, is the distance factor between the obstacle and the current detector;

[0029] is the water depth influence function, the specific formula is:

[0030] ;

[0031] in is the initial value of water depth, is the scale of water depth attenuation, is the volume factor of the obstacle.

[0032] Furthermore, the method for setting the distance factor between the obstacle and the current monitor and the volume factor of the obstacle is as follows:

[0033] Obtain the obstacle volume and the actual position of the obstacle from the monitor obtained by the monitor from the data preprocessing module, as well as the predicted position of the same monitor in the same period calculated by the data calculation module, and provide the data to the deep learning model for training;

[0034] The deep learning model is based on the flow velocity and depth, and adds the influence factors of the distance from the obstacle to the monitor and the obstacle volume on the obstacle position. Through linear regression fitting, the distance factor between the obstacle and the current monitor and the volume factor of the obstacle are obtained.

[0035] Furthermore, the data calculation module calculates the current route of the ship based on the information obtained by the monitor in the following specific method:

[0036] Based on the positions and sizes of all obstacles acquired by the monitor, a grid map is established, and static obstacles are marked in the grid map;

[0037] For dynamic obstacles, the positions of dynamic obstacles obtained by the monitor are sorted according to time to generate the position changes of dynamic obstacles in time series. The motion trajectory of the current dynamic obstacle is predicted by using a polynomial regression model based on the historical position of any dynamic obstacle. The random interference value is added as a parameter to obtain the predicted positions of a preset number of current dynamic obstacles, thus completing the position prediction of all dynamic obstacles.

[0038] Mark the grids corresponding to the locations where all dynamic obstacles may reach in the future, set a danger score for the obstacle based on the size of the obstacle and the probability of reaching the target location, and mark the danger score on the grid map;

[0039] Based on the predicted results of the future positions of static obstacles and dynamic obstacles, a passage path is planned for the current ship. The 8-neighborhood of the grid where the current ship is located is used as the connection point. The depth-first algorithm is used to calculate the lowest accumulated danger score of the current ship passing through the current waters, and the route with the lowest value is selected as the current ship's route.

[0040] Furthermore, the specific method of using the polynomial regression model to predict the motion trajectory of the current dynamic obstacle is:

[0041] For the current dynamic obstacle, the flow velocity and depth parameters are introduced to calculate the position of the dynamic obstacle in the future. The specific formula is:

[0042] ;

[0043] in is the factor affected by the water flow velocity. The factors affected by water depth are: and is the regression coefficient to be sought, is the time difference between the future time node and the current time node, is the time node, For the The time point of the moment;

[0044] Add disturbance parameters based on the influence of water velocity and depth. The specific formula is:

[0045] ;

[0046] in and As the regression coefficient after disturbance, multiple disturbance parameters are set to obtain the dynamic obstacle in the future. Multiple locations that may appear after a time point.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] The water traffic safety warning system provided by the present invention can obtain information such as water flow velocity, water depth, and location and size of nearby obstacles in the water area. The crew can clearly understand the environmental conditions around the ship, so as to better cope with the complex water environment, and help the crew to reasonably control the speed and navigation direction of the ship to avoid stranding or loss of control due to excessive water flow or insufficient water depth.

[0049] The data calculation module in the water traffic safety early warning system provided by the present invention takes into account the water flow mechanics model, including the continuity equation, momentum equation and energy equation, and combines the influence of water flow velocity, water depth and obstacles when calculating the monitor position. At the same time, the maximum limit distance between the monitor displacement distance and the initial starting point of monitoring is set, and the combination of these factors can more accurately determine the monitor position.

[0050] The data calculation module in the water traffic safety warning system provided by the present invention uses a polynomial regression model to predict the motion trajectory of the current dynamic obstacle, adds a random interference value as a parameter, and combines these factors to more accurately determine the future position of the dynamic obstacle, providing more reliable data support for subsequent ship route simulation and collision prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a system module structure diagram 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] like Figure 1 As shown, the technical solution adopted by the present invention is as follows: a water traffic safety warning system.

