Visibility-based ship collision risk assessment method

Abstract

The invention provides a visibility-based ship collision risk assessment method, and relates to the technical field of ship navigation safety, and the method comprises the steps: constructing a current quaternary field of a ship through a navigation data set of the ship; the visibility of the environment image is obtained through calculation, the current quaternary field is zoomed through the visibility of the environment image, a new quaternary field is obtained, and the minimum safe meeting distance and the minimum safe passing distance are obtained through the new quaternary field; according to the navigation data set of the ship, the navigation data set of the target ship, the minimum safety encounter distance and the minimum safety passing distance, a danger membership degree set is obtained through calculation; and calculating a collision risk degree through the risk membership degree set, and adjusting the route of the ship through the collision risk degree to avoid the target ship. According to the method, a more accurate and real-time collision risk assessment result can be provided in a complex and changeable inland water area environment, and the safety performance and decision-making ability of the unmanned ship in a complex navigation environment are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of ship navigation safety, and in particular to a method for assessing ship collision risk based on visibility. Background Art

[0002] The safe navigation of ships is highly dependent on accurate collision risk assessment algorithms to ensure effective avoidance of other ships or obstacles. However, existing assessment methods are mainly based on the minimum safe clearance distance (DCPA) and the minimum approach time (TCPA), and fail to fully consider the impact of environmental conditions (such as lighting changes) on risk assessment, resulting in insufficient accuracy and adaptability of assessment results. In inland waterway shipping, traditional collision risk assessment methods mostly rely on driver experience, and have problems such as strong subjectivity and unstable assessment results. Although there are some assessment models based on geometric principles, ship domain models and fuzzy logic in the existing technology, the accuracy and reliability of assessment results still need to be further improved in the complex and changeable inland water environment. Summary of the Invention

[0003] In order to solve the above problems, the present invention provides a method for assessing the risk of ship collision based on visibility, comprising the steps of:

[0004] S1: Obtain the navigation data set of the own ship, and construct the current four-element domain of the own ship through the navigation data set of the own ship;

[0005] S2: Acquire the environment image of the target ship, calculate the visibility of the environment image, scale the current four-element domain according to the visibility of the environment image, obtain a new four-element domain, and obtain the minimum safe encounter distance and the minimum safe passing distance according to the new four-element domain;

[0006] S3: Obtain the target ship's navigation data set, and calculate the hazard membership set based on the own ship's navigation data set, the target ship's navigation data set, the minimum safe encounter distance, and the minimum safe passing distance;

[0007] S4: Obtain the collision risk through the risk membership set calculation, and adjust the ship's route according to the collision risk to avoid the target ship.

[0008] Optionally, step S1 specifically includes:

[0009] S11: Establish a coordinate system with the center of the own ship as the origin, the direction of the own ship's bow as the positive direction of the y-axis, and the right direction perpendicular to the own ship's bow as the positive direction of the x-axis;

[0010] S12: Calculate the longitudinal front radius R through the navigation data set of the own ship fore , longitudinal rear radius R aft , lateral left radius Rport and the lateral right radius R starb ;

[0011] S13: In the first quadrant of the coordinate system, through the longitudinal front radius R fore and the lateral right radius R starb Construct the first 1 / 4 ellipse; in the second quadrant of the coordinate system, through the longitudinal front radius R fore and the lateral left radius R port Construct the second 1 / 4 ellipse; in the third quadrant of the coordinate system, through the longitudinal rear radius R aft and the lateral left radius R port Construct the third 1 / 4 ellipse; in the fourth quadrant of the coordinate system, through the longitudinal rear radius R aft and the lateral right radius R starb Construct the fourth 1 / 4 ellipse;

[0012] S14: The first 1 / 4 ellipse, the second 1 / 4 ellipse, the third 1 / 4 ellipse and the fourth 1 / 4 ellipse are spliced ​​together to obtain the current quaternion domain of the own ship.

[0013] Optionally, step S2 specifically includes:

[0014] S21: Obtain the R, G, and B three-channel histogram of the target ship's environment image, and calculate the average brightness μ of the environment image through the R, G, and B three-channel histogram;

[0015] S22: Get the average brightness μ on a sunny day 晴天 and visibility on clear days V s , through the average brightness μ of the environment image and the average brightness μ of the sunny day 晴天 and visibility on clear days V s , calculate the visibility V of the environment image;

[0016] S23: Set the current quaternary field according to V s / V ratio is scaled to obtain a new quaternion field;

[0017] S24: The distance from the center of the ship to the boundary of the new four-dimensional area is used as the minimum safe encounter distance, and twice the minimum safe encounter distance is used as the minimum safe passing distance.

