A ship night navigation supervision system

Abstract

The present invention discloses a nighttime ship navigation monitoring system, comprising: an acquisition control module for acquiring motion images and environmental images of a ship sailing at night, and detecting and tracking light spots in the motion images and environmental images to obtain trajectory information of the ship sailing at night; a monitoring module for receiving the trajectory information, monitoring the ship's navigation speed and navigation environment based on the trajectory information, and feeding back the obtained first monitoring result to an early warning module; the early warning module includes an anti-collision warning module and a speed warning module, the anti-collision warning module using a spatial and temporal urgency model to assess collision risks and issue early warnings for the navigation environment in the first monitoring result; and a speed warning module using a ship speed control range module to compare the actual ship speed in the first monitoring result with a speed range threshold, thereby assessing the actual ship speed and issuing early warnings. The present invention improves the safety of ships during navigation by quantitatively monitoring the ship's status.

Description

Technical Field

[0001] The present invention relates to the technical field of ship navigation supervision, and in particular to a ship nighttime navigation supervision system. Background Art

[0002] Navigation safety has always been a hot topic of global concern. Human factors, particularly fatigue, have been identified as a major contributing factor in maritime accidents. Fatigue-induced operational errors by ship watchkeepers are the root cause of most maritime accidents. Ship detection and risk identification are crucial for ensuring safe navigation, reducing maritime accidents, and enabling information-based ship management. While there are numerous methods for ship detection and tracking, and complex and diverse detection equipment such as radar, AIS (Automatic Identification System), and video, the effectiveness of all types of monitoring is affected in challenging environments. For example, at night, visibility is low and vision is obscured. While radar can aid navigation, its targets are not intuitive enough to identify the target object. AIS signals are also susceptible to loss and are affected by onboard equipment.

[0003] There are also many methods for identifying ship risks, and the equipment is complex and diverse. For example, traditional equipment radar and ARPA (Automatic Radar Plotting Aid) can identify collision risks by calculating TCPA and DCPA. At the same time, they can also test the ship and make judgments on the ship's collision avoidance behavior. However, traditional radar still has shortcomings such as limited resolution and delays in processing target information. In addition, the driver often lacks standard supervision of the safety of the ship's speed. Summary of the Invention

[0004] The purpose of the present invention is to overcome the above-mentioned technical deficiencies and provide a ship night navigation supervision system to solve the problems in the existing technology of limited resolution ability, delayed processing of target information, and lack of systematic standard supervision based on the driver's experience judgment of traditional methods of identifying ship risks.

[0005] In order to solve the above problems, the present invention discloses a ship night navigation supervision system, comprising: an acquisition control module, a monitoring module and an early warning module;

[0006] The acquisition control module is used to acquire motion images and environmental images of a ship sailing at night, detect and track light spots in the motion images and environmental images, obtain trajectory information of the ship sailing at night, and send the trajectory information to the monitoring module;

[0007] The monitoring module is configured to receive the trajectory information, monitor the navigation speed and navigation environment of the vessel according to the trajectory information, obtain a first monitoring result, and feed the first monitoring result back to the early warning module;

[0008] The warning module includes an anti-collision warning module and a speed warning module. The anti-collision warning module performs collision risk assessment and warning on the navigation environment in the first monitoring result through a spatial and temporal urgency model; the speed warning module compares the actual speed of the ship in the first monitoring result with the speed range threshold through a ship speed control range module, and evaluates and warns the actual speed of the ship.

[0009] In some embodiments, the acquisition control module detects and tracks the light spots in the motion image and the environment image, including:

[0010] Using a preset LOG operator edge detection algorithm to perform spot detection on the motion image and the environment image to obtain a ship moving image and a spot image;

[0011] The light spot image is tracked using a preset Kalman filter algorithm, and the light spot image is matched with the ship driving image to obtain the trajectory information of the ship sailing at night.

[0012] In some embodiments, the performing spot detection on the motion image and the environment image using a preset LOG operator edge detection algorithm includes:

[0013] grayscale processing is performed on the motion image and the environment image to obtain a grayscale image;

[0014] Performing a convolution operation on the grayscale image using a preset convolution operation method to obtain a convolution image result;

[0015] According to the convolution image result, the convolution image result is filtered using a preset spot image grayscale threshold to obtain an initial spot image;

[0016] Amplifying the initial spot image using a preset maximum value filtering method to obtain an amplified spot image;

[0017] Determining the ship spot and image background in the amplified light spot image according to a change in pixel values ​​between the amplified light spot image and the initial light spot image, and corroding the image background to obtain an actual light spot image;

[0018] determining the center position and radius of the ship light spot according to the difference relationship between the amplified light spot image and the actual light spot image;

[0019] Based on the ship light spot radius and the distance between the center positions of two adjacent light spots, overlapping light spots are determined, and the overlapping light spots are removed to obtain a target light spot image.