[0054] It includes a monitor, a ship end and a computing end. The monitor is deployed in the monitored waters and sends a response signal when receiving a request signal. The monitor can be an IoT sensor, buoy, underwater sonar and other devices deployed on the bridge pier. The ship end includes a signal sending module, a data acquisition module, a data preprocessing module and a response module. The ship end is located on the ship and performs limited calculations. The computing end includes a data computing module. The computing end is deployed in a computing server and can perform large-scale data calculations with strong computing power. The ship end and the computing end are connected through a data transmission module.

[0055] The signal sending module receives a computing end command including the position of the monitor, and simultaneously sends request signals with different names to the monitors within the target direction range through two signal sending devices at different positions.

[0056] The data acquisition module receives the response signal sent by the monitor. The response signal includes the water area condition information obtained by the monitor and the request signal name. The water area condition information includes water flow velocity, water depth, location and size of obstacles within the monitoring range.

[0057] The data acquisition module receives the response signal sent by the monitor. The response signal includes the water condition information obtained by the monitor, the time when the response signal is sent, and the corresponding request signal name. The water condition information includes water flow velocity, water depth, location and size of obstacles within the monitoring range.

[0058] The data transmission module sends the current distance between the ship and the target direction monitor to the data calculation module, and receives the processing result sent by the data calculation module.

[0059] The data calculation module calculates the position of the monitor and the current route of the ship based on the information obtained by the monitor.

[0060] The response module prompts the current navigation of the ship based on the calculation results of the data calculation module.

[0061] When warning the water traffic safety of the current ship, the computing end first sends a signal to the ship end. The computing end commands the position of the monitor, including the relative position of the monitor to the current ship. The computing end predicts the direction of the current ship and sends the monitor information in the direction of the current ship to the ship end corresponding to the current ship.

[0062] The signal sending module receives the command from the computing end, and according to the information from the computing end, simultaneously sends the request signal to the monitor in the target direction through two signal sending devices at different positions. The signal generating devices include radar, high-frequency ray meter, etc. The two signal sending devices can avoid problems such as signal shielding due to errors and environmental factors.

[0063] After the signal sending module sends the request signal, the data acquisition module receives the response signal sent by the monitor. The response signal includes the water area information obtained by the monitor, the response signal sending time and the corresponding request signal name. The water area information includes water flow velocity, water depth, location and volume of obstacles within the monitoring range, etc. Through the operation of the monitor, detailed information of the current water area can be accurately obtained, thereby providing strong data support and guarantee for subsequent related work.

[0064] After the signal sending module sends the request signal and the data acquisition module receives the response signal sent by the monitor, the time point of sending the data and the time point of receiving the data are submitted to the data preprocessing module. The data preprocessing module determines the difference between the actual position and the expected position of the monitor based on the time point when the signal sending module sends the request signal and the time point when the data acquisition module receives the response signal.

[0065] According to the type of request signal sent by the signal sending module, the signal type may include laser signal, electromagnetic wave signal and other signal types. And obtain the time point when the two signal sending modules send the request signal , the time point when the data acquisition module receives the response signal , ,and ;

[0066] Will Obtain the time difference between the two response signals and determine the distance difference between the two signal sending devices and the target detector;

[0067] pass The distance between the two signal sending devices and the target detector is calculated, and the actual distance between the monitor and the current ship in the current waters is obtained through cross positioning, where v is the propagation speed of the current signal type in the air. The actual distance between the current ship and the monitor is determined by the signal type, and the specific location of the monitor can be accurately determined. At the same time, because the ship sails at a slow speed, the distance problem during the ship's movement is not considered.

[0068] Calculate the ratio of the actual distance to the expected distance and set the user preset error to , if the ratio of the actual distance to the expected distance is If the actual distance is within the interval, the content in the computing end command is marked as trusted data. If the ratio of the actual distance to the expected distance is not within this interval, the data in the computing end command is marked as untrustworthy.

