A method and device for early warning of navigation risk of an inland waterway vessel

Abstract

The application relates to an inland river ship navigation risk early warning method and device, which comprises the following steps: quantifying ship navigation risk factors, constructing a three-dimensional dynamic ship field model; acquiring radar feature information, AIS feature information, river water depth information and navigation environment influence information of the ship; performing track prediction based on the AIS feature information and a preset BP neural network model, and performing track fitting with the radar feature data to obtain a predicted track; generating a current ship field and a predicted ship field according to the three-dimensional dynamic ship field model, the predicted track and the channel environment influence information; judging whether the current ship field and the predicted ship field have the risk of being invaded; and generating risk early warning information and giving a risk early warning prompt to the user when the risk of being invaded exists. The three-dimensional dynamic ship field model and the predicted track based on multi-factor quantization under the condition of time and space change are combined, and the accuracy, rapidity and universality of the inland river ship navigation risk early warning are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ship navigation safety, and in particular to an inland ship navigation risk early warning method and device. BACKGROUND

[0002] The inland navigation conditions are complex, and there is a certain safety risk when the ship navigates. Once a water traffic accident occurs, it will cause serious personnel casualties and economic losses. Therefore, the problem of inland ship navigation safety has been widely valued.

[0003] The ship field refers to defining a subjective or objective area around the ship that is not infringed, which is widely used in micro water traffic safety fields such as ship navigation safety early warning. Due to the complexity of inland navigation conditions, the inland ship field is quite different from the sea ship field, but most of the current research and application related to the inland ship field still refer to the related content of the sea ship field, and lack of consideration of the characteristics of the inland navigation environment and the universality of the whole river section, which leads to poor accuracy of the inland ship navigation risk early warning. SUMMARY

[0004] Therefore, it is necessary to provide an inland ship navigation risk early warning method and device to solve the technical problem of poor risk early warning accuracy and universality of the whole navigation section when the ship navigates in the inland river.

[0005] To solve the above problems, the present application provides an inland ship navigation risk early warning method, which comprises:

[0006] Quantify the ship navigation risk factors based on the signal function and the pull-type transformation, and construct a three-dimensional dynamic ship field model;

[0007] Obtain the radar feature information, AIS feature information, river depth information and navigation environment influence information of the ship;

[0008] According to the AIS feature information, the ship track is predicted based on the preset BP neural network model, and the track fitting is performed with the radar feature data to obtain the predicted track;

[0009] According to the three-dimensional dynamic ship field model, the predicted track and the channel environment influence information, the current ship field and the predicted ship field of the ship are generated;

[0010] According to the radar feature information, the AIS feature information and the river depth information, it is judged whether the current ship field and the predicted ship field of the ship exist the risk of being invaded;

[0011] Generate risk warning information when the current ship field or the predicted ship field of the ship is at risk of being invaded, and give a risk warning prompt to the user.

[0012] Optionally, the ship navigation risk factors are quantified based on the signal function and the pull-type transformation, and a three-dimensional dynamic ship field model is constructed, including:

[0013] According to the pulse signal function and the step signal function, the ship navigation risk factors are quantified by using the pull-type transformation to obtain a plurality of environmental impact quantity mathematical expressions;

[0014] The plurality of environmental impact quantity mathematical expressions are matrix synthesized to obtain an environmental impact element overall quantization model;

[0015] Based on the preset initial basic ship field model and the environmental impact element overall quantization model, the three-dimensional dynamic ship field model with scale changing with space and time is constructed.

[0016] Optionally, according to the pulse signal function and the step signal function, the ship navigation risk factors are quantified by using the pull-type transformation to obtain a plurality of environmental impact quantity mathematical expressions, including:

[0017] Based on the pulse signal function, the wind-induced drift is quantified by using the pull-type transformation to obtain a wind-induced drift mathematical expression;

[0018] Based on the pulse signal function, the flow-induced drift is quantified by using the pull-type transformation to obtain a flow-induced drift mathematical expression;

[0019] Based on the step signal function, the channel condition impact element is quantified by using the pull-type transformation to obtain a channel condition impact mathematical expression;

[0020] Based on the step signal function, the underwater topography is quantified by using the pull-type transformation to obtain an underwater topography impact mathematical expression;

[0021] Based on the step signal function, the other ship impact is quantified by using the pull-type transformation to obtain an other ship impact quantity mathematical expression;

[0022] The ship navigation risk factors include the wind-induced drift, the flow-induced drift, the channel condition, the underwater topography and the other ship impact, and the channel condition impact element includes a channel edge, a bridge pier, a navigation mark, an obstacle and a bend.