[0020] In some embodiments, the tracking of the light spot image using a preset Kalman filter algorithm and the matching of the light spot image with the ship traveling image to obtain the trajectory information of the ship sailing at night include:

[0021] Obtain the ship's navigation speed and position, and construct the state vector of the navigation speed and position;

[0022] Performing matrix transformation on the state vector to determine a filter matrix set, wherein the filter matrix set includes a state transfer matrix, a control input matrix, a system control quantity, a system control quantity matrix, and a corresponding covariance matrix;

[0023] Based on the filtering parameters, predict the state vector at the next moment and determine the initial prediction covariance matrix;

[0024] Determining a gain of a Kalman filter according to the predicted covariance matrix;

[0025] Determining a target prediction covariance matrix according to the gain;

[0026] According to the target prediction covariance matrix, the trajectory information of the ship sailing at night is determined.

[0027] In some embodiments, the anti-collision warning module includes a risk assessment module and a first warning module;

[0028] The risk assessment module is configured to perform a collision risk assessment on the navigation environment in the first monitoring result using a spatial and temporal urgency model to obtain a collision assessment result, and if the assessment result indicates that the ship is at risk of collision, the collision assessment result is sent to the early warning module;

[0029] The first warning module includes a first light warning unit and a sound warning unit. The first warning module is used to receive the evaluation result and respond to the evaluation result through the first light warning unit and the sound warning unit.

[0030] In some embodiments, the risk assessment module is configured to perform a collision risk assessment on the navigation environment in the first monitoring result using a spatial and temporal urgency model to obtain an assessment result, including:

[0031] A multi-layered elliptical ship field is established based on the length and width dimensions of the neighboring ships of the target ship;

[0032] Assigning weights to the multiple layers of ship fields, and determining a weight value of the superimposed layer according to the weight value of each layer of ship fields;

[0033] Determine the area of ​​the superposition region according to the superposition layer weight value;

[0034] Establishing a spatial urgency model for ship collision based on the area of ​​the superimposed region;

[0035] Determine the collision time of the ships according to the relative position vector sum and relative speed vector sum of the target ship and the adjacent ships;

[0036] According to the relationship between the risk of ship collision and the urgency of collision time, a time urgency model of ship collision is established;

[0037] Based on catastrophe theory, a collision risk calculation model between the target ship and adjacent ships is established according to the spatial urgency model and temporal urgency model of ship collision.

[0038] In some embodiments, the speed warning module includes a speed control module and a second warning module;

[0039] The speed control module is configured to compare the actual ship speed in the first monitoring result with the speed range threshold through the ship speed control range module, evaluate the actual ship speed, obtain a speed evaluation result, and send the speed evaluation result to the second early warning module if the speed evaluation result is not within the speed range threshold;

[0040] The second warning module includes a second light warning unit, and the second warning module is used to receive the speed assessment result and respond to the assessment result through the second light warning unit.

[0041] In some embodiments, the comparing the actual ship speed in the first monitoring result with the speed range threshold by the ship speed control range module to evaluate the actual ship speed includes:

[0042] Based on the ship's sinking model and the limitation of the channel water depth on the ship's excess water depth, the ship's maximum speed control model is determined;

[0043] According to the influence of wind drift on ship speed, a minimum speed control model of the ship is established;

[0044] Based on the maximum speed control model and the minimum speed control model, with the ship spacing as the constraint condition and according to the car-following theory, the maximum speed threshold and the minimum speed threshold of the ship are determined.

[0045] In some embodiments, the acquisition control module includes at least an automatic identification system, a wind sensor, a water flow sensor, an information receiver, an electronic chart display and information system, and a radar identification unit.

[0046] In some embodiments, the monitoring module includes at least a panoramic camera, a light sensor, an infrared detector, an ultrasonic detector, and a night flight device.

[0047] Compared with the prior art, the ship night navigation supervision system provided by the present invention first obtains the trajectory information of the ship navigation at night through the acquisition control module, and sends the trajectory information to the monitoring module, monitors and manages the ship sailing at night through the monitoring module, and feeds back the monitoring results to the early warning module in real time. Finally, the early warning module evaluates the collision risk of the ship through the anti-collision warning module. If there is a collision risk, a warning is issued in time to enable the ship driver to make correct operations; at the same time, the speed of the ship is monitored through the speed warning module. If the actual speed of the ship is not within the range of the maximum speed threshold and the minimum speed threshold, the speed abnormality is issued in time to enable the ship driver to make correct operations; the present invention improves the safety of the ship during navigation by quantitatively monitoring the ship status. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 This is a structural diagram of an embodiment of a ship nighttime navigation monitoring system provided by the present invention;

[0049] Figure 2 This is a flow chart of an embodiment of the working process of the acquisition control module in the ship night navigation monitoring system provided by the present invention;

[0050] Figure 3 This is a schematic diagram of an implementation of calculating the light spot overlapping area in the ship night navigation monitoring system provided by the present invention;

[0051] Figure 4 This is a structural diagram of an embodiment of an anti-collision warning module in the ship nighttime navigation monitoring system provided by the present invention;

