Ship collision early warning method, device, equipment and medium

Abstract

The invention discloses a ship collision early warning method, device and equipment and a medium. The method is applied to a first ship and comprises the steps that a first state parameter of the first ship in a first time period and first position information of a first area where a second ship is located at the first moment are obtained, and the first position information is obtained through prediction based on second state information of the second ship in the first time period; the first time period is before the first moment; predicting second position information of a second area where the first ship is located at the first moment based on the first state parameter; determining the overlap ratio of the first area and the second area according to the first position information and the second position information; and when the overlap ratio exceeds an overlap ratio threshold value, outputting collision early warning information. Therefore, the accuracy of collision early warning can be improved.

Description

Technical Field

[0001] The present application belongs to the field of information processing technology, and in particular relates to a ship collision warning method, device, equipment and medium. Background Art

[0002] As ships are used more and more widely, ship collision incidents are gradually increasing.

[0003] In order to avoid the occurrence of ship collisions, researchers have proposed a series of ship collision warning methods, such as: speed barrier method, model predictive control and other mathematical modeling methods, path planning algorithm, fuzzy reasoning algorithm, multi-objective optimization algorithm and other intelligent optimization algorithms, as well as deep learning, supervised learning and other machine learning methods.

[0004] However, when the ship collision warning is performed by the above method, the ship will be regarded as a single particle, so the accuracy of the ship collision warning is low. Summary of the invention

[0005] The embodiments of the present application provide a ship collision warning method, device, equipment and medium, which can improve the accuracy of ship collision warning.

[0006] In a first aspect, an embodiment of the present application provides a ship collision warning method, which is applied to a first ship, and includes:

[0007] Acquire a first state parameter of a first ship in a first time period and first position information of a first area where a second ship is located at a first moment, wherein the first position information is predicted based on second state information of the second ship in the first time period, and the first time period is before the first moment;

[0008] Predicting second position information of a second area where the first ship is located at a first moment based on the first state parameter;

[0009] Determining the overlap between the first area and the second area according to the first position information and the second position information;

[0010] When the overlap exceeds the overlap threshold, a collision warning message is output.

[0011] In a second aspect, an embodiment of the present application provides a ship collision warning device, which is applied to a first ship, and includes:

[0012] An acquisition module, used for acquiring a first state parameter of a first ship in a first time period and first position information of a first area where a second ship is located at a first moment, wherein the first position information is predicted based on second state information of the second ship in the first time period, and the first time period is before the first moment;

[0013] A prediction module, configured to predict second position information of a second area where the first ship is located at a first moment based on the first state parameter;

[0014] A determination module, configured to determine the overlap between the first area and the second area according to the first location information and the second location information;

[0015] The output module is used to output collision warning information when the overlap exceeds the overlap threshold.

[0016] In a third aspect, an embodiment of the present application provides an electronic device, the device comprising: a processor and a memory storing computer program instructions;

[0017] When the processor executes the computer program instructions, the ship collision warning method as shown in any one of the embodiments of the first aspect is implemented.

[0018] In a fourth aspect, an embodiment of the present application provides a computer storage medium having computer program instructions stored thereon, and when the computer program instructions are executed by a processor, the ship collision warning method shown in any one of the embodiments of the first aspect is implemented.

[0019] In the ship collision warning method, device, equipment and medium of the embodiment of the present application, the first ship can obtain its first state parameter in the first time period and the first position information of the first area where the second ship is located at the first moment, and predict the second position information of the second area where the first ship is located at the first moment based on the first state parameter, and then determine the overlap between the first area and the second area based on the first position information and the second position information, and output the collision warning information when the overlap exceeds the overlap threshold. Since the embodiment of the present application performs collision warning based on the overlap between the second area where the first ship is located and the first area where the second ship is located, it takes into account that the ship needs to occupy a certain area, rather than treating the ship as a single particle, thereby improving the accuracy of the collision warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solution of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0021] Figure 1 This is one of the flow charts of a ship collision warning method provided by an embodiment of the present application;

[0022] Figure 2 It is a schematic diagram of a ship state change provided by an embodiment of the present application;

[0023] Figure 3It is a schematic diagram of a two-dimensional shape fitting of a hull provided by an embodiment of the present application;

[0024] Figure 4 This is one of the schematic diagrams of a ship collision scenario provided by an embodiment of the present application;

[0025] Figure 5 This is a second schematic diagram of a ship collision scenario provided by an embodiment of the present application;

[0026] Figure 6 This is a third schematic diagram of a ship collision scenario provided by an embodiment of the present application;

[0027] Figure 7 This is a schematic diagram of a scenario of ship information interaction provided by an embodiment of the present application;

[0028] Figure 8 This is the second flowchart of a ship collision warning method provided by an embodiment of the present application;

[0029] Fig. 9 It is a structural schematic diagram of a ship collision warning device provided by an embodiment of the present application;

[0030] Fig.10 It is a structural schematic diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0031] The features and exemplary embodiments of various aspects of the present application will be described in detail below. In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, rather than to limit the present application. For those skilled in the art, the present application can be implemented without the need for some of these specific details. The following description of the embodiments is only to provide a better understanding of the present application by illustrating the examples of the present application.

[0032] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "include..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0033] As the background technology, inland waterways are usually narrow and winding, with dense traffic and complex and changeable environment, so collision accidents are prone to occur. According to statistics from the State Maritime Safety Administration, collision accidents account for 48.7% of maritime accidents, and are often accompanied by total loss of ships and casualties, which is a more serious type of ship accident.

[0034] In order to reduce the loss of life and property caused by ship collisions, researchers have proposed a series of ship collision warning technologies, which can be divided into three categories based on the methods: ① Mathematical modeling method: speed barrier method, model predictive control, etc.; ② Intelligent optimization algorithm: path planning algorithm, fuzzy reasoning algorithm, multi-objective optimization algorithm, etc.; ③ Machine learning method: deep learning, supervised learning, etc.

[0035] Although there are many studies on collision warning technology, there are still some shortcomings. Most studies regard ships as single particles and do not consider that the hull needs to occupy a certain area. Therefore, the accuracy of the warning is poor, which is difficult to meet the warning requirements in the case of dense ships in inland waterways, and the actual application value is low.

[0036] The embodiment of the present application provides a ship collision warning method, device, equipment and medium, wherein a first ship can obtain its first state parameter in a first time period and the first position information of a first area where a second ship is located at a first moment, and predict the second position information of a second area where the first ship is located at a first moment based on the first state parameter, and then determine the overlap between the first area and the second area based on the first position information and the second position information, and output collision warning information when the overlap exceeds the overlap threshold. Since the embodiment of the present application performs collision warning based on the overlap between the second area where the first ship is located and the first area where the second ship is located, it takes into account that the ship needs to occupy a certain area, rather than treating the ship as a single particle, thereby improving the accuracy of the collision warning.