[0018] Optional:

[0019] The expression of the visibility V of the environment image is:

[0020]

[0021] Visibility on a clear day V s The expression is:

[0022]

[0023] Among them, β is the atmospheric extinction coefficient and ε is the contrast threshold.

[0024] Optional:

[0025] The risk membership set includes: DCPA risk membership, TCPA risk membership, relative distance risk membership, relative bearing risk membership and ship speed ratio risk membership.

[0026] Optionally, step S3 specifically includes:

[0027] S31: Calculate the DCPA value, TCPA value, relative distance, relative bearing, and ship speed ratio using the navigation data set of the own ship and the navigation data set of the target ship;

[0028] S32: Calculate the DCPA hazard membership by using the DCPA value, minimum safe encounter distance, and minimum safe passing distance;

[0029] S33: Calculate the TCPA hazard membership by using the DCPA value, TCPA value, minimum safe encounter distance, and minimum safe passing distance;

[0030] S34: Calculate the relative distance hazard membership based on the navigation data set of the own ship, the navigation data set of the target ship, the relative distance, and the minimum safe passing distance;

[0031] S35: Calculate the relative bearing hazard membership based on the own ship's navigation data set, the target ship's navigation data set, and the relative bearing;

[0032] S36: Calculate and obtain the ship speed ratio hazard membership through the navigation data set of the own ship, the navigation data set of the target ship, and the ship speed ratio.

[0033] Optional:

[0034] The expression of collision risk index CRI is:

[0035]

[0036] in, is the DCPA hazard membership, is the TCPA hazard membership, k R is the relative distance hazard membership, k α is the relative position hazard membership, k K is the ship speed ratio hazard membership, w DCPA 、w TCPA 、w R 、w α and w Kare the weight values ​​of DCPA risk membership, TCPA risk membership, relative distance risk membership, relative bearing risk membership and ship speed ratio risk membership respectively.

[0037] The present invention also provides a device for assessing the risk of ship collision based on visibility, for implementing the method for assessing the risk of ship collision based on visibility, the device comprising:

[0038] The four-element domain construction module is used to obtain the navigation data set of the ship and construct the current four-element domain of the ship through the navigation data set of the ship;

[0039] The four-element field adjustment module is used to obtain the environmental image of the target ship, calculate the visibility of the environmental image, scale the current four-element field according to the visibility of the environmental image, obtain a new four-element field, and obtain the minimum safe encounter distance and the minimum safe passing distance through the new four-element field;

[0040] The danger membership calculation module is used to obtain the navigation data set of the target ship, and calculate the danger membership set based on the navigation data set of the own ship, the navigation data set of the target ship, the minimum safe encounter distance and the minimum safe passing distance;

[0041] The avoidance module is used to calculate the collision risk by using the risk membership set, and adjust the ship's route according to the collision risk to avoid the target ship.

[0042] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the visibility-based ship collision risk assessment method when executing the program.

[0043] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for assessing the risk of ship collision based on visibility is implemented.

[0044] The present invention has the following beneficial effects:

[0045] 1. Obtain visibility through environmental image calculation of the target ship, use visibility as a dynamic correction factor, and dynamically scale the quaternary domain to apply to collision risk assessment of ships in various weather conditions;

[0046] 2. The collision risk is calculated through the DCPA risk membership, TCPA risk membership, relative distance risk membership, relative bearing risk membership and ship speed ratio risk membership, and the ship route is adjusted according to the collision risk. In the process of ship avoidance, dynamic parameters such as DCPA value, TCPA value, relative distance, relative bearing and ship speed ratio are fully considered, which can provide more accurate and real-time collision risk assessment results in complex and changeable inland water environment, significantly improving the safety performance and decision-making ability of unmanned ships in complex navigation environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a flow chart of a method according to an embodiment of the present invention;

[0048] Figure 2 This is a schematic diagram of the current four-element field;

[0049] Figure 3 Typical inland river surface images and corresponding histograms under different visibility conditions;

[0050] Figure 4 Schematic diagram of relative motion and safe passing distance between ships;

[0051] Figure 5 A schematic diagram of the definition of critical time threshold and safety time threshold;

[0052] Figure 6 Schematic diagram of the four-dimensional ship collision field under different ambient brightness conditions of cross-encounter;

[0053] Figure 7 Schematic diagram of the four-dimensional ship collision field under different ambient brightness conditions in an encounter situation;