[0052] Figure 5 This is a schematic diagram of an embodiment of ship field stratification in the ship night navigation monitoring system provided by the present invention;

[0053] Figure 6 This is a structural diagram of an embodiment of a speed warning module in the ship night navigation monitoring system provided by the present invention;

[0054] Figure 7 It is a schematic diagram of an embodiment of the application of the car-following theory in the ship night navigation monitoring system provided by the present invention. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0056] The embodiment of the present invention provides a ship night navigation monitoring system 1, comprising: an acquisition control module 11, a monitoring module 12 and an early warning module 13;

[0057] The acquisition control module 11 is used to acquire motion images and environmental images of a ship sailing at night, detect and track light spots in the motion images and environmental images, obtain trajectory information of the ship sailing at night, and send the trajectory information to the monitoring module;

[0058] The monitoring module 12 is configured to receive the trajectory information, monitor the navigation speed and navigation environment of the vessel according to the trajectory information, obtain a first monitoring result, and feed the first monitoring result back to the early warning module;

[0059] The warning module 13 includes an anti-collision warning module 131 and a speed warning module 132. The anti-collision warning module performs collision risk assessment and warning on the navigation environment in the first monitoring result through a spatial and temporal urgency model; the speed warning module compares the actual speed of the ship in the first monitoring result with the speed range threshold through a ship speed control range module, and evaluates and warns the actual speed of the ship.

[0060] In an embodiment of the present invention, first, the trajectory information of the ship sailing at night is obtained through the acquisition control module, and the trajectory information is sent to the monitoring module. The monitoring module monitors and manages the ship sailing at night, and the monitoring results are fed back to the early warning module in real time. Finally, the early warning module evaluates the collision risk of the ship through the anti-collision warning module. If there is a collision risk, a warning is issued in time to enable the ship driver to make correct operations; at the same time, the speed of the ship is monitored through the speed warning module. If the actual speed of the ship is not within the range of the maximum speed threshold and the minimum speed threshold, the speed abnormality is issued in time to enable the ship driver to make correct operations; the present invention improves the safety of the ship during navigation by quantitatively monitoring the ship status.

[0061] It should be noted that the ship night navigation monitoring system provided in this embodiment further includes a display module, wherein the display module includes a display device, and the display device is installed in the front instrument panel of the ship's central control room.

[0062] Furthermore, the monitoring module includes a night navigation device, a light sensor, a panoramic camera, an image processing module, a ship light detection module, a ship light tracking module, an infrared detector and an ultrasonic detector; wherein the night navigation device is electrically connected to the light sensor, which is in turn electrically connected to the panoramic camera. The night navigation device can penetrate the darkness and clearly see the water conditions within a few nautical miles around the ship at night, enabling the driver to more accurately analyze the water conditions and assess risks; the light sensor can sense the brightness changes of the ship's lights and output a weak electrical signal, which can be amplified by a simple circuit to more clearly detect the light spot of the ship; the panoramic camera includes eight fixed-point image sensors for 360° With the monitoring function without blind spots, the eight image sensors can be integrated into a complete 360-degree monitoring motion image and environmental image through stitching technology; the image processing module is connected to the panoramic camera, and is composed of a clarity optimization unit and a continuous shooting unit. The clarity optimization unit can optimize the clarity of the navigation situation ahead and the motion image and environmental image of the ship captured by the panoramic camera, ensuring that the driver and the ship's driving computer can clearly and intuitively observe the information ahead. The continuous shooting unit can continuously shoot and process the road conditions ahead and ship information, ensuring that the information ahead can be collected in real time to ensure the safety of the driver and the ship.

[0063] In some embodiments, see Figure 2 The acquisition control module detects and tracks the light spots in the motion image and the environment image, including:

[0064] S201, performing spot detection on the motion image and the environment image using a preset LOG operator edge detection algorithm to obtain a ship moving image and a spot image;

[0065] S202: Track the light spot image using a preset Kalman filter algorithm, and match the light spot image with the ship's traveling image to obtain track information of the ship's navigation at night.

[0066] In this embodiment, by inputting the motion image and environmental image processed by the image processing module, the light spot is first detected based on preprocessing (including image cropping and grayscale conversion). The light spot is then tracked based on the light spot detection result (the X / Y coordinates of the light spot in the image) to obtain the ship's nighttime navigation trajectory.

[0067] In some embodiments, the performing spot detection on the motion image and the environment image using a preset LOG operator edge detection algorithm includes:

[0068] grayscale processing is performed on the motion image and the environment image to obtain a grayscale image;

[0069] Performing a convolution operation on the grayscale image using a preset convolution operation method to obtain a convolution image result;

[0070] According to the convolution image result, the convolution image result is filtered using a preset spot image grayscale threshold to obtain an initial spot image;

[0071] Amplifying the initial spot image using a preset maximum value filtering method to obtain an amplified spot image;

[0072] Determining the ship spot and image background in the amplified light spot image according to a change in pixel values ​​between the amplified light spot image and the initial light spot image, and corroding the image background to obtain an actual light spot image;

[0073] determining the center position and radius of the ship light spot according to the difference relationship between the amplified light spot image and the actual light spot image;

[0074] Based on the ship light spot radius and the distance between the center positions of two adjacent light spots, overlapping light spots are determined, and the overlapping light spots are removed to obtain a target light spot image.