[0037] Figure 1A flow chart of a ship collision warning method provided by an embodiment of the present application is shown. It should be noted that the executor of the ship collision warning method may be a first ship.

[0038] like Figure 1 As shown, the ship collision warning method may include the following steps:

[0039] S110, obtaining a first state parameter of a first ship in a first time period and first position information of a first area where a second ship is located at a first moment;

[0040] S120, predicting second position information of a second area where the first ship is located at a first moment based on the first state parameter;

[0041] S130, determining the overlap between the first area and the second area according to the first position information and the second position information;

[0042] S140: When the overlap degree exceeds the overlap degree threshold, output collision warning information.

[0043] Thus, the first ship can obtain its first state parameter in the first time period and the first position information of the first area where the second ship is located at the first moment, and predict the second position information of the second area where the first ship is located at the first moment based on the first state parameter, and then determine the overlap between the first area and the second area based on the first position information and the second position information, and output collision warning information when the overlap exceeds the overlap threshold. Since the embodiment of the present application performs collision warning based on the overlap between the second area where the first ship is located and the first area where the second ship is located, it takes into account that the ship needs to occupy a certain area, rather than treating the ship as a single particle, thereby improving the accuracy of the collision warning.

[0044] In S110, the first ship and the second ship may be ships sailing in the same waters. For example, the first ship and the second ship may sail in the same inland waterway. Compared with ocean shipping, inland waterways require higher accuracy in ship trajectory prediction.

[0045] The first time period may be before the first moment. For example, the first moment may be a future moment, and the first time period may be a historical time period. That is, the position information of the area occupied by the ship at the future moment may be predicted based on the state information of the ship in the historical time period.

[0046] The first state parameter in the first time period may include state parameters corresponding to multiple moments of the first ship in the first time period. The first state parameter may include at least one of speed, acceleration, second heading angle, first drift angle, first heading drift angle, and heading intention.

[0047] The first area may be an area occupied by the hull of the second ship, and the first position information may represent the position of the first area.

[0048] The first position information may be predicted based on the second state information of the second ship in the first time period. Specifically, the first position information may be predicted and broadcasted by the second ship based on the second state information of the second ship in the first time period, based on the position information of the second area where the second ship is located at the first moment. The first ship may receive the first position information broadcasted by the second ship.

[0049] The process by which the second ship predicts the first position information based on the second state information may be the same as the process by which the first ship predicts the second position information based on the first state information, and therefore reference may be made to the subsequent introduction to predicting the second position information.

[0050] Currently, ships mainly rely on Automatic Identification System (AIS), marine radar, electronic nautical charts and other equipment to obtain information about other ships. However, the information transmission interval of such equipment is long and the delay is also high, which makes it less suitable for use in busy inland waterways.

[0051] In some embodiments, in order to shorten the interval between information transmission and reduce latency, the first location information may be acquired by a communication device utilizing vehicle to everything (V2X) wireless communication technology.

[0052] Here, the communication device using vehicle to everything (V2X) wireless communication technology may be a V2X communication device.

[0053] Specifically, both the first ship and the second ship can be connected vessels (CV) equipped with V2X communication equipment. When the distance between the first ship and the second ship is within the communication range of the V2X communication equipment, the second ship can broadcast the first position information through the V2X communication equipment, and the first ship can receive the first position information through the V2X communication equipment.

[0054] In this way, since the interval and time of sending information by the V2X communication device are shorter, the first position information obtained through the V2X communication device is more real-time.

[0055] In addition, the communication between ships and the communication between ships and the road side unit (RSU) deployed on the shore involved in the embodiments of the present application can be implemented based on V2X communication equipment.

[0056] For example, in order to accurately reflect the navigation risks within the communication range, a CV equipped with a V2X communication device can broadcast a ship safety package through the V2X communication device. The ship safety package may include: package count, ship identification, timestamp, on board unit (OBU) position, speed, heading angle, drift angle, bow drift angle, rudder angle, four-axis acceleration, hull two-dimensional profile, hull type, historical trajectory and predicted path.

[0057] Among them, packet counting can be used to detect packet loss rate and air interface communication stability; ship identification can be the unique identification of the ship, which can be standardized by the industry; timestamp can be used for trajectory prediction and time synchronization of collision model; on-board unit (OBU) position can be used for model simplification and navigation-level prediction of long-distance collision. When there are multiple OBUs, the OBU position can be the position of the main OBU, which can be an OBU manually designated from multiple OBUs; speed can be obtained through the Global Navigation Satellite System (GNSS). The heading angle can be the angle between the projection of the ship's center of mass along the bow direction on the horizontal plane and the coordinate axis in the standard coordinate system; the drift angle can be the angular difference of the ship's center of mass in a certain time period, and the drift angle can be used as an input parameter of the collision model; the bow drift angle can be the angular difference of the bow in a certain time period, and the bow drift angle can be used as an input parameter of the collision model; the rudder angle can be used as one of the input sources of trajectory prediction; the four-axis acceleration can mainly be the longitudinal acceleration and the lateral acceleration; the two-dimensional contour of the hull can be the length and width of the hull or the contour of the hull; the hull type can facilitate the establishment of a collision model; the historical trajectory can be the historical path of the ship's center of mass over a period of time; the predicted path can be generated by a trajectory prediction model. If the ship is an autonomous driving ship, the predicted path can be obtained by reading the planned path point list through a predetermined interface.

[0058] Currently, ships mainly rely on AIS, marine radar, electronic nautical charts and other equipment to obtain ship status parameters, but the positioning accuracy and accuracy of parameters such as heading angle of such equipment are not high.

[0059] In some implementations, in order to improve the positioning accuracy of the ship and the accuracy of determining state parameters such as the heading angle, the first ship may be installed with multiple OBUs.

[0060] Exemplarily, the first ship is equipped with two OBUs, and the two OBUs can be installed on the bow and stern lines of the ship and are symmetrical relative to the center of the bow and stern lines of the ship.

[0061] In this way, the first ship can determine its own position, heading angle and other status parameters through multiple OBUs with higher accuracy.

[0062] In some embodiments, the first state parameter may include a second heading angle. In order to obtain a more accurate heading angle, obtaining the first state parameter of the first ship at any time within the first time period may include:

[0063] Respectively obtain the first positions of multiple OBUs at the time;

[0064] A second heading angle of the first vessel at a time is determined according to the plurality of first positions.

[0065] Here, the heading angle may refer to the angle between the bow and stern line of the ship in the horizontal plane and the X-axis of the fixed coordinate system.

[0066] Specifically, at any moment in the first time period, the first ship can obtain the first positions of multiple OBUs at that moment through GNSS, and then calculate the second heading angle of the first ship at that moment according to the first positions corresponding to the multiple OBUs.