[0054] Figure 8 Schematic diagram of the four-dimensional ship collision field under different ambient brightness conditions in an overtaking situation;

[0055] Figure 9 The present invention is a structural diagram of a device for assessing the risk of ship collision based on visibility;

[0056] Figure 10 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application;

[0057] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0058] The following will be combined with the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0059] Reference Figure 1 The present invention provides a method for assessing the risk of ship collision based on visibility, comprising the steps of:

[0060] S1: Obtain the navigation data set of the own ship, and construct the current four-element domain of the own ship through the navigation data set of the own ship;

[0061] In some embodiments:

[0062] Step S1 specifically includes:

[0063] S11: Establish a coordinate system with the center of the own ship as the origin, the direction of the own ship's bow as the positive direction of the y-axis, and the right direction perpendicular to the own ship's bow as the positive direction of the x-axis;

[0064] S12: Calculate the longitudinal front radius R through the navigation data set of the own ship fore , longitudinal rear radius R aft , lateral left radius R port and the lateral right radius R starb ;

[0065] In some embodiments, the calculation formula is:

[0066]

[0067] Where: L is the length of the ship; v is the speed of the ship; k AD is the initial diameter coefficient of gyration; k DT is the ship's advance coefficient.

[0068] S13: In the first quadrant of the coordinate system, through the longitudinal front radius R fore and the lateral right radius R starb Construct the first 1 / 4 ellipse; in the second quadrant of the coordinate system, through the longitudinal front radius R fore and the lateral left radius R port Construct the second 1 / 4 ellipse; in the third quadrant of the coordinate system, through the longitudinal rear radius R aft and the lateral left radius R port Construct the third 1 / 4 ellipse; in the fourth quadrant of the coordinate system, through the longitudinal rear radius R aft and the lateral right radius R starb Construct the fourth 1 / 4 ellipse;

[0069] S14: The first 1 / 4 ellipse, the second 1 / 4 ellipse, the third 1 / 4 ellipse and the fourth 1 / 4 ellipse are spliced ​​together to obtain the current quaternion domain of the own ship.

[0070] In some embodiments, the current quaternion field is as follows Figure 2 As shown, it consists of four ellipses. The length of the semi-axis is dynamically adjusted with the speed and the encounter situation to achieve adaptive optimization. The boundary condition equation of the current quaternion field is:

[0071]

[0072] S2: Acquire the environment image of the target ship, calculate the visibility of the environment image, scale the current four-element domain according to the visibility of the environment image, obtain a new four-element domain, and obtain the minimum safe encounter distance and the minimum safe passing distance according to the new four-element domain;

[0073] In some embodiments:

[0074] Step S2 specifically includes:

[0075] S21: Obtain the R, G, and B three-channel histogram of the target ship's environment image, and calculate the average brightness μ of the environment image through the R, G, and B three-channel histogram;

[0076] In some embodiments, four typical inland river surface images under different visibility conditions are selected for analysis, such as Figure 3 As shown, they are Figure 3 (a1) is an image of an inland river with heavy fog. Figure 3 (b1) is the image of the inland river at night. Figure 3 (c1) is the image of an inland river on a clear day. Figure 3 (d1) is an image of an inland river on a cloudy day. Image analysis was performed on these four typical inland river images. First, the R, G, and B channel histograms of the four images were plotted, and then the average brightness of the images was calculated using the histograms.

[0077] The left side is the original image, the right side is the Figure 3 (a2)- Figure 3 (d2) is the corresponding histogram. In order to more intuitively obtain the characteristics of the inland river surface image from the data level, the average brightness of the image is calculated by the average brightness calculation formula of the image, which can well measure the brightness of the image.

[0078]

[0079] Where μ is the average brightness of the image, M and N are the number of rows and columns of the image, respectively. I(i, j) is the brightness value of pixel (i, j), which is usually in the range [0, 255].

[0080] S22: Get the average brightness μ on a sunny day 晴天 and visibility on clear days V s , through the average brightness μ of the environment image and the average brightness μ of the sunny day 晴天 and visibility on clear days V s , calculate the visibility V of the environment image;

[0081] In some embodiments, the average brightness of the four types of images is calculated as shown in Table 1.

[0082] Table 1 Average brightness of four types of images

[0083]

[0084] Normalize the average brightness of the image by the average brightness of the inland river surface image on a sunny day:

[0085]

[0086] In some embodiments, the effect of ambient brightness on visibility is considered in conjunction with Koschmieder's law, which describes the relationship between visibility V and the atmospheric extinction coefficient β.