[0075] In this embodiment, the LOG operator can be expressed by the following formula: ,in, is the spatial standard deviation of the Gaussian function; is the Laplace operator; is a Gaussian function.

[0076] Furthermore, since the original image captured by the camera is RGB, it must first be converted to grayscale. Since the spot size is undetermined, filters of different scales can be used to detect edges in the image. The filter scale depends on the convolution kernel parameter σ, so equally spaced vectors are constructed for use in filters of different scales.

[0077] Furthermore, the specific process of the preset convolution operation method is: first, a Gaussian Laplace convolution operation is performed in the vertical direction; secondly, a Gaussian convolution operation is performed in the horizontal direction, and the image after the convolution operation is subjected to a Gaussian convolution operation in the horizontal direction; then, a Gaussian Laplace convolution operation is performed in the horizontal direction; finally, the convolution operation results in the two directions are summed to obtain a convolution image result.

[0078] Furthermore, the selection of the grayscale threshold can be determined based on the size and number of light spots. Specifically, if the threshold is too small, the detected light spots may contain other types of lights, resulting in an excessive number of light spots and making it difficult to effectively track them. If the threshold is too large, the number of detected light spots is too small, making it impossible to detect and subsequently track ships.

[0079] Furthermore, the overlapping spots can be determined as follows: first calculate the sum of the two spot radii , the absolute value of the difference between the radii The distance between the two light spot centers .like , indicating that the small spot is located in the ring of the large spot, and the small spot can be ignored; if , indicating that there is no overlap, otherwise calculate the overlapping area between the two spots. The calculation of the spot overlapping area is shown in the figure below. Figure 3 As shown in the figure 、 and are the radius of the two light spots and the distance between the centers of the light spots respectively; on this basis, the ratio of the overlapping area to the area of ​​each light spot is calculated respectively. If the ratio is greater than the set threshold, it is considered that the light spots overlap. Specifically, in this embodiment, the initial value is 0.1.

[0080] In some embodiments, the tracking of the light spot image using a preset Kalman filter algorithm and the matching of the light spot image with the ship traveling image to obtain the trajectory information of the ship sailing at night include:

[0081] Obtain the ship's navigation speed and position, and construct the state vector of the navigation speed and position;

[0082] Performing matrix transformation on the state vector to determine a filter matrix set, wherein the filter matrix set includes a state transfer matrix, a control input matrix, a system control quantity, a system control quantity matrix, and a corresponding covariance matrix;

[0083] Based on the filtering parameters, predict the state vector at the next moment and determine the initial prediction covariance matrix;

[0084] Determining a gain of a Kalman filter according to the predicted covariance matrix;

[0085] Determining a target prediction covariance matrix according to the gain;

[0086] According to the target prediction covariance matrix, the trajectory information of the ship sailing at night is determined.

[0087] In this embodiment, the Kalman filter algorithm is first used to track the light spot; then the target light spot is matched with the motion image and the environment image; finally, the unstable light spot is detected and the new light spot is updated to obtain the trajectory of the ship.

[0088] In a specific embodiment of the present invention, the algorithm steps of the spot tracking module are as follows:

[0089] Step 1: Construct the state vector of ship position and speed. Perform Kalman filter tracking on a single frame of video image at a given moment. Assumptions: and They are The position of the center of the light spot at that moment; and They are and Speed ​​in direction; and They are and The acceleration in the direction. The vector used to describe the state of the moving target is defined as:

[0090] (1)

[0091] Observation vector for:

[0092] (2)

[0093] The state transition matrix is ​​determined from this , control input matrix , The amount of control over the system at all times , system measurement matrix And its corresponding covariance matrix is:

[0094] (3)

[0095] Once the filter parameters are determined, the process of using Kalman filtering to achieve target tracking is as follows:

[0096] First, predict the state at the next moment and calculate the covariance matrix,

[0097] (4)

[0098] After updating and adding observation value correction, the state variable value matrix and covariance matrix are obtained.

[0099] (5)

[0100] The gain of the Kalman filter is calculated as

[0101] (6)

[0102] After the prediction and update of equations (5) and (6), the correction value at the next moment can be obtained. The above two steps are repeated continuously to realize Kalman filter tracking.

[0103] Step 2: Match the light spots of the motion image and the environment image.

[0104] Let M and N be two frames of images to be matched, is the characteristic point of p light spots in M, is the feature point of p spots in N. Construct the similarity matrix D of the feature points of M and N spots, the elements of this matrix are (7)

[0105] in for and The distance between them.