[0023] Optionally, the plurality of environmental impact quantity mathematical expressions are matrix synthesized to obtain an environmental impact element overall quantization model, including:

[0024] According to the corresponding x-axis, y-axis and z-axis of the ship field, a plurality of diagonal matrices corresponding to the mathematical expressions of the plurality of environmental impact quantities are constructed;

[0025] The plurality of diagonal matrices are accumulated to obtain an environmental impact total quantity expression;

[0026] According to the environmental impact total quantity expression, the overall quantitative model of the environmental impact element is constructed.

[0027] Optionally, the mathematical expression of the three-dimensional dynamic ship field model is:

[0028]

[0029] In the formula, L is the length of the ship; a is the short semi-axis of the static ship field; b is the long semi-axis of the static ship field; z is the draft of the ship; f t (x, y, z) is the range size of the three-dimensional dynamic ship field; k w (t) is the influence of wind-induced drift on the ship field; f c (t) is the influence of flow-induced drift on the ship field; is the influence of a curve on the ship field; ωσ () is the influence of another ship on the ship field of the ship; is the influence of underwater topography on the ship field; (h1+

[0030] is the influence of an obstacle above the waterline on the ship field; is the influence of the edge of the channel, the pier, the beacon and the horizontal plane on the ship field.

[0031] Optionally, according to the AIS feature information, the ship track is predicted based on a preset BP neural network model, and the ship track is fitted with the radar feature data to obtain a predicted track, including:

[0032] The AIS feature information corresponding to the ship is input into the BP neural network prediction model to obtain a first predicted track;

[0033] A second predicted track of the ship is obtained based on the radar feature information;

[0034] The first predicted track and the second predicted track are fitted by using the least square method to obtain the predicted track of the ship.

[0035] Optionally, the current ship field and the predicted ship field of the ship are generated according to the three-dimensional dynamic ship field model, the predicted track and the navigation environmental impact information, including:

[0036] acquire corresponding navigation environment influence information of a plurality of predicted navigation points in the predicted track and a current position of the ship;

[0037] input the corresponding navigation environment influence information into the three-dimensional dynamic ship field model to obtain a current ship field of the ship and a plurality of predicted ship fields corresponding to the plurality of predicted navigation points.

[0038] Optionally, the judging whether the current ship field and the predicted ship field of the ship have a risk of being invaded according to the radar feature information, the AIS feature information and the river water depth information comprises:

[0039] calculating coordinates of a target relative to the predicted navigation point according to the radar feature information to obtain coordinate information of the target;

[0040] calculating a coordinate range of the predicted ship field relative to the predicted navigation point to obtain coordinate range information;

[0041] judging whether the target is located in the predicted ship field according to the coordinate information of the target and the coordinate range information of the predicted ship field;

[0042] when the target is located in the predicted ship field, the predicted ship field has a risk of being invaded;

[0043] when the target is located outside the predicted ship field, the predicted ship field does not have a risk of being invaded.

[0044] Optionally, the judging whether the current ship field and the predicted ship field of the ship have a risk of being invaded according to the radar feature information, the AIS feature information and the river water depth information further comprises:

[0045] acquiring draft depth information of the ship from the AIS feature information;

[0046] acquiring corresponding river water depth information of the plurality of predicted navigation points based on an electronic channel chart system;

[0047] comparing the draft depth of the ship with the corresponding river water depth of the plurality of navigation points according to the draft depth information and the corresponding river water depth information;

[0048] when the draft depth is smaller than the river water depth, the predicted ship field does not have a risk of being invaded;

[0049] when the draft depth is greater than or equal to the river water depth, the predicted ship field has a risk of being invaded.

[0050] Further, the application also provides an inland ship navigation risk early warning device, comprising:

[0051] A model construction unit is configured to quantize ship navigation risk factors based on a signal function and a pull-type transformation, and construct a three-dimensional dynamic ship field model;

[0052] An information acquisition unit is configured to acquire radar feature information, AIS feature information, river water depth information, and channel environment impact information of the ship;

[0053] A track prediction unit is configured to predict a ship track of the ship based on a preset BP neural network model according to the AIS feature information, and perform track fitting with the radar feature data to obtain a predicted track;

[0054] A ship field prediction unit is configured to generate a current ship field and a predicted ship field of the ship according to the three-dimensional dynamic ship field model, the predicted track, and the channel environment impact information;

[0055] A risk judgment unit is configured to judge whether the current ship field and the predicted ship field of the ship exist a risk of being invaded according to the radar feature information, the AIS feature information, and the river water depth information;

[0056] A risk early warning unit is configured to generate risk early warning information when the current ship field or the predicted ship field of the ship exists a risk of being invaded, and perform risk early warning prompting to a user.