[0067] For example, Figure 2 As shown, state1 can be the state of the first ship at time t1, and state2 can be the state of the first ship at time t2. θ1 is the second heading angle of the first ship at time t1, and θ2 is the second heading angle of the first ship at time t2. θ1 and θ2 can be calculated by the following formula:

[0068]

[0069] Among them, (x1, y1) is the position of OBU 210 at time t1, (x2, y2) is the position of OBU 220 at time t1, (x3, y3) is the position of OBU 210 at time t2, and (x4, y4) is the position of OBU 220 at time t2.

[0070] In this way, based on the positions of multiple OBUs on the ship at a certain moment, the heading angle of the ship at that moment can be determined more accurately.

[0071] In some embodiments, the first state parameter may include a first bow drift angle. In order to obtain a more accurate bow drift angle, obtaining the first state parameter of the first ship at any second moment in the first time period may include:

[0072] respectively obtaining second positions of the plurality of OBUs at the second moment, respectively obtaining third positions of the plurality of OBUs at the third moment, and obtaining a calibrated distance between the main OBU and the bow of the first ship;

[0073] A first bow drift angle of the first ship at a second moment is determined according to the plurality of second positions, the plurality of third positions and the calibrated distance.

[0074] Here, the bow drift angle may refer to the angle between the bow velocity vector and the heading angle in the horizontal plane. The bow drift angle may characterize the degree of bow deflection. The larger the bow drift angle, the greater the deviation between the ship's direction of travel and the expected course. If the bow drift angle is 0, it means that the ship is traveling in the expected course and is in a stable state.

[0075] The third moment may be before the second moment. The master OBU may be any one of the multiple OBUs. The master OBU may be pre-specified manually, and the other OBUs except the master OBU among the multiple OBUs may be slave OBUs. In addition, the distance between the master OBU and the bow may be pre-calibrated to obtain a calibrated distance. The first ship may obtain the second positions of the multiple OBUs at the second moment, and obtain the third positions of the multiple OBUs at the third moment, respectively, through GNSS.

[0076] Specifically, a calculation formula for the bow drift angle may be constructed in advance, and then the calculation formula for the bow drift angle may be used to calculate the first bow drift angle of the first ship according to the plurality of second positions, the plurality of third positions and the calibration distance.

[0077] For example, when constructing the calculation formula of the bow drift angle, it can be assumed that the master and slave OBUs are installed on the bow and stern line of the ship and are symmetrical relative to the center of the bow and stern line of the ship. Figure 2 As shown, OBU210 can be the master OBU, and OBU 220 can be the slave OBU. State1 can be the state of the first ship at the third moment, and state2 can be the state of the first ship at the second moment. Based on this, the calculation formula of the bow drift angle can be constructed as follows:

[0078]

[0079] θ3=θ2-Δθ

[0080] Wherein, Δθ is the angle of the bow velocity vector, (x1, y1) is the position of OBU 210 at the third moment, (x3, y3) is the position of OBU 210 at the second moment, L is the calibrated distance between OBU 210 and the bow, θ2 is the heading angle, and θ3 is the bow drift angle.

[0081] θ2 can be calculated by the above formula for calculating the second heading angle. (x1+L cosθ1, y1+L sinθ1) may represent the position of the bow at the second moment, and (x3+L cosθ2, y3+L sinθ2) may represent the position of the bow at the third moment.

[0082] In this way, based on the positions of multiple OBUs on the ship at two certain moments, the bow drift angle of the ship can be determined more accurately.

[0083] At present, the drift angle of a ship can usually be obtained by comparing the heading angle obtained by the bow compass with the track obtained from the GPS or other navigation systems. However, for large ships with long hulls, there is a certain deviation between the heading angle obtained by the bow compass and the ship speed vector at the center of gravity of the ship, which will cause a deviation in the drift angle.

[0084] In some embodiments, the first state parameter may include a first drift angle. In order to obtain a more accurate drift angle, obtaining the first state parameter of the first ship at any fourth moment in the first time period may include:

[0085] Respectively obtaining fourth positions of the plurality of OBUs at a fourth moment, and respectively obtaining fifth positions of the plurality of OBUs at a fifth moment;

[0086] A first drift angle of the first vessel at a fourth moment is determined according to the plurality of fourth positions and the plurality of fifth positions.

[0087] Here, the drift angle may refer to the angle between the ship's velocity vector at the ship's center of gravity and the ship's bow and stern line. The drift angle may be a small time-varying angle that is difficult to measure.

[0088] The fifth moment may be before the fourth moment. The first ship may respectively obtain the fourth positions of the plurality of OBUs at the fourth moment and respectively obtain the fifth positions of the plurality of OBUs at the fifth moment through GNSS.

[0089] Specifically, a calculation formula for the drift angle may be constructed in advance, and then the calculation formula for the drift angle may be used to calculate the first drift angle of the first ship according to the plurality of second positions, the plurality of third positions and the calibration distance.

[0090] For example, when constructing the calculation formula of the drift angle, it can be assumed that the center of gravity of the ship coincides with the center. Figure 2 As shown, state1 can be the state of the first ship at the fifth moment, and state2 can be the state of the first ship at the fourth moment. Based on this, the calculation formula of the drift angle can be constructed as follows:

[0091]

[0092] β=θ2-θ z

[0093] Among them, θ zis the angle of the ship speed vector at the center of gravity of the ship, (x1, y1) is the position of OBU 210 at the fifth moment, (x2, y2) is the position of OBU 220 at the fifth moment, (x3, y3) is the position of OBU 210 at the fourth moment, (x4, y4) is the position of OBU220 at the fourth moment, θ2 is the heading angle, and β is the drift angle.

[0094] In this way, based on the positions of multiple OBUs on the ship at two certain moments, the drift angle of the ship can be determined more accurately.

[0095] The heading angle, bow drift angle and bow drift angle obtained above are all very important parameters in inland waterway navigation. When the ship sails in a straight line in an ideal waterway without wind and waves, θ1=θ2, θ3=β=0, and the ship is in a stable state; when the ship turns right and sails in a waterway without wind and waves, θ1>θ2, θ3>β≠0, and the ship is in a continuous changing state; when the ship sails straight in a waterway with wind and waves, θ1≠θ2, θ3≠β≠0, and the ship is in an unstable state. The dual OBU deployment method can more accurately measure the ship's heading angle, drift angle and bow drift angle.

[0096] In some embodiments, the first state parameter may include the center position of the second ship. In order to obtain a more accurate center position of the ship, obtaining the first state parameter of the first ship at any time within the first time period may include:

[0097] Respectively obtain the sixth positions of multiple OBUs at the time;

[0098] The second ship center position of the first ship at a time is determined based on the multiple sixth positions.

[0099] Here, the first ship can obtain the positions of multiple OBUs at the moment through GNSS, and then perform projection transformation on the multiple positions into two-dimensional coordinates to obtain the sixth position, and then obtain the center position of the second ship by aggregating the multiple sixth positions.

[0100] Specifically, a calculation formula for the center position of a ship may be pre-constructed, and then the calculation formula for the center position of a ship may be used to calculate the second center position of the first ship based on a plurality of sixth positions.