[0087] The expression of the visibility V of the environment image is:

[0088]

[0089] Visibility on a clear day V s The expression is:

[0090]

[0091] Among them, β is the atmospheric extinction coefficient and ε is the contrast threshold.

[0092] In some embodiments, ε is typically taken as 0.02;

[0093] The above Koschmieder law only considers the effect of atmospheric extinction coefficient on visibility, but does not consider the effect of ambient light intensity. In order to consider the effect of real-time environment on the quaternary ship field, the real-time ambient brightness is introduced to correct the visibility, and finally the normalized brightness μ is introduced. (0,1) Corrected the effect of ambient brightness on visibility.

[0094] For three ambient brightness conditions: sunny day, cloudy day and night, the visibility ratio relationship under different ambient brightness is established:

[0095]

[0096] Where Vs is the visibility on a clear day, V c is the visibility on a cloudy day, V n is the visibility at night, μ (0,1)s is the normalized brightness of a sunny day (equal to 1), μ (0,1)c is the normalized brightness of cloudy sky, μ (0,1)n is the normalized brightness at night. Assuming good weather quality, that is, the atmospheric extinction coefficient in different environments is the same, the visibility ratio of different ambient brightness to the brightness of a sunny day can be further derived:

[0097]

[0098] S23: Set the current quaternary field according to V s / V ratio is scaled to obtain a new quaternion field;

[0099] In some embodiments, the quaternary ship domain is constructed based on good visibility conditions (i.e., visibility on a clear day) without considering the ambient brightness. The present invention combines the ratio of visibility on a cloudy day to that on a clear day, and the ratio of visibility at night to that on a clear day, to scale the quaternary ship dynamic domain; thereby dynamically adjusting the safe collision avoidance range of the quaternary ship under different ambient brightness conditions, ensuring that the minimum safe encounter distance between ships can still be reasonably set in low visibility environments (such as cloudy days or at night), thereby improving navigation safety and the adaptability of collision avoidance strategies.

[0100] S24: The distance from the center of the ship to the boundary of the new four-dimensional area is used as the minimum safe encounter distance, and twice the minimum safe encounter distance is used as the minimum safe passing distance.

[0101] In some embodiments, the distance r from the origin of the ellipse to the edge is calculated according to the formula:

[0102]

[0103] Where: a is the major semi-axis of the ellipse; b is the minor semi-axis of the ellipse; α is the angle with the major semi-axis a.

[0104] The calculation formula for the distance from the ship to the boundary of the area in each quadrant is derived:

[0105]

[0106] Where: α is the angle between the line connecting own ship and the nearest encounter point and the major semi-axis of the ellipse. x and y are the coordinate differences between own ship and the nearest encounter point.

[0107] Relative motion between ships and safe passing distance Figure 4As shown in the figure, the ship domain established based on the four-element ship domain model above denies intrusion by other ships. That is, the collision risk within the domain is set to 1, and the domain boundary is usually set to the minimum safe encounter distance r1. The safe passing distance r2 is twice r1. When the DCPA is greater than r2, it indicates that there is no collision risk to the ship.

[0108] S3: Obtain the target ship's navigation data set, and calculate the hazard membership set based on the own ship's navigation data set, the target ship's navigation data set, the minimum safe encounter distance, and the minimum safe passing distance;

[0109] In some embodiments, a risk membership set is constructed based on fuzzy logic theory, which is often used to handle unclear boundaries between knowledge and experience. Because ship collision risk is highly ambiguous and uncertain, it is applied to collision risk assessment and collision avoidance decision-making. Common fuzzy evaluation models use DCPA, TCPA, ship speed ratio, target ship distance, and bearing as evaluation parameters.

[0110] The risk membership set includes: DCPA risk membership, TCPA risk membership, relative distance risk membership, relative bearing risk membership and ship speed ratio risk membership.

[0111] In some embodiments:

[0112] Step S3 specifically includes:

[0113] S31: Calculate the DCPA value, TCPA value, relative distance, relative bearing, and ship speed ratio using the navigation data set of the own ship and the navigation data set of the target ship;

[0114] In some embodiments, the basic information of the ship when encountering the situation is obtained through sensors, such as the longitude and latitude coordinates, the ship speed v O 、Course φ O . The latitude and longitude coordinates of the target ship, ship speed v T 、Course φ T With the unmanned vessel as the origin, a coordinate system is established, where the x-axis is positive eastward along the equator and the y-axis is positive northward perpendicular to the equator.