[0069] In the actual environment, the position of the monitor will be affected by factors such as water flow and wind, and will deviate from the original position. When the monitor deviates from the original position, unreliable data will be generated when monitoring obstacles in the water area, resulting in position deviation when simulating obstacles, which may easily lead to the risk of ship collision. Therefore, the position of the monitor is confirmed to ensure that the obstacle information obtained by the monitor is reliable.

[0070] If the calculation end accurately judges the position of the monitor, it means that when the data calculation module simulates the position of the monitor, the calculation result is accurate and the distance deviation between the calculated monitor position and the actual detector position is small.

[0071] After adding an untrusted mark to the data in the computing end command, the untrusted mark is sent to the computing end, and the data calculation module of the computing end corrects and locates the position of the monitor.

[0072] The data calculation module sets the maximum limit distance m of the monitor. Within the maximum limit distance, the positions of all monitors are determined by the hydrodynamic model. The maximum limit distance of the monitor is set. Combining these factors can more accurately determine the position of the monitor, providing more reliable data support for subsequent ship route simulation and collision prediction.

[0073] Based on the continuity equation, momentum equation and energy equation, the position of the detector under the influence of water flow and water depth in the current water area is described , introduce the constraint condition of maximum limit distance to create a function.

[0074] Define the objective function to find out the influence of water velocity and water depth on the position of the current monitor in the current water area. The objective function is:

[0075] ;

[0076] in Represents the combined factors of water velocity and water depth, .

[0077] is the water velocity function, It is the square of the water flow velocity, indicating the magnitude of the flow velocity. The greater the flow velocity, the greater the impact on the position change of the monitor. The water flow velocity function The specific formula is:

[0078] ;

[0079] in is the Manning roughness coefficient, is the hydraulic radius, is the slope, is the distance factor between the obstacle and the current detector.

[0080] It is the water depth effect function. When the water depth is shallow, the water flow is more violent and the monitor becomes unstable. When the water depth is deep, the flow rate is slower and the monitor position is stable. The water depth effect function The specific formula is:

[0081] ;

[0082] in is the initial value of water depth, is the scale of water depth attenuation, is the volume factor of the obstacle.

[0083] Method for setting the distance factor between the obstacle and the current monitor and the volume factor of the obstacle:

[0084] First, the obstacle volume data obtained by the monitor and the actual position distance information between the obstacle and the monitor are accurately extracted from the data preprocessing module. At the same time, the predicted position data calculated by the data calculation module for the same period and the same monitor is obtained. Then, these data are fully provided to the deep learning model for its training process.

[0085] The deep learning model not only takes into account the basic factors such as water flow velocity and depth, but also includes the distance between the obstacle and the monitor and the volume of the obstacle as the influencing factors on the location of the obstacle. The data is fitted with the help of the mathematical method of linear regression to obtain the distance factor that can characterize the relationship between the obstacle and the current monitor, and the volume factor that is closely related to the volume of the obstacle.

[0086] After the position of the monitor is calculated by the data calculation module, the current route of the ship is calculated based on the information obtained by the monitor.

[0087] Based on the positions and sizes of all obstacles acquired by the monitor, a grid map is established, and static obstacles are marked in the grid map.

[0088] Static obstacles include water bridges, water culverts, tunnels, water power windmills, water gas stations, docks, harbors and other facilities. Static obstacles can be marked in the grid map according to their size and relative position to the monitor.

[0089] For dynamic obstacles, the positions of dynamic obstacles obtained by the monitor are sorted according to time to generate the position changes of dynamic obstacles in a time series. The motion trajectory of the current dynamic obstacle is predicted by using a polynomial regression model based on the historical position of any dynamic obstacle, and random interference values ​​are added as parameters.