[0101] For example, when constructing the calculation formula of the center position of the ship, it can be assumed that multiple OBUs are located on the bow and stern line of the ship. Based on this, the calculation formula of the center position of the ship can be constructed as follows:

[0102]

[0103]

[0104] Among them, (xcenter ,y center ) is the center position of the ship, (x major ,y major ) and (x minor ,y minor ) are the two sixth positions.

[0105] In this way, based on the positions of multiple OBUs on the ship at a certain moment, the center position of the ship at that moment can be determined more accurately.

[0106] In some embodiments, the first state parameter may include a heading intention. In order to obtain a more accurate heading intention, obtaining the first state parameter of the first ship at any time within the first time period may include:

[0107] Obtaining a third heading angle, a second drift angle, and a second bow drift angle of the first ship at time instant;

[0108] The heading intention of the first ship at the time is determined by using a Gaussian Mixed Model (GMM) based on the third heading angle, the second drift angle and the second bow drift angle.

[0109] Here, the heading intention may be one of a left turn, a right turn, and a straight line.

[0110] The third heading angle can be obtained in the same manner as the second heading angle, the second drift angle can be obtained in the same manner as the first drift angle, and the second bow drift angle can be obtained in the same manner as the first bow drift angle, which will not be described in detail here.

[0111] Specifically, a GMM may be pre-constructed, and then the GMM may be used to determine the heading intention of the first ship based on the third heading angle, the second drift angle, and the second bow drift angle.

[0112] Exemplarily, the construction process of GMM may be as follows:

[0113] Cache Time a The heading angle θ2, drift angle β, and heading drift angle θ3 information of the time window length are calculated respectively for their first-order Gaussian distribution probability densities:

[0114]

[0115] Where θ is the heading angle θ2, drift angle β or heading drift angle θ3, μ and σ 2 are the mean and variance of the Gaussian distribution respectively.

[0116] Calculate the joint probability density distribution of the multidimensional variable x = (θ2, β, θ3) composed of the heading angle θ2, the drift angle β and the heading drift angle θ3:

[0117]

[0118] X=(x1,x2,...,x n )

[0119] Where d is the variable dimension. This model belongs to a three-dimensional Gaussian distribution, so d can be 3;

[0120] Represents the mean of each dimensional variable; ∑ is the covariance matrix, which represents the correlation between each dimensional variable. For a three-dimensional Gaussian distribution, we have:

[0121]

[0122] The above joint probability density function contains only one Gaussian model. The following three Gaussian distributions (left turn, straight, right turn, a total of 3 categories) form a GMM, and its probability density function is as follows:

[0123]

[0124] in, is the probability density function of the k-th Gaussian model, which can be regarded as the probability that the model will produce x after the k-th model is selected; p(x) = π k is the weight of the k-th Gaussian model, which is regarded as the prior probability of selecting the k-th model and satisfies

[0125] In this way, GMM is constructed. According to GMM, Time can be calculated a The maximum heading probability p of the ship is taken as the heading intention.

[0126] In this way, GMM can be used to more accurately determine the ship's heading intention at a certain moment based on the ship's heading angle, drift angle and bow drift angle at that moment.

[0127] In addition, in some implementations, the rudder steering angle may be directly acquired through the CAN bus of the first ship, and the heading intention of the first rudder may be determined according to the rudder steering angle.

[0128] Regarding S120, the second area may be an area occupied by the hull of the first ship, and the second position information may represent the position of the second area.

[0129] In some implementations, in order to more accurately determine the location information of the area where the ship is located, S120 may include:

[0130] Predicting a first ship center position and a first heading angle of the first ship at a first moment based on the first state parameter;

[0131] Based on the calibrated hull data of the first ship, the center position of the first ship and the first heading angle, the two-dimensional shape of the first ship is fitted to obtain second position information.

[0132] Here, the calibration hull data may be the length and width of the hull or the hull contour line. The length and width of the hull may specifically be the maximum length and maximum width of the hull.

[0133] Exemplarily, a two-dimensional coordinate system with the bow vertex as a reference point can be determined based on the hull structure diagram, and the two-dimensional coordinates of the center position of the first ship in the two-dimensional coordinate system can be determined by calibration.

[0134] If the hull data is the length and width of the hull, the hull rectangular envelope can be fitted based on the maximum length and width of the hull, such as Figure 3 As shown in the dotted rectangular box 310 in FIG. 1 , the second position information can be the coordinates of r1, r2, r3 and r4. If the two-dimensional coordinates of the center position r of the first ship in the two-dimensional coordinate system are (x, y), and the first heading angle is θ, the rectangular space solution of the ship can be obtained:

[0135]

[0136] Among them, (r 1，x , r 1，y ) is the coordinate of r1, (r 2，x , r 2，y ) is the coordinate of r2, (r 3，x , r 3，y ) is the coordinate of r3, (r 4，x , r 4，y ) is the coordinate of r4, S1 can be the distance between r and r1 or r2, S2 can be the distance between r and r3 or r4, α1 is the angle between the bow and stern line of the ship and the line connecting rr1, and α2 is the angle between the bow and stern line of the ship and the line connecting rr3. Among them, S1, S2, α1 and α2 can all be determined based on the length and width of the hull.

[0137] If the hull data is the hull contour line, the 2D shape of the hull can be fitted based on the hull contour line. The fitting method is as follows: take the bow vertex as (0, 0), discretize the hull contour line into several points, calculate the position of each point relative to the bow vertex according to the hull structure diagram, and use the least squares method to fit the curve. The formula is as follows:

[0138]

[0139] Among them, i is the serial number of the discrete point, n is the total number of discrete points, and h θ (x i ) is a function of the horizontal coordinate of point i, which can be a linear function or a nonlinear function, y i is the ordinate of point i.

[0140] By solving the minimum value of the above formula, we can get the fitting curve function of the hull.

[0141] For hulls with irregular contours, the least squares method can be used in sections to fit the hull curve using multiple functions.

[0142] In this way, the two-dimensional shape of the hull is taken into consideration, and the accurate position of the area where the first ship is located is obtained by accurately fitting the two-dimensional shape of the first ship, which is more in line with the needs of collision warning for inland waterway shipping in complex environments.

[0143] In some embodiments, in order to more accurately predict the ship center position and heading angle of the ship, S120 may include:

[0144] The trajectory prediction model is used to predict the ship center position and the heading angle of the first ship at the first moment based on the first state parameter to obtain the first ship center position and the first heading angle of the first ship at the first moment.

[0145] Here, the trajectory prediction model can be obtained by improving the transformer model based on the Long Short Term Memory (LSTM) network.

[0146] Specifically, the trajectory prediction model may include a position encoding model, a position encoding model, and a prediction model.