[0115] Based on the above-established coordinate system and the ship information collected by the sensor, the target ship’s position information (x T ,y T ) and own ship's position information (x O ,y O ), and then solve the relative speed v between the ship and the target ship R and its relative heading In addition, it is necessary to calculate the distance R between the target ship and the own ship, the true bearing α of the target ship relative to the own shipTO , the true bearing of own ship relative to target ship α OT 、DCPA(denoted as d CPA ), TCPA (denoted as t CPA ) and the speed ratio K between the target ship and own ship. The calculation formula is as follows:

[0116]

[0117] Where:

[0118]

[0119]

[0120] S32: Calculate the DCPA hazard membership by using the DCPA value, minimum safe encounter distance, and minimum safe passing distance;

[0121] In some embodiments, the DCPA hazard membership refers to a measure of the spatial collision risk between the target ship and the own ship at the closest approach point in collision avoidance decisions based on fuzzy logic. This indicator maps the DCPA value to a membership between 0 and 1, reflecting its contribution to the collision risk. A higher membership indicates a greater potential collision risk. In the spatial dimension, DCPA is an intuitive and effective reference indicator. After determining the minimum safe passing distance through the safety field, the DCPA hazard membership can be calculated using the following formula:

[0122]

[0123] Among them, r1 is the minimum safe encounter distance after considering visibility optimization, and r2 is the absolute safe encounter distance, which is usually 2-3 times of r1. In this paper, r2 is taken as twice r1, and the two constitute the safety domain.

[0124] S33: Calculate the TCPA hazard membership by using the DCPA value, TCPA value, minimum safe encounter distance, and minimum safe passing distance;

[0125] In some embodiments, the TCPA risk membership is used as a fuzzy quantitative indicator to assess the potential collision risk between the target vessel and the own vessel at the moment of closest approach in the future. This membership reflects the impact of approach time on the degree of collision risk by mapping the TCPA value to a fuzzy membership value between 0 and 1. A higher membership indicates a more imminent collision possibility and greater danger. The formula for the TCPA risk membership is as follows:

[0126]

[0127] in

[0128]

[0129] Among them, T1 represents the critical time threshold, T2 represents the safety time threshold, and the definitions of the critical time threshold and the safety time threshold are as follows: Figure 5 shown.

[0130] S34: Calculate the relative distance hazard membership based on the navigation data set of the own ship, the navigation data set of the target ship, the relative distance, and the minimum safe passing distance;

[0131] In some embodiments, the relative distance danger membership is an important indicator reflecting the potential spatial collision risk between the target ship and the own ship. The calculation formula is as follows:

[0132]

[0133] The relative distance (R) measures the spatial proximity between the target vessel and the own vessel during navigation, reflecting the spatial distribution of potential collision hazards. A higher relative distance hazard membership function value indicates that the target vessel is closer to the own vessel and the navigational state is more dangerous. d1 represents the minimum safe distance, indicating the minimum collision avoidance distance required for the yielding vessel to take emergency collision avoidance measures. d2 represents the safe distance, indicating the relatively safe distance between the two vessels. d1 and d2 are calculated as follows:

[0134] d1=K1K2K3D min

[0135] d2=K1K2K3r2

[0136] Among them, r2 is the safe passing distance of the ship, K1, K2, and K3 are human factors. The research object of this paper is an unmanned ship, which is assumed to have a high degree of intelligence. Therefore, K1 = K2 = K3 = 1 by default; D min Indicates the distance between the two ships at the latest steering point, usually 12 times the length of the ship.

[0137] S35: Calculate the relative bearing hazard membership based on the own ship's navigation data set, the target ship's navigation data set, and the relative bearing;

[0138] In some embodiments, the relative azimuth hazard membership primarily measures the degree of danger posed to the owning vessel by vessels approaching from different azimuths. Generally speaking, a vessel approaching from the starboard side poses a greater threat than one approaching from the port side, and a vessel approaching directly ahead poses a greater threat than one approaching directly behind. Under relative angle analysis, under the same conditions, the vessel is most dangerous when it is 19° from the owning vessel, with a hazard membership of 1; while the vessel is safest when it is 199° from the owning vessel, with a hazard membership of 0. Based on this, the relative azimuth hazard membership can be obtained as follows:

[0139]

[0140] S36: Calculate and obtain the ship speed ratio hazard membership through the navigation data set of the own ship, the navigation data set of the target ship, and the ship speed ratio.

[0141] In some embodiments, the ship speed is higher than the risk membership:

[0142]

[0143] Where: C is the collision angle between the two ships, that is, the true bearing of the target ship relative to the own ship. When the ship speed ratio K>0, M=2.