[0106] make for Whether and If the condition matches,

[0107] (8)

[0108] Setting up the objective function

[0109] (9)

[0110] In order to achieve one-to-one matching of ship light spots, the following constraints are set:

[0111] (10)

[0112] Therefore, the Hungarian algorithm is used to solve the optimal matching problem of video spots, which is a 0-1 programming problem. The corresponding mathematical model is

[0113] (11)

[0114] Step 3: Detect unstable light spots and update new light spots.

[0115] First, Kalman filtering is performed on the tracking spot of the previous frame of video image, as shown in step 1. Then, the observation result and the measurement result are matched with the video image spot, as shown in step 2. The spot is determined to be retained based on the matching result and the set distance threshold. The newly detected spot that is not matched is regarded as the new spot. At the same time, the original spot that has not been matched for a number of times after reaching the threshold is regarded as a spot that has left the video monitoring area and is deleted.

[0116] In some embodiments, see Figure 4 The anti-collision warning module 131 includes a risk assessment module 131a and a first warning module 131b;

[0117] The risk assessment module 131a is configured to perform a collision risk assessment on the navigation environment in the first monitoring result using a spatial and temporal urgency model to obtain a collision assessment result. If the assessment result indicates that the ship has a collision risk, the collision assessment result is sent to the early warning module.

[0118] The first warning module 131b includes a first light warning unit and a sound warning unit. The first warning module is used to receive the evaluation result and respond to the evaluation result through the first light warning unit and the sound warning unit.

[0119] In this embodiment, the risk assessment module is electrically connected to the first warning module; the risk assessment module collects AIS data from other ships and calculates the ship field. Based on the values ​​of the superimposed area of ​​the actual ship field and the superimposed area of ​​the ship field when multiple ships collide, the module first establishes a collision space urgency model, and then establishes a collision time urgency model based on the relative positions and relative speeds of other ships. Finally, according to the catastrophe theory, the ship collision risk calculation based on time and space urgency is realized.

[0120] In a specific embodiment, the algorithm steps of the risk assessment module are as follows:

[0121] Step 1: Establish a spatial urgency model for ship collisions.

[0122] Obtain the length of other ships through AIS , ship width , the ship domain is equivalent to the domain long axis is the ship length , the short axis is the width of the ship The elliptical ship area is divided into different regions from the center to the outside. The stratification effect is as follows: Figure 5 The stratification standards are shown in the following table (Ship field stratification standards and weight assignment), and weight area assignment is performed on each layer of the elliptical ship field.

[0123]

[0124] According to the ship domain regional stratification and ship internal level weight assignment, the cross-addition method is used to calibrate the 25 ship domain overlay layer weights, as shown in the following table (elliptical ship domain overlay layer weights).

[0125]

[0126] Therefore, the calculation method of the layered weighted superposition area of ​​the elliptical fields of the two ships is as follows:

[0127] (12)

[0128] Where, It is the superimposed area of ​​the ship field; For a ship Layer and another ship The overlapping area of ​​the layers; The weights of different levels are superimposed. The collision space is standardized by calculating the superimposed area of ​​the two ships when they collide in the same direction. The spatial urgency model of ship collision is established as follows: (13)

[0129] Where, for The space urgency of ship collision at the moment, ranging from 0 to 1; for The overlapping area of ​​the two ships' territories at that moment; for The overlapping area of ​​two ships colliding at the same time and direction.

[0130] Step 2: Establish a time urgency model for ship collision.

[0131] By extracting data from the AIS system, the ship's position, speed and relative direction are obtained. The position of vessel A (own vessel) at the moment is , the position of vessel B is ,but The relative position of ship A to ship B at time

[0132] (14)

[0133] set up The speed of ship A at the moment is , heading is , the speed of ship B is , heading is ,but The relative speed of ship A to ship B at time

[0134] (15)

[0135] The collision time of the ships is solved according to the relative position vector and relative speed vector of the ships, as shown below

[0136] (16)

[0137] Where, is the time of ship collision; for The relative position vector of ship A to ship B at time t; for The distance between ships at the moment, n mile; for The minimum distance between ship A and ship B at the moment of collision, nmile; for The relative speed of ship A to ship B at time t, kn.

[0138] Finally, based on the functional relationship between ship collision risk and collision time, an exponential function is used to establish a time urgency model for ship collision.

[0139] (17)

[0140] Where, for The time urgency of ship collision at moment, ranging from 0 to 1; is the ship collision time, min; The adjustment coefficient is not less than 0 and can be fitted based on data statistics or expert consultation.

[0141] Step 3: Establish a multi-ship collision risk assessment method based on catastrophe theory.

[0142] According to the spatial urgency model and temporal urgency model of ship collision, a collision risk calculation model of two ships is established based on catastrophe theory.

[0143] (18)

[0144] Where, for Ship collision risk value at each moment; for The space urgency of ship collision at the moment, ranging from 0 to 1; for The time urgency of ship collision, ranging from 0 to 1.