[0147] Based on this, in some embodiments, in order to more accurately predict the ship center position and the heading angle of the ship, the predicting the first ship center position and the first heading angle of the first ship at the first moment based on the first state parameter may include:

[0148] Performing position encoding on the first state parameter using a position encoding model to obtain a first feature;

[0149] Use the self-attention model to perform self-attention operation on the first feature to obtain the second feature;

[0150] The prediction model is used to predict the center position and heading angle of the first ship at the first moment based on the second feature to obtain the center position and the first heading angle of the first ship.

[0151] For example, the state frame corresponding to the first state parameter of the first ship at time t in the first time period can be defined as: in, is the position coordinate of the first ship at time t after coordinate transformation and smoothing, v t 、a t , β t , pt The first ship's speed, acceleration, heading angle, drift angle, heading drift angle and heading intention at time t can be defined as X=(X -H , X -H+1 , ..., X t , ..., X 0 ), where t≤0, X 0 is the reference frame, i.e., the last frame of the historical trajectory sequence.

[0152] Then, the trajectory coordinate point of the first ship at the future time t can also be defined as Y = (Y 1 , Y 2 , ..., Y t , ..., Y T ), where t>0.

[0153] Specifically, the position encoding model may include an embedding layer and a position encoding layer. The embedding layer may be a single linear layer, which performs dimension-upgrading processing on the historical trajectory sequence X, and then adds position encoding to each frame of data in the historical trajectory sequence X. The position encoding value interval may be controlled between [0, 1], and a variable frequency sine-cosine function with no optimization space may be used for position encoding, as shown below:

[0154]

[0155] Where loc is the relative position of the current data in the sequence, d is the dimension of the data after dimensionality increase, and i represents the i-th dimension of the data.

[0156] The self-attention model can be an encoder, which can include a multi-head attention module, a residual connection, a normalization module, and a fully connected layer. Each self-attention mechanism contained in the multi-head attention module is independently calculated during the calculation process, and the parameters are not shared. When the calculations of multiple self-attention mechanisms are completed, the outputs of the multiple self-attention mechanisms are linearly combined and output to the residual, normalization module, and fully connected layer. The calculation formula is as follows:

[0157] MultiHead(Q,K,V)=Concat(head1,...,head n )W 0

[0158] head i =Attention(Q i , K i , V i )

[0159] Among them, W 0is the linear combination weight matrix, and the other parameters are typical parameters of the transformer model and will not be described here.

[0160] The prediction model can be a decoder, which can be composed of an LSTM sequence model and can be used to generate a prediction sequence. Add random Gaussian noise z to get the hidden feature Then the state frame at time t-1 is converted into a spatial vector through a fully connected layer LSTM will transform the space vector Calculate and get the current state vector Finally, it is converted into a coordinate sequence through a multilayer perceptron (MLP), which is the final predicted sequence. The calculation process is as follows:

[0161]

[0162]

[0163]

[0164]

[0165] Among them, W1 and W4 are different weight parameters of MLP, W2 is the weight parameter of the fully connected layer, and W3 is the weight parameter of LSTM.

[0166] It includes the first ship center position and the first heading angle of the first ship at time t in the future.

[0167] In this way, the midpoint position and heading angle of the first ship at a certain moment are predicted by the above-mentioned machine learning model based on the first state parameter. Since the first state parameter includes very rich information and the accuracy of the above-mentioned machine learning model is also high, a more accurate midpoint position and heading angle of the ship can be obtained.

[0168] In some implementations, before S120 , the initial trajectory prediction model may be trained to obtain the above trajectory prediction model.

[0169] Specifically, the decoder can input the historical trajectory sequence of the ship and the real future trajectory marked for the historical trajectory sequence in parallel to learn the connection between the data, and then the decoder outputs the predicted future trajectory of the ship frame by frame. During the model inference process, the input of the decoder is the predicted future trajectory output by the decoder at the previous moment, thereby realizing trajectory prediction and outputting prediction information.

[0170] In some implementations, before the initial trajectory prediction model is trained, the training data may be preprocessed, and the training samples may include the historical trajectory sequence of the ship and the real future trajectory annotated for the historical trajectory sequence.

[0171] Here, preprocessing can include data cleaning, valid data extraction, and trajectory optimization.

[0172] Among them, data cleaning can include V2X data decoding and outlier deletion; effective data extraction can include filtering stationary trajectories and multi-OBU position aggregation; trajectory optimization can include local coordinate transformation, trajectory inflection point smoothing and normalization processing.

[0173] Specifically, for normalization processing, the ship's position, speed, relative distance and other parameters can be normalized to the expected range to reduce the inflection point data noise. The formula is as follows:

[0174]

[0175] Among them, x is the input value of the normalized network after smoothing, y is the output value of the normalized network, η is the normalized weight after training, γ is the normalized offset value after training, mean() is the mean of multiple input values ​​in the historical time period, f() is the variance of multiple input values ​​in the historical time period, and Δk is the normalization process constant.

[0176] In addition, the specific method of aggregating multiple OBU positions can refer to the method for determining the center position of the second ship mentioned above, which will not be repeated here.

[0177] In S130, the first ship may determine the degree of overlap between its second area and the second area of ​​the second ship according to its second location information and the first location information of the second ship.

[0178] For example, Figure 4 As shown, if the current time is t3 and the first time is t4, the shaded area 400 is the overlapped area of ​​the second area where the first ship 410 is located and the first area where the second ship 420 is located at t4. Based on this, the calculation formula of the overlap degree C can be:

[0179]

[0180] Among them, S a，b is the area of ​​the overlapped region 400, S a is the area of ​​the second region where the first ship 410 is located, S b is the area of ​​the first region where the second ship 420 is located.

[0181] In S140, the collision warning information may include a collision position and a collision time, wherein the collision position may be a center point position of an overlapping area of ​​the first area and the second area.

[0182] Here, when the overlap exceeds the overlap threshold, collision warning information can be output; when the overlap does not exceed the overlap threshold, it indicates that the predicted trajectories of the first ship and the second ship have no intersection and are in a safe heading state, and collision warning information may not be output.

[0183] Exemplarily, the overlap threshold may be 0. If C=0, no collision warning information is output; if C>0, collision warning information is output.

[0184] Specifically, the first ship may broadcast collision warning information to other ships (including the second ship) and shore-based RSUs.

[0185] In some implementations, in order to perform collision warning more accurately, S140 may include:

[0186] When the overlap degree is greater than the overlap degree threshold, the duration between the first moment and the current moment is calculated to obtain a first duration;

[0187] When the first duration is greater than the first preset duration, outputting a first-level collision warning message;

[0188] When the first duration is not greater than the first preset duration and greater than the second preset duration, outputting a second-level collision warning message;

[0189] When the first duration is not greater than the second preset duration, a third-level collision warning message is output.

[0190] Here, the first preset duration may be greater than the second preset duration. The first preset duration and the second preset duration may be set according to actual needs.

[0191] The danger levels represented by the first-level collision warning information, the second-level collision warning information, and the third-level collision warning information may increase in sequence.