[0090] For the current dynamic obstacle, the flow velocity and depth parameters are introduced to calculate the position of the dynamic obstacle in the future. The specific formula is:

[0091]

[0092] in is the factor affected by the water flow velocity. The factors affected by water depth are: and is the regression coefficient to be sought, is the time difference between the future time node and the current time node, is the time node, For the The time node of the moment.

[0093] Add disturbance parameters based on the influence of water velocity and depth. The specific formula is:

[0094]

[0095] in and As the regression coefficient after disturbance, multiple disturbance parameters are set to obtain the dynamic obstacle in the future. Multiple positions that may appear after a time node. Adding disturbance factors can reasonably predict the disturbance of dynamic obstacles with power size changes, and add expected simulation of the change of the movement direction of dynamic obstacles.

[0096] The predicted positions of a preset number of current dynamic obstacles are obtained by the above method, and the position prediction of all dynamic obstacles is completed by this method. The randomness of dynamic obstacles can be predicted, ensuring that the disordered interference of dynamic obstacles is considered during the navigation of the current ship, and the possible positions of dynamic obstacles can be avoided comprehensively.

[0097] Mark the grids corresponding to the locations where all dynamic obstacles may reach in the future. Set a danger score for the obstacle based on the size of the dynamic obstacle and the probability of reaching the target location, and mark the danger score in the grid map.

[0098] Based on the predicted results of the future positions of static obstacles and dynamic obstacles, a passage path is planned for the current ship. The 8-neighborhood of the grid where the current ship is located is used as the connection point, and the minimum accumulated danger score of the current ship passing through the current waters is calculated through the depth-first algorithm.

[0099] During the exploration process, when the depth-first algorithm encounters a new node, it will first select one of the adjacent nodes that have not been visited for in-depth exploration, mark it as visited, and recursively apply the depth-first strategy on the node. If the target cannot be found on a branch, it will backtrack to the previous step and select another adjacent node that has not been visited to continue exploring. This process is repeated, and all routes from the current ship's location to the target point are gradually constructed through continuous in-depth exploration and backtracking. A hazard score is added during the exploration process, and the route with the lowest hazard score is selected as the current ship's route.

[0100] After the calculation of the current ship's route is completed, the response module on the ship side prompts the current ship's navigation based on the calculation results of the data calculation module, ensuring that the current ship's route can avoid static and dynamic obstacles in the current waters with the lowest risk.

[0101] Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. 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 water traffic safety warning system, comprising a monitor, a vessel end and a computing end, wherein the monitor is arranged in the monitored waters, and is characterized in that: The ship end includes a signal sending module, a data acquisition module, a data preprocessing module and a response module, the computing end includes a data computing module, and the ship end and the computing end are connected via a data transmission module; The signal sending module receives a computing end command, the computing end command includes the position of the monitor, and simultaneously sends request signals with different names to the monitors within the target direction range through two signal sending devices at different positions; The data acquisition module receives the response signal sent by the monitor. The response signal includes the water area condition information obtained by the monitor, the time when the response signal is sent, and the corresponding request signal name. The water area condition information includes the flow velocity, water depth, and the location and volume of obstacles within the monitoring range. The data preprocessing module determines the difference between the actual position and the expected position of the monitor according to the time point when the signal sending module sends the request signal and the time point when the data acquisition module receives the response signal; The data transmission module sends the distance between the current ship and the target direction monitor to the data calculation module, and receives the processing result sent by the data calculation module; The data calculation module calculates the position of the monitor and the current route of the ship based on the information obtained by the monitor; The response module prompts the current navigation of the ship based on the calculation results of the data calculation module.

2. A water traffic safety warning system according to claim 1, characterized in that: The specific method of the data preprocessing module to determine the difference between the actual position and the expected position of the monitor is: According to the type of request signal sent by the signal sending module, and the time point when the two signal sending modules send the request signal , the time point when the data acquisition module receives the response signal , ,and ; Will Obtain the time difference between the two response signals and determine the distance difference between the two signal sending devices and the target detector; pass and The distances between the two signal sending devices and the target detector are calculated, and the actual distance between the monitor and the current ship in the current waters is obtained through cross positioning, where v is the propagation speed of the current signal type in the air; Calculate the ratio of the actual distance to the expected distance and set the user preset error to , if the ratio of the actual distance to the expected distance is If the actual distance is within the interval, the content in the computing end command is marked as trusted data. If the ratio of the actual distance to the expected distance is not within this interval, the data in the computing end command is marked as untrusted and transmitted to the computing end through the data transmission module.