[0145] Before establishing a risk assessment model, the following factors must be considered: ① For any single ship, the maximum collision risk assessment is 1, i.e., the risk in the event of a collision, and other collision risks should not be added; ② For any single ship, if no collision occurs with another ship, the collision risk of that ship should be less than 1 and greater than the maximum collision risk between that ship and any other ship; ③ The collision risk assessment value of any single ship should not be less than 0. This allows for the establishment of a collision risk calculation model for multiple ships.

[0146] (19)

[0147] Where, For ships exist The overall collision risk value with other ships at all times; They are Different time, different ship to ship the risk of collision, and ; The number of ships that the current ship will encounter.

[0148] In some embodiments, see Figure 6 The speed warning module 132 includes a speed control module 132a and a second warning module 132b;

[0149] The speed control module 132a is configured to compare the actual ship speed in the first monitoring result with the speed range threshold through the ship speed control range module, evaluate the actual ship speed, obtain a speed evaluation result, and send the speed evaluation result to the second early warning module if the speed evaluation result is not within the speed range threshold;

[0150] The second warning module 132b includes a second light warning unit, and the second warning module is used to receive the speed assessment result and respond to the assessment result through the second light warning unit.

[0151] In this embodiment, the speed control module 132a is electrically connected to the second warning module 132b; by considering the limitation of the water depth of the channel water area, that is, considering the influence of the ship sinking amount, a maximum speed control model of the ship is established; then by considering the limitation of the ship channel width, that is, the influence of the wind drift amount, a minimum speed control model of the ship is established; finally, by considering the distance between ships, that is, according to the following theory, a speed control range model of the ship is established.

[0152] In some embodiments, the comparing the actual ship speed in the first monitoring result with the speed range threshold by the ship speed control range module to evaluate the actual ship speed includes:

[0153] Based on the ship's sinking model and the limitation of the channel water depth on the ship's excess water depth, the ship's maximum speed control model is determined;

[0154] According to the influence of wind drift on ship speed, a minimum speed control model of the ship is established;

[0155] Based on the maximum speed control model and the minimum speed control model, with the ship spacing as the constraint condition and according to the car-following theory, the maximum speed threshold and the minimum speed threshold of the ship are determined.

[0156] In this embodiment, by connecting AIS equipment, information collection module, ECDIS and other equipment to obtain ship navigation data, nautical chart data and water area data, the surplus water depth of different ships is analyzed based on the surplus water depth determination method of the "Overall Design Code for Seaports". , calculate the ship's surplus water depth as shown in formula (20)

[0157] (20)

[0158] (twenty one)

[0159] (twenty two)

[0160] Actual navigable water depth for ships As shown in formula (23)

[0161] (twenty three)

[0162] Where, is the designed ship's fully loaded draft, in m, which is input from AIS; The required navigation depth, m, is input from ECDIS; is the designed water depth of the channel, in m, which is input by ECDIS; is the amount of hull sinking when the ship is sailing, m; is the minimum excess depth under the keel during navigation, m; is the wave excess depth, in m. Generally, it is 0 in fully sheltered waters, 0.15m in semi-sheltered waters, and 0.3m in unsheltered waters. This parameter is input by the information acquisition module; The ship loading trim margin depth, in m, is 0.15 m for bulk carriers and oil tankers. This parameter is input by the information collection module; The reserve depth for siltation, in meters. For ports without siltation, the reserve depth for siltation can be ignored. For ports with siltation, the reserve depth for siltation should not be less than 0.4 meters. This parameter is input by the information collection module. is the tide height, m, which is input by the information acquisition module; The chart water depth, m, is input from ECDIS.

[0163] The BARRASS model is selected as the ship's sinking model, and the safety margin of the model is reduced by 10%. The calculation model of the ship's sinking is:

[0164] (twenty four)

[0165] Where, is the ship's square coefficient, which is calculated based on AIS input; is the ship speed, kn.

[0166] According to the above-mentioned ship sinking model and considering the limitation of channel water depth on the ship’s spare water depth, the maximum speed control model of the ship can be determined as follows:

[0167] (25)

[0168] Where, is the maximum speed of the ship, kn.

[0169] Step 2: Establish a minimum speed control model considering wind drift.

[0170] During the voyage, the ship is affected by natural conditions and its own navigation conditions, and will deflect at a certain angle Swinging forward, the width of the water area occupied during the process of sailing is mainly composed of the transverse width occupied by the ship and total wind drift The compositions are shown in Equations (26) and (27) respectively.

[0171] (26)

[0172] (27)

[0173] Where, is the transverse width of the water area occupied by the ship, m; is the ship width, m; is the length of the ship, m; is the total amount of wind drift, m; is the wind-exposed area of ​​the hull above water, m2, which is calculated based on the wind sensor input; is the underwater area of ​​the ship, m2, which is input by the water flow sensor; is the relative wind speed, m / s, which is input by the wind sensor; is the flow rate, m / s, which is input by the water flow sensor; is the ship maneuverability index; is the wind flow pressure deflection angle, (°).