[0144] S4: Obtain the collision risk through the risk membership set calculation, and adjust the ship's route according to the collision risk to avoid the target ship.

[0145] In some embodiments, the membership of DCPA and relative distance R reflects the complexity of ship collision risk in the spatial dimension; TCPA membership reflects the change of ship collision risk in the temporal dimension; and relative bearing α OT and the ship speed ratio K affect the difficulty of collision avoidance between ships. Based on this, the present invention proposes a collision risk calculation model based on the quaternary dynamic ship field;

[0146] The expression of collision risk index CRI is:

[0147]

[0148] in, is the DCPA hazard membership, is the TCPA hazard membership, k R is the relative distance hazard membership, k α is the relative position hazard membership, k K is the ship speed ratio hazard membership, w DCPA 、w TCPA 、w R 、w α and w K are the weight values ​​of DCPA risk membership, TCPA risk membership, relative distance risk membership, relative bearing risk membership and ship speed ratio risk membership respectively.

[0149] In some embodiments, w DCPA ,w TCPA ,w R ,w α ,w K The weight values ​​are 0.36, 0.32, 0.14, 0.10, and 0.08 respectively.

[0150] Based on the quantitative calculation model for ship collision risk constructed above, the model, based on the four-dimensional dynamic ship domain and taking into account real-time visibility, was validated under three different visibility conditions to assess its effectiveness and rationality. Table 2 lists the initial position, heading, and speed of the host ship and target ship in a crossing encounter scenario.

[0151] The weight matrix shows that the DCPA membership function has the highest weight, indicating its most significant impact on collision risk assessment. Therefore, higher accuracy is required when calculating the DCPA membership function. The collision risk assessment model based on the four-dimensional ship domain considers the impact of changes in ship length and real-time speed on the minimum safe approach distance, which increases the adaptability of the ship domain model to a certain extent. However, it ignores the impact of real-time visibility. The minimum safe approach distance does not change with changes in visibility caused by the environment, resulting in large errors in scenarios with large variations in visibility.

[0152] Table 2 Initial ship parameters in the cross-encounter situation

[0153]

[0154]

[0155] Table 3 Comparative analysis of ship collision risk parameters under different ambient brightness conditions in cross-encounter situations

[0156]

[0157] Schematic diagram of the four-dimensional ship collision field under different environmental brightness conditions of cross-encounter Figure 6 As shown, in Figure 6 In the figure, the solid blue line represents the quaternary ship collision domain without considering the influence of real-time brightness, while the dashed red line represents the improved quaternary ship collision domain after considering real-time brightness. It can be seen that under clear skies, the blue implementation coincides with the dashed red line. However, under cloudy and nighttime conditions, the quaternary ship domain expands compared to the clear skies, the minimum safe distance for collisions between ships decreases, and the collision risk increases. This shows that, with other conditions remaining unchanged, the real-time brightness of the environment significantly affects the minimum safe passing distance (ship domain), which in turn leads to significant changes in the collision risk. Combining Tables 2 and 3, it can be seen that under the three different ambient brightness conditions of clear skies, cloudy skies, and nighttime, the DCPA hazard membership increases with decreasing ambient brightness, and therefore the final collision risk also increases. The improved collision risk calculation model based on ambient brightness is consistent with inland waterway collision avoidance practices.

[0158] Table 4 Initial parameters of ships in encounter situations

[0159]

[0160] Table 5 Comparative analysis of ship collision risk parameters under different ambient brightness conditions in encounter situations

[0161]

[0162] Table 6 Initial ship parameters in overtaking situation

[0163]

[0164] Table 7 Comparative analysis of ship collision risk parameters under different ambient brightness conditions in overtaking situations

[0165]

[0166] Schematic diagram of the four-dimensional ship collision field under different ambient brightness conditions in the encounter situation Figure 7 As shown in the figure, the schematic diagram of the four-dimensional ship collision field under different ambient brightness conditions in the overtaking situation is as follows: Figure 8 As shown, in order to further verify the effectiveness of the proposed method, for the two-ship encounter and overtaking situations set in Table 4 and Table 6, the collision risk model diagram is also drawn with the help of software (as shown in Figure 7 and Figure 8 By comparing experimental results under three different ambient brightness conditions, we analyze the impact of ambient brightness on the risk of ship collision. As shown in the figure, as the real-time ambient brightness decreases, the ship's domain boundary expands, and the ship's collision risk increases accordingly. The experimental results show that, from the perspective of the ship's domain, all other conditions being equal, the lower the real-time ambient brightness, the higher the risk of ship collision. This conclusion is consistent with actual navigation practice and regulatory requirements.