3. A water traffic safety warning system according to claim 1, characterized in that: The specific method for the data calculation module to calculate the position of the monitor is: The data calculation module sets the maximum limit distance m of the monitor, and determines the current position of the monitor through the water flow mechanics model within the maximum limit distance; Based on the continuity equation, momentum equation and energy equation, the position of the detector under the influence of water flow and water depth in the current water area is described , introduce the constraint of maximum limiting distance; Define the objective function to find out the influence of water velocity and water depth on the position of the current monitor in the current water area. The objective function is: ； in Represents the combined factors of water velocity and water depth, ; is the water velocity function, is the square of the water velocity; Water flow velocity function The specific formula is: ； in is the Manning roughness coefficient, is the hydraulic radius, is the slope, is the distance factor between the obstacle and the current detector; is the water depth influence function, the specific formula is: ； in is the initial value of water depth, is the scale of water depth attenuation, is the volume factor of the obstacle.

4. A water traffic safety warning system according to claim 3, characterized in that: Method for setting the distance factor between the obstacle and the current monitor and the volume factor of the obstacle: Obtain the obstacle volume and the actual position of the obstacle from the monitor obtained by the monitor from the data preprocessing module, as well as the predicted position of the same monitor in the same period calculated by the data calculation module, and provide the data to the deep learning model for training; The deep learning model is based on the flow velocity and depth, and adds the influencing factors of the distance between the obstacle and the monitor and the obstacle volume on the obstacle position. Through linear regression fitting, the distance factor between the obstacle and the current monitor and the volume factor of the obstacle are obtained.

5. A water traffic safety warning system according to claim 4, characterized in that: The specific method by which the data calculation module calculates the current route of the vessel based on the information obtained by the monitor is as follows: Based on the positions and sizes of all obstacles acquired by the monitor, a grid map is established, and static obstacles are marked in the grid map; For dynamic obstacles, the positions of dynamic obstacles obtained by the monitor are sorted according to time to generate the position changes of dynamic obstacles in time series. The motion trajectory of the current dynamic obstacle is predicted by using a polynomial regression model based on the historical position of any dynamic obstacle. The random interference value is added as a parameter to obtain the predicted positions of a preset number of current dynamic obstacles, thus completing the position prediction of all dynamic obstacles. Mark the grids corresponding to the locations where all dynamic obstacles may reach in the future, set a danger score for the obstacle based on the size of the obstacle and the probability of reaching the target location, and mark the danger score on the grid map; Based on the predicted results of the future positions of static obstacles and dynamic obstacles, a passage path is planned for the current ship. The 8-neighborhood of the grid where the current ship is located is used as the connection point. The depth-first algorithm is used to calculate the lowest accumulated danger score of the current ship passing through the current waters, and the route with the lowest value is selected as the current ship's route.

6. A water traffic safety warning system according to claim 5, characterized in that: The specific method of using the polynomial regression model to predict the motion trajectory of the current dynamic obstacle is: For the current dynamic obstacle, the flow velocity and depth parameters are introduced to calculate the position of the dynamic obstacle in the future. The specific formula is: ； in is the factor affected by the water flow velocity. The factors affected by water depth are: and is the regression coefficient to be sought, is the time difference between the future time node and the current time node, is the time node, For the The time point of the moment; Add disturbance parameters based on the influence of water velocity and depth. The specific formula is: ； in and As the regression coefficient after disturbance, multiple disturbance parameters are set to obtain the dynamic obstacle in the future. Multiple locations that may appear after a time point.

Images