[0174] Then quote Zhang Xianku's ship maneuverability index The estimation method is

[0175] (28)

[0176] (29)

[0177] Where, is the dimensionless manipulability index; is the length between two columns, m, which is input by the information input module; is the rudder area, m2, which is input by the information input module; is the ship's draft, m.

[0178] In order to avoid threatening navigation safety by being too close to the edge of the channel, ships should maintain a certain safety distance. If the current channel is a double-lane channel, ships should also maintain a certain safety distance between each other. It should satisfy formula (30).

[0179] (30)

[0180] Where, is the channel width, m, which is input by ECDIS analysis.

[0181] Therefore, the minimum control speed model for ship navigation is determined as

[0182] (31)

[0183] Where, is the relative wind speed, m / s; is the flow velocity, m / s.

[0184] If the speed ∈

[0185] If the speed ∈

[0186] Step 3: Establish a speed control range model considering the car-following theory.

[0187] According to the car-following theory, when two ships sail one after the other on a straight line in the same channel, when the speed of the rear ship decreases to the same speed as the front ship, a certain minimum safety margin can still be maintained. Figure 7 As shown in the figure, 1 is the initial position of the leading ship; 2 is the initial position of the trailing ship; 3 is the position of the leading ship when braking is completed; 4 is the position of the trailing ship when braking begins; and 5 is the position of the trailing ship when braking is completed.

[0188] Therefore, the safe distance between ships should satisfy the condition (32).

[0189] (32)

[0190] Where, is the distance between the two ships before the leading ship brakes, m; is the braking distance of the following ship, m; is the distance traveled by the following ship during the reaction time when braking, m; is the safety margin between ships, m; is the braking distance of the leading ship, m.

[0191] When a ship brakes in an emergency, it must stop at the minimum distance. The braking distance can be determined based on the stroke, and the empirical estimation method for the ship's reverse stroke and stroke time is shown in Equations (34) and (35).

[0192] (33)

[0193] (34)

[0194] Where, is the reverse stroke, m; is the displacement of the ship, t; is the virtual mass coefficient in the ship's forward direction, which is taken as 1.07; is the acceleration due to gravity, take 9.8m / s2; is the propeller reverse pull, take Estimate, is the power of reversing the ship; is the speed of the ship when it is astern, m / s; When reversing, s.

[0195] When the ship is close to the berth, the ship speed should be controlled from the minimum speed, but the upper limit of the speed control range should not be greater than the maximum speed. When the speed of the following ship is The distance between the two ships is , the distance between the following ship and the slowed-down ship is , the distance traveled by the leading ship at a fixed speed is ,have

[0196] (35)

[0197] (36)

[0198] (37)

[0199] (38)

[0200] Can be sorted

[0201] (39)

[0202] When the ship is far away from the berth, the ship speed should be controlled from the maximum speed, but the upper limit of the speed control range should not be less than the minimum speed. When the leading ship's speed is The distance between the two ships is , the distance between the following ship and the slowed-down ship is , the distance traveled by the leading ship at a fixed speed is ,have

[0203] (40)

[0204] (41)

[0205] (42)

[0206] (43)

[0207] Can be sorted

[0208] (44)

[0209] Therefore and This is the speed control range of the ship, that is, the minimum and maximum speeds of the ship.

[0210] In some embodiments, the acquisition control module includes at least an automatic identification system, a wind sensor, and a water flow sensor.

[0211] In this embodiment, the function of the wind sensor is to identify the magnitude and direction of the wind force currently experienced by the ship, the function of the water flow sensor is to identify the magnitude and direction of the water flow currently experienced by the ship, and the function of the information receiver is to input navigation information including wave surplus depth, ship loading longitudinal surplus depth, siltation surplus depth, tide height, ship surface area, rudder area, length between two columns and main engine power.

[0212] The specific embodiments of the present invention described above do not limit the scope of protection of the present invention. Any other corresponding changes and modifications made based on the technical concept of the present invention should be included in the scope of protection of the claims of the present invention.