[0192] Specifically, if the overlap is greater than the overlap threshold, the first time duration between the collision moment (the first performance moment) and the current moment can be further determined. The smaller the first time duration is, the higher the collision warning level is.

[0193] Exemplarily, the current time is t5, the collision time is t6, the first preset time length is λ2, and the second preset time length is λ1.

[0194] If 0<C and λ2<t6-t5, it can be said that the ship is in a slightly dangerous state, and the first-level collision warning information can be output and a first-level warning can be issued. Among them, λ2<t6-t1 means that the time of collision is far away, and there is enough time to adjust the course and speed.

[0195] If 0<C and λ1<t6-t5≤λ2, it can be said that the ship is in a general dangerous state, and the secondary collision warning information can be output and a secondary warning can be issued. Among them, λ1<t6-t5≤λ2 means that the distance to the collision moment is moderate, and the ship needs to adjust the speed and course in time.

[0196] If 0<C and t6-t5≤λ1, it can be said that the ship is in a very dangerous state, and the third-level collision warning information can be output and a third-level warning can be issued. Among them, t6-t5≤λ1 means that the collision moment is very close, and the ship needs to decelerate urgently, adjust the course, and send a distress signal.

[0197] For example, in Figure 5 In the scenario shown, a secondary collision warning warn2 can be performed. Figure 6 In the scenario shown, a first-level collision warning warn1 can be performed.

[0198] In this way, when it is determined that a collision is about to occur, different levels of collision warning information are output based on the urgency of the collision, and different levels of warnings are issued, which can facilitate the immediate handling of more urgent collision warnings.

[0199] In some implementations, S140 may include:

[0200] When the corresponding overlaps of the first ship and the second ship at the multiple first moments respectively exceed the overlap threshold, determining the maximum overlap among the multiple overlaps; and outputting collision warning information corresponding to the maximum overlap;

[0201] or,

[0202] When the overlaps corresponding to the first ship and the plurality of second ships respectively exceed the overlap threshold, the collision moments of the plurality of second ships with the first ship are respectively determined; the earliest collision moment among the plurality of collision moments is determined; and the collision warning information corresponding to the earliest collision moment is output.

[0203] Specifically, since the collision has a process, the collision process may include multiple moments, so the number of the first moments may be multiple. In order to avoid repeated warnings, S140 may include:

[0204] When the corresponding overlaps of the first ship and the second ship at the plurality of first moments respectively exceed an overlap threshold, determining a maximum overlap among the plurality of overlaps;

[0205] Output the collision warning information corresponding to the maximum overlap.

[0206] Here, the overlaps corresponding to multiple collision moments (ie, the first moment) can be determined. The overlaps at different collision moments can be different. The collision warning information corresponding to the maximum overlap can be output, that is, the collision warning information corresponding to the most serious collision.

[0207] For example, the overlap degrees corresponding to the multiple collision moments may be C1, C2, ..., C n , the maximum overlap can be C max =max(C1, C2, ..., C n ), the collision warning information corresponding to the maximum overlap can be T(x c ,y c , t c ).

[0208] In this way, by outputting the collision warning information corresponding to the maximum overlap, it is possible to avoid outputting the collision warning information at multiple moments in the predicted collision process, thereby avoiding wasting communication resources.

[0209] In addition, since there may be multiple ships near the first ship, the number of the second ships may be multiple. In order to improve the safety of the ships, S140 may include:

[0210] When the coincidence degrees respectively corresponding to the first ship and the plurality of second ships exceed a coincidence degree threshold, respectively determining the collision moments of the plurality of second ships with the first ship;

[0211] determining an earliest collision time among a plurality of collision times;

[0212] Output the collision warning information corresponding to the earliest collision time.

[0213] Here, when it is predicted that the first ship is about to collide with multiple second ships, the collision moments of the first ship and the multiple second ships can be determined respectively, and then the earliest collision moment among the multiple collision moments can be determined, and collision warning information corresponding to the earliest collision moment can be output.

[0214] In this way, by outputting the collision warning information corresponding to the earliest collision moment, relevant personnel can be reminded to pay attention to the most urgent collision events, thereby improving the safety of the ship.

[0215] In some embodiments, the continuous broadcast message mechanism of all CVs can include the location information of multiple OBUs and the area where the ship is located, so as to build situational awareness of inland waterway ships, such as Figure 7As shown. Ships can share information with each other, and the shore-based RSU can obtain the navigation status of ships in real time to achieve dynamic management of waterways. RSU broadcasts basic safety information such as waterway obstacles, dangerous areas, hydrology, and meteorology to ships to improve ship traffic safety. Obstacles, dangerous areas, etc. can be marked according to map information. The marking can be as follows Figure 7 The dashed box 710 shown in FIG. The ship-shore collaboration method is conducive to the supervision and coordinated dispatch of ships.

[0216] In order to better describe the entire solution, based on the above embodiments, a specific example is given. Figure 8 As shown, the ship collision warning method may include S801-S807, which is explained in detail below.

[0217] S801, obtaining first state parameters of a first ship and first position information of a second ship.

[0218] S802: Perform trajectory prediction based on the first state parameter to obtain second position information of the first ship.

[0219] S803: Calculate the overlap degree C based on the first position information and the second position information.

[0220] S804, determine whether C is greater than 0.

[0221] If yes, execute S805; if no, do not output the collision warning information and return to execute S801.

[0222] S805, calculating a first time duration t6-t5 between the collision time t6 and the current time t5.

[0223] S806, determining whether the first time length t6-t5 is greater than a first preset time length λ2.

[0224] If yes, output the first-level collision warning information; if no, execute S807.

[0225] S807 , determining whether the first time length t6 - t5 is not greater than the first preset time length λ2 and greater than the second preset time length λ1 .

[0226] If yes, then a second-level collision warning message is output; if no, then it indicates that the first time duration t6-t5 is not greater than the second preset time duration λ1, and a third-level collision warning message is output.

[0227] Thus, the first ship can obtain its first state parameter in the first time period and the first position information of the first area where the second ship is located at the first moment, and predict the second position information of the second area where the first ship is located at the first moment based on the first state parameter, and then determine the overlap between the first area and the second area based on the first position information and the second position information, and output collision warning information when the overlap exceeds the overlap threshold. Since the embodiment of the present application performs collision warning based on the overlap between the second area where the first ship is located and the first area where the second ship is located, it takes into account that the ship needs to occupy a certain area, rather than treating the ship as a single particle, thereby improving the accuracy of the collision warning.

[0228] The ship collision warning method provided in the embodiment of the present application utilizes V2X communication equipment to exchange ship status information, considers the two-dimensional shape of the hull, adopts a trajectory prediction method, constructs a multi-level warning model, realizes ship network collision warning, and improves the safety of inland waterway shipping.

[0229] Based on the same inventive concept, the embodiment of the present application also provides a ship collision warning device, which can be applied to a first ship. Fig. 9 The ship collision warning device provided in the embodiment of the present application is described in detail.