[0167] like Figure 9 As shown, the present invention further provides a visibility-based ship collision risk assessment device 90 for implementing the visibility-based ship collision risk assessment method. The device comprises:

[0168] The quaternary domain construction module 901 is used to obtain the navigation data set of the own ship and construct the current quaternary domain of the own ship based on the navigation data set of the own ship;

[0169] The quaternary domain adjustment module 902 is configured to obtain an environmental image of the target ship, calculate the visibility of the environmental image, scale the current quaternary domain based on the visibility of the environmental image to obtain a new quaternary domain, and obtain the minimum safe encounter distance and the minimum safe passing distance based on the new quaternary domain;

[0170] The danger membership calculation module 903 is used to obtain the navigation data set of the target ship, and calculate the danger membership set based on the navigation data set of the own ship, the navigation data set of the target ship, the minimum safe encounter distance, and the minimum safe passing distance;

[0171] The avoidance module 904 is configured to obtain a collision risk by calculating the risk membership set, and adjust the own ship's route according to the collision risk to avoid the target ship.

[0172] In some embodiments, see Figure 10 , Figure 10 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device 100 provided in an embodiment of the present application includes a memory 101 and a processor 102. The memory 101 stores a computer program, wherein the computer program, when executed by the processor, implements the visibility-based ship collision risk assessment method.

[0173] Specifically, the processor 102 may include, for example, a general-purpose microprocessor, an instruction set processor and / or a related chipset and / or a dedicated microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 102 may also include onboard memory for caching purposes. The processor 102 may be a single processing unit or multiple processing units for executing different actions of the method flow according to the embodiments of the present application.

[0174] Memory 101 can be, for example, any medium capable of containing, storing, conveying, disseminating, or transmitting instructions. For example, memory 101 can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, components, or propagation media. Specific examples of memory 101 include: magnetic storage devices, such as magnetic tape or hard disk drives (HDDs); optical storage devices, such as compact discs (CD-ROMs); random access memory (RAM) or flash memory; and / or wired or wireless communication links.

[0175] This application also provides a computer-readable medium storing a computer program that, when executed by a processor, implements the visibility-based ship collision risk assessment method. This computer-readable medium may be included in the device / apparatus / system described in the above embodiments, or it may exist independently and not incorporated into the device / apparatus / system. The computer-readable medium carries one or more programs that, when executed, implement the method described in the embodiments of this application.

[0176] According to an embodiment of the present application, a computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium or any combination thereof. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or component, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device, or component. In the present application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical cable, radio frequency signals, or any suitable combination thereof.

[0177] Those skilled in the art will understand that the features described in the various embodiments and / or claims of the present application may be combined and / or combined in a variety of ways, even if such combinations or combinations are not explicitly described in the present application. In particular, without departing from the spirit and teachings of the present application, the features described in the various embodiments and / or claims of the present application may be combined and / or combined in a variety of ways. All of these combinations and / or combinations fall within the scope of the present application. Therefore, the scope of the present application should not be limited to the above-mentioned embodiments, but should be determined not only by the attached claims, but also by the equivalents of the attached claims. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A method for assessing ship collision risk based on visibility, characterized in that: Including steps: S1: Obtain the navigation data set of the own ship, and construct the current four-element domain of the own ship through the navigation data set of the own ship; S2: Acquire the environment image of the target ship, calculate the visibility of the environment image, scale the current four-element domain according to the visibility of the environment image, obtain a new four-element domain, and obtain the minimum safe encounter distance and the minimum safe passing distance according to the new four-element domain; S3: Obtain the target ship's navigation data set, and calculate the hazard membership set based on the own ship's navigation data set, the target ship's navigation data set, the minimum safe encounter distance, and the minimum safe passing distance; S4: Obtain the collision risk through the risk membership set calculation, and adjust the ship's route according to the collision risk to avoid the target ship.