Claims

1. A ship night navigation supervision system, characterized in that: include: Acquisition control module, monitoring module and early warning module; The acquisition control module is used to acquire motion images and environmental images of a ship sailing at night, detect and track light spots in the motion images and environmental images, obtain trajectory information of the ship sailing at night, and send the trajectory information to the monitoring module; The monitoring module is configured to receive the trajectory information, monitor the navigation speed and navigation environment of the vessel according to the trajectory information, obtain a first monitoring result, and feed the first monitoring result back to the early warning module; The warning module includes an anti-collision warning module and a speed warning module. The anti-collision warning module performs collision risk assessment and warning on the navigation environment in the first monitoring result through a spatial and temporal urgency model; The speed warning module compares the actual speed of the ship in the first monitoring result with the speed range threshold through the ship speed control range module, and evaluates the actual speed of the ship and issues a warning; The anti-collision warning module also includes a risk assessment module and a first warning module; The risk assessment module is configured to perform a collision risk assessment on the navigation environment in the first monitoring result using a spatial and temporal urgency model to obtain a collision assessment result, and if the assessment result indicates that the ship is at risk of collision, the collision assessment result is sent to the early warning module; The first warning module includes a first light warning unit and a sound warning unit, and the first warning module is used to receive the evaluation result and respond to the evaluation result through the first light warning unit and the sound warning unit; The risk assessment module is used to perform a collision risk assessment on the navigation environment in the first monitoring result using a spatial and temporal urgency model to obtain an assessment result, including: A multi-layered elliptical ship field is established based on the length and width dimensions of the neighboring ships of the target ship; Assigning weights to the multiple layers of ship fields, and determining a weight value of the superimposed layer according to the weight value of each layer of ship fields; Determine the area of ​​the superposition region according to the superposition layer weight value; Establishing a spatial urgency model for ship collision based on the area of ​​the superimposed region; Determine the collision time of the ships according to the relative position vector sum and relative speed vector sum of the target ship and the adjacent ships; According to the relationship between the risk of ship collision and the urgency of collision time, a time urgency model of ship collision is established; Based on catastrophe theory, a collision risk calculation model between the target ship and adjacent ships is established according to the spatial urgency model and temporal urgency model of ship collision.

2. The ship night navigation monitoring system according to claim 1, characterized in that: The acquisition control module detects and tracks the light spots in the motion image and the environment image, including: Using a preset LOG operator edge detection algorithm to perform spot detection on the motion image and the environment image to obtain a ship moving image and a spot image; The light spot image is tracked using a preset Kalman filter algorithm, and the light spot image is matched with the ship driving image to obtain the trajectory information of the ship sailing at night.

3. The ship night navigation monitoring system according to claim 2, characterized in that: The step of performing spot detection on the motion image and the environment image using a preset LOG operator edge detection algorithm includes: grayscale processing is performed on the motion image and the environment image to obtain a grayscale image; Performing a convolution operation on the grayscale image using a preset convolution operation method to obtain a convolution image result; According to the convolution image result, the convolution image result is filtered using a preset spot image grayscale threshold to obtain an initial spot image; Amplifying the initial spot image using a preset maximum value filtering method to obtain an amplified spot image; Determining the ship spot and image background in the amplified light spot image according to a change in pixel values ​​between the amplified light spot image and the initial light spot image, and corroding the image background to obtain an actual light spot image; determining the center position and radius of the ship light spot according to the difference relationship between the amplified light spot image and the actual light spot image; Based on the ship light spot radius and the distance between the center positions of two adjacent light spots, overlapping light spots are determined, and the overlapping light spots are removed to obtain a target light spot image.

4. The ship night navigation monitoring system according to claim 3 is characterized in that: The method of tracking the light spot image using a preset Kalman filter algorithm and matching the light spot image with the ship driving image to obtain the trajectory information of the ship sailing at night includes: Obtain the ship's navigation speed and position, and construct the state vector of the navigation speed and position; Performing matrix transformation on the state vector to determine a filter matrix set, wherein the filter matrix set includes a state transfer matrix, a control input matrix, a system control quantity, a system control quantity matrix, and a corresponding covariance matrix; Based on the filtering parameters, predict the state vector at the next moment and determine the initial prediction covariance matrix; Determining a gain of a Kalman filter according to the predicted covariance matrix; Determining a target prediction covariance matrix according to the gain; According to the target prediction covariance matrix, the trajectory information of the ship sailing at night is determined.

5. The ship night navigation monitoring system according to claim 1, characterized in that: The speed warning module includes a speed control module and a second warning module; The speed control module is configured to compare the actual ship speed in the first monitoring result with the speed range threshold through the ship speed control range module, evaluate the actual ship speed, obtain a speed evaluation result, and send the speed evaluation result to the second early warning module if the speed evaluation result is not within the speed range threshold; The second warning module includes a second light warning unit, and the second warning module is used to receive the speed assessment result and respond to the assessment result through the second light warning unit.

6. The ship night navigation monitoring system according to claim 1, characterized in that: The ship speed control range module compares the actual ship speed in the first monitoring result with the speed range threshold to evaluate the actual ship speed, including: Based on the ship's sinking model and the limitation of the channel water depth on the ship's excess water depth, the ship's maximum speed control model is determined; According to the influence of wind drift on ship speed, a minimum speed control model of the ship is established; Based on the maximum speed control model and the minimum speed control model, with the ship spacing as the constraint condition and according to the car-following theory, the maximum speed threshold and the minimum speed threshold of the ship are determined.

7. The ship night navigation monitoring system according to claim 1, characterized in that: The acquisition control module at least includes an automatic identification system, a wind sensor, a water flow sensor, an information receiver, an electronic chart display and information system, and a radar identification unit.

8. The ship night navigation monitoring system according to claim 1, characterized in that: The monitoring module at least includes a panoramic camera, a light sensor, an infrared detector, an ultrasonic detector and a night flight instrument.

Images