[0230] Fig. 9 A schematic structural diagram of a ship collision warning device provided by an embodiment of the present application is shown.

[0231] like Fig. 9 As shown, the ship collision warning device may include:

[0232] An acquisition module 901 is used to acquire a first state parameter of a first ship in a first time period and first position information of a first area where a second ship is located at a first moment, wherein the first position information is predicted based on second state information of the second ship in the first time period, and the first time period is before the first moment;

[0233] A prediction module 902, configured to predict second position information of a second area where the first ship is located at a first moment based on the first state parameter;

[0234] A determination module 903, configured to determine the overlap between the first area and the second area according to the first location information and the second location information;

[0235] The output module 904 is used to output collision warning information when the overlap exceeds the overlap threshold.

[0236] Thus, the first ship can obtain its first state parameter in the first time period and the first position information of the first area where the second ship is located at the first moment, and predict the second position information of the second area where the first ship is located at the first moment based on the first state parameter, and then determine the overlap between the first area and the second area based on the first position information and the second position information, and output collision warning information when the overlap exceeds the overlap threshold. Since the embodiment of the present application performs collision warning based on the overlap between the second area where the first ship is located and the first area where the second ship is located, it takes into account that the ship needs to occupy a certain area, rather than treating the ship as a single particle, thereby improving the accuracy of the collision warning.

[0237] In some implementations, in order to more accurately determine the location information of the area where the ship is located, the prediction module 902 may include:

[0238] A prediction submodule, configured to predict a first ship center position and a first heading angle of the first ship at a first moment based on a first state parameter;

[0239] The fitting submodule is used to fit the two-dimensional shape of the first ship based on the calibrated hull data of the first ship, the center position of the first ship and the first heading angle to obtain the second position information, and the calibrated hull data is the length and width of the hull or the hull contour line.

[0240] In some embodiments, in order to more accurately predict the ship center position and heading angle of the ship, the prediction submodule may include:

[0241] An encoding unit, configured to perform position encoding on the first state parameter using a position encoding model to obtain a first feature;

[0242] A computing unit, configured to perform a self-attention operation on the first feature using a self-attention model to obtain a second feature;

[0243] The prediction unit is used to predict the center position and heading angle of the first ship at the first moment based on the second feature by using the prediction model to obtain the center position and the first heading angle of the first ship.

[0244] In some implementations, the first ship is equipped with multiple OBUs, and the first state parameter includes a second heading angle. In order to obtain a more accurate heading angle, the acquisition module 901 may include:

[0245] A first acquisition submodule is used to respectively acquire the first positions of multiple OBUs at a time;

[0246] The first determination submodule is used to determine a second heading angle of the first ship at a time according to the multiple first positions.

[0247] In some implementations, the first state parameter includes a first bow drift angle. In order to obtain a more accurate bow drift angle, the acquisition module 901 may include:

[0248] A second acquisition submodule is used to respectively acquire the second positions of the plurality of OBUs at the second moment, respectively acquire the third positions of the plurality of OBUs at the third moment, and acquire the calibrated distance between the main OBU and the bow of the first ship, wherein the main OBU is any one of the plurality of OBUs at the third moment before the second moment;

[0249] The second determining submodule is used to determine a first bow drift angle of the first ship at a second moment according to the multiple second positions, the multiple third positions and the calibrated distance.

[0250] In some implementations, the first state parameter includes a first drift angle. To obtain a more accurate drift angle, the acquisition module 901 may include:

[0251] A third acquisition submodule is used to respectively acquire fourth positions of the plurality of OBUs at a fourth moment, and to respectively acquire fifth positions of the plurality of OBUs at a fifth moment, the fifth moment being before the fourth moment;

[0252] The third determining submodule is used to determine a first drift angle of the first ship at a fourth moment according to the multiple fourth positions and the multiple fifth positions.

[0253] In some implementations, the first state parameter includes the center position of the second ship. In order to obtain a more accurate center position of the ship, the acquisition module 901 may include:

[0254] The fourth acquisition submodule is used to respectively acquire the sixth positions of the plurality of OBUs at the time;

[0255] The fourth determination submodule is used to determine the center position of the second ship of the first ship at a time according to the multiple sixth positions.

[0256] In some implementations, the first state parameter includes a heading intention. In order to obtain a more accurate heading intention, the acquisition module 901 may include:

[0257] A fifth acquisition submodule is used to acquire a third heading angle, a second drift angle, and a second bow drift angle of the first ship at a time;

[0258] The fifth determination submodule is used to determine the heading intention of the first ship at the moment based on the third heading angle, the second drift angle and the second bow drift angle by using a mixed Gaussian model, where the heading intention is one of turning left, turning right and going straight.

[0259] In some embodiments, in order to avoid repeated warnings or further improve the safety of the ship, the output module 904 may include:

[0260] A sixth determination submodule is used to determine a maximum degree of overlap among a plurality of degrees of overlap when the degrees of overlap corresponding to the first ship and the second ship at a plurality of first moments respectively exceed a threshold value of the degree of overlap; a first output submodule is used to output collision warning information corresponding to the maximum degree of overlap;

[0261] or,

[0262] The seventh determination submodule is used to determine the collision moments of the multiple second ships with the first ship when the overlap degrees corresponding to the first ship and the multiple second ships respectively exceed the overlap degree threshold; the eighth determination submodule is used to determine the earliest collision moment among the multiple collision moments; the second output submodule is used to output the collision warning information corresponding to the earliest collision moment.

[0263] Fig.10 A schematic structural diagram of an electronic device provided by an embodiment of the present application is shown.

[0264] like Fig.10 As shown, the electronic device 10 is a structural diagram of an exemplary hardware architecture of an electronic device that can implement the ship collision warning method and ship collision warning device in the embodiment of the present application. The electronic device can refer to the electronic device in the embodiment of the present application.

[0265] The electronic device 10 may include a processor 1001 and a memory 1002 storing computer program instructions.

[0266] Specifically, the processor 1001 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0267] The memory 1002 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 1002 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In appropriate cases, the memory 1002 may include a removable or non-removable (or fixed) medium. In appropriate cases, the memory 1002 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 1002 is a non-volatile solid-state memory. In a specific embodiment, the memory 1002 may include a read-only memory (ROM), a random access memory (RAM), a disk storage medium device, an optical storage medium device, a flash memory device, an electrical, optical, or other physical / tangible memory storage device. Thus, typically, memory 1002 includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to a method according to an aspect of the present application.

[0268] The processor 1001 implements any one of the ship collision warning methods in the above embodiments by reading and executing computer program instructions stored in the memory 1002 .

[0269] In one example, the electronic device may further include a communication interface 1003 and a bus 1004. Fig.10 As shown, the processor 1001, the memory 1002, and the communication interface 1003 are connected via a bus 1004 and communicate with each other.