2. The method for assessing ship collision risk based on visibility according to claim 1, characterized in that: Step S1 specifically includes: S11: Establish a coordinate system with the center of the own ship as the origin, the direction of the own ship's bow as the positive direction of the y-axis, and the right direction perpendicular to the own ship's bow as the positive direction of the x-axis; S12: Calculate the longitudinal front radius R through the navigation data set of the own ship fore , longitudinal rear radius R aft , lateral left radius R port and the lateral right radius R starb ; S13: In the first quadrant of the coordinate system, through the longitudinal front radius R fore and the lateral right radius R starb Construct the first 1 / 4 ellipse; in the second quadrant of the coordinate system, through the longitudinal front radius R fore and the lateral left radius R port Construct the second 1 / 4 ellipse; in the third quadrant of the coordinate system, through the longitudinal rear radius R aft and the lateral left radius R port Construct the third 1 / 4 ellipse; in the fourth quadrant of the coordinate system, through the longitudinal rear radius R aft and the lateral right radius R starb Construct the fourth 1 / 4 ellipse; S14: The first 1 / 4 ellipse, the second 1 / 4 ellipse, the third 1 / 4 ellipse and the fourth 1 / 4 ellipse are spliced ​​together to obtain the current quaternion domain of the own ship.

3. The method for assessing ship collision risk based on visibility according to claim 1, characterized in that: Step S2 specifically includes: S21: Obtain the R, G, and B three-channel histogram of the target ship's environment image, and calculate the average brightness μ of the environment image through the R, G, and B three-channel histogram; S22: Get the average brightness μ on a sunny day 晴天 and visibility on clear days V s , through the average brightness μ of the environment image and the average brightness μ of the sunny day 晴天 and visibility on clear days V s , calculate the visibility V of the environment image; S23: Set the current quaternary field according to V s / V ratio is scaled to obtain a new quaternion field; S24: The distance from the center of the ship to the boundary of the new four-dimensional area is used as the minimum safe encounter distance, and twice the minimum safe encounter distance is used as the minimum safe passing distance.

4. The method for assessing ship collision risk based on visibility according to claim 3, characterized in that: The expression of the visibility V of the environment image is: Visibility on a clear day V s The expression is: Among them, β is the atmospheric extinction coefficient and ε is the contrast threshold.

5. The method for assessing ship collision risk based on visibility according to claim 1, characterized in that: The risk membership set includes: DCPA risk membership, TCPA risk membership, relative distance risk membership, relative bearing risk membership and ship speed ratio risk membership.

6. The method for assessing ship collision risk based on visibility according to claim 5, characterized in that: Step S3 specifically includes: S31: Calculate the DCPA value, TCPA value, relative distance, relative bearing, and ship speed ratio using the navigation data set of the own ship and the navigation data set of the target ship; S32: Calculate the DCPA hazard membership by using the DCPA value, minimum safe encounter distance, and minimum safe passing distance; S33: Calculate the TCPA hazard membership based on the DCPA value, TCPA value, minimum safe encounter distance, and minimum safe passing distance; S34: Calculate the relative distance hazard membership based on the navigation data set of the own ship, the navigation data set of the target ship, the relative distance, and the minimum safe passing distance; S35: Calculate the relative bearing hazard membership based on the own ship's navigation data set, the target ship's navigation data set, and the relative bearing; S36: Calculate and obtain the ship speed ratio hazard membership through the navigation data set of the own ship, the navigation data set of the target ship, and the ship speed ratio.

7. The method for assessing ship collision risk based on visibility according to claim 5, characterized in that: The expression of collision risk index CRI is: in, is the DCPA hazard membership, is the TCPA hazard membership, k R is the relative distance hazard membership, k α is the relative position hazard membership, k K is the ship speed ratio hazard membership, w DCPA 、w TCPA 、w R 、w α and w K are the weight values ​​of DCPA risk membership, TCPA risk membership, relative distance risk membership, relative bearing risk membership and ship speed ratio risk membership respectively.

8. A device for assessing the risk of ship collision based on visibility, for implementing the method for assessing the risk of ship collision based on visibility according to any one of claims 1 to 7, characterized in that: The device comprises: The four-element domain construction module is used to obtain the navigation data set of the ship and construct the current four-element domain of the ship through the navigation data set of the ship; The four-element field adjustment module is used to obtain the environmental image of the target ship, calculate the visibility of the environmental image, scale the current four-element field according to the visibility of the environmental image, obtain a new four-element field, and obtain the minimum safe encounter distance and the minimum safe passing distance through the new four-element field; The danger membership calculation module is used to obtain the navigation data set of the target ship, and calculate the danger membership set based on the navigation data set of the own ship, the navigation data set of the target ship, the minimum safe encounter distance and the minimum safe passing distance; The avoidance module is used to calculate the collision risk by using the risk membership set, and adjust the ship's route according to the collision risk to avoid the target ship.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the visibility-based ship collision risk assessment method according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for assessing the risk of ship collision based on visibility as claimed in any one of claims 1 to 7 is implemented.

Images