[0270] The communication interface 1003 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0271] Bus 1004 includes hardware, software or both, and the parts of electronic equipment are coupled to each other.For example, but not limitation, bus may include accelerated graphics port (AGP) or other graphics bus, enhanced industrial standard architecture (EISA) bus, front side bus (FSB), hypertransport (HT) interconnection, industrial standard architecture (ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations. In appropriate cases, bus 1004 may include one or more buses. Although the present application embodiment describes and shows a specific bus, the application considers any suitable bus or interconnection.

[0272] The electronic device can execute the ship collision warning method in the embodiment of the present application, thereby realizing the combination Figures 1 to 9 The invention describes a ship collision warning method and device.

[0273] In addition, in combination with the ship collision warning method in the above embodiment, the embodiment of the present application can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when the computer program instructions are executed by the processor, any one of the ship collision warning methods in the above embodiment is implemented.

[0274] It should be clear that the present application is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present application.

[0275] The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0276] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiment, or in a different order from the embodiment, or several steps can be performed simultaneously.

[0277] The above reference is according to the method of the embodiment of the present application, the flow chart of the device (system) and the computer program product and / or the block diagram described various aspects of the present application.It should be understood that each square box in the flow chart and / or the block diagram and the combination of each square box in the flow chart and / or the block diagram can be realized by computer program instructions.These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer or other programmable data processing device to produce a machine so that these instructions executed by the processor of the computer or other programmable data processing device enable the realization of the function / action specified in one or more square boxes of the flow chart and / or the block diagram.Such a processor can be but is not limited to a general-purpose processor, a special-purpose processor, a special application processor or a field programmable logic circuit.It can also be understood that each square box in the block diagram and / or the flow chart and the combination of the square boxes in the block diagram and / or the flow chart can also be realized by the dedicated hardware that performs the specified function or action, or can be realized by the combination of dedicated hardware and computer instructions.

[0278] The above is only a specific implementation of the present application. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the protection scope of the present application is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in this application, and these modifications or replacements should be included in the protection scope of this application.

Claims

1. A ship collision warning method, characterized in that: The method is applied to a first vessel, and comprises: Acquire a first state parameter of the first ship in a first time period and first position information of a first area where the second ship is located at a first moment, wherein the first position information is predicted based on the second state information of the second ship in the first time period, and the first time period is before the first moment; Predicting second position information of a second area where the first ship is located at the first moment based on the first state parameter; Determining, according to the first location information and the second location information, a degree of overlap between the first area and the second area; When the overlap exceeds the overlap threshold, collision warning information is output.

2. The method according to claim 1, characterized in that The predicting, based on the first state parameter, second position information of the second area where the first ship is located at the first moment includes: Predicting a first ship center position and a first heading angle of the first ship at the first moment based on the first state parameter; Based on the calibrated hull data of the first ship, the center position of the first ship and the first heading angle, the two-dimensional shape of the first ship is fitted to obtain the second position information, wherein the calibrated hull data is the length and width of the hull or the hull contour line.

3. The method according to claim 2, characterized in that The predicting, based on the first state parameter, a first ship center position and a first heading angle of the first ship at the first moment includes: Performing position encoding on the first state parameter using a position encoding model to obtain a first feature; Performing a self-attention operation on the first feature using a self-attention model to obtain a second feature; The prediction model is used to predict the center position and heading angle of the first ship at the first moment based on the second feature to obtain the center position of the first ship and the first heading angle.

4. The method according to any one of claims 1 to 3, characterized in that The first ship is equipped with a plurality of on-board units OBU, the first state parameter includes a second heading angle, and obtaining the first state parameter of the first ship at any time within the first time period includes: Respectively obtain the first positions of the plurality of OBUs at the time; A second heading angle of the first ship at the moment is determined according to the plurality of first positions.

5. The method according to claim 4, characterized in that The first state parameter includes a first bow drift angle, and obtaining the first state parameter of the first ship at any second moment in the first time period includes: respectively obtaining the second positions of the plurality of OBUs at the second moment, respectively obtaining the third positions of the plurality of OBUs at the third moment, and obtaining the calibrated distance between the main OBU and the bow of the first ship, wherein the third moment is before the second moment, and the main OBU is any one of the plurality of OBUs; A first bow drift angle of the first ship at the second moment is determined according to the plurality of second positions, the plurality of third positions and the calibrated distance.

6. The method according to claim 4, characterized in that The first state parameter includes a first drift angle, and obtaining the first state parameter of the first ship at any fourth moment in the first time period includes: Respectively obtaining fourth positions of the plurality of OBUs at the fourth moment, and respectively obtaining fifth positions of the plurality of OBUs at a fifth moment, the fifth moment being before the fourth moment; A first drift angle of the first ship at the fourth moment is determined according to the plurality of fourth positions and the plurality of fifth positions.

7. The method according to claim 4, characterized in that The first state parameter includes the center position of the second ship, and obtaining the first state parameter of the first ship at any time within the first time period includes: respectively obtaining the sixth positions of the plurality of OBUs at the time; The second ship center position of the first ship at the moment is determined based on the multiple sixth positions.

8. The method according to any one of claims 1 to 3, characterized in that The first state parameter includes a heading intention, and obtaining the first state parameter of the first ship at any time within the first time period includes: Acquire a third heading angle, a second drift angle, and a second bow drift angle of the first ship at the moment; A mixed Gaussian model is used to determine the heading intention of the first ship at the moment based on the third heading angle, the second drift angle and the second bow drift angle, where the heading intention is one of turning left, turning right and going straight.

9. The method according to claim 1, characterized in that: When the overlap exceeds the overlap threshold, outputting collision warning information includes: When the overlaps between the first ship and the second ship respectively corresponding to the first moments exceed the overlap threshold, determining a maximum overlap among the plurality of overlaps; and outputting collision warning information corresponding to the maximum overlap; or, When the overlap degrees respectively corresponding to the first ship and the plurality of second ships exceed an overlap degree threshold, respectively determine the collision moments of the plurality of second ships with the first ship; determine the earliest collision moment among the plurality of collision moments; and output collision warning information corresponding to the earliest collision moment.

10. A ship collision warning device, characterized in that: The device is applied to a first ship, and comprises: an acquisition module, configured to acquire a first state parameter of the first ship in a first time period and first position information of a first area where the second ship is located at a first moment, wherein the first position information is predicted based on the second state information of the second ship in the first time period, and the first time period is before the first moment; a prediction module, configured to predict second position information of a second area where the first ship is located at the first moment based on the first state parameter; a determination module, configured to determine a degree of overlap between the first area and the second area according to the first location information and the second location information; The output module is used to output collision warning information when the overlap exceeds the overlap threshold.

11. An electronic device, characterized in that: The device comprises: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, the ship collision warning method as described in any one of claims 1-9 is implemented.

12. A computer storage medium, characterized in that: The computer storage medium stores computer program instructions, and when the computer program instructions are executed by the processor, the ship collision warning method according to any one of claims 1 to 9 is implemented.

Images