[0057] The inland ship navigation risk early warning method provided by the application quantizes ship navigation risk factors based on a signal function and a pull-type transformation, and constructs a three-dimensional dynamic ship field model; radar feature information, AIS feature information, river water depth information, and navigation environment impact information of the ship are acquired, a track is predicted based on AIS feature information and a preset BP neural network model, and track fitting is performed with radar feature data to obtain a predicted track; a current ship field and a predicted ship field of the ship are generated according to a three-dimensional dynamic ship field model, a predicted track, and channel environment impact information; whether the current ship field and the predicted ship field of the ship exist a risk of being invaded is judged according to the radar feature information, the AIS feature information, and the river water depth information, and risk early warning information is generated when the current ship field or the predicted ship field of the ship exists a risk of being invaded, and risk early warning prompting is performed to a user. The application combines a three-dimensional dynamic ship field model and a predicted track based on multi-factor quantization under time-space changes, and improves the accuracy, rapidity, and universality of inland ship navigation risk early warning. BRIEF DESCRIPTION OF DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed in the embodiment description. Obviously, the drawings in the following description only represent some embodiments of the present application, and for those skilled in the art, other drawings can be obtained from these drawings without any creative effort.

[0059] Figure 1 A flowchart of an embodiment of the method for early warning of navigation risk of inland waterway vessels provided by the present application is shown in Figure 1.

[0060] Figure 2 A flowchart of an embodiment of step S101 of the method for early warning of navigation risk of inland waterway vessels provided by the present application is shown in Figure 2.

[0061] Figure 3 An influence diagram of other vessels on the ship domain of the subject vessel is shown in Figure 3.

[0062] Figure 4 A flowchart of an embodiment of step S202 of the method for early warning of navigation risk of inland waterway vessels provided by the present application is shown in Figure 4.

[0063] Figure 5 A structural diagram of an embodiment of the device for early warning of navigation risk of inland waterway vessels provided by the present application is shown in Figure 5. DETAILED DESCRIPTION

[0064] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments only represent some embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without any creative effort fall within the scope of the present application.

[0065] It should be understood that the schematic drawings are not drawn to scale. The flowcharts used in the present application show the operations implemented according to some embodiments of the present application. It should be understood that the operations of the flowcharts can be implemented in no particular order, and the steps without logical context relationship can be reversed in order or implemented simultaneously. In addition, one or more other operations can be added to the flowcharts or removed from the flowcharts by those skilled in the art under the guidance of the content of the present application.

[0066] Reference to "an embodiment" in this text means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the present application. The phrase appears in various places in the specification does not necessarily all refer to the same embodiment, nor is it necessarily independent or alternative embodiments to other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0067] The embodiment of the present application provides an inland ship navigation risk early warning method and device, which are described below.

[0068] Figure 1 An embodiment of the inland ship navigation risk early warning method provided by the present application is shown in a flowchart, as shown in the figure, which comprises: Figure 1

[0069] S101, quantifying ship navigation risk factors based on a signal function and a pull-type transformation, and constructing a three-dimensional dynamic ship field model;

[0070] S102, acquiring radar feature information, AIS feature information, river water depth information and navigation environment influence information of the ship;

[0071] S103, predicting the ship's track based on a preset BP neural network model according to the AIS feature information, and performing track fitting with the radar feature data to obtain a predicted track;

[0072] S104, generating a current ship field and a predicted ship field of the ship according to the three-dimensional dynamic ship field model, the predicted track and the channel environment influence information;

[0073] S105, judging whether the current ship field and the predicted ship field of the ship exist the risk of being invaded according to the radar feature information, the AIS feature information and the river water depth information;

[0074] S106, generating risk early warning information when the current ship field or the predicted ship field of the ship exists the risk of being invaded, and prompting the user for risk early warning.

[0075] ​Compared with the prior art, the inland ship navigation risk early warning method provided by the application quantifies ship navigation risk factors based on a signal function and a pull-type transformation, and constructs a three-dimensional dynamic ship field model; radar feature information, AIS feature information, river depth information and navigation environment influence information of the ship are acquired, a track is predicted based on the AIS feature information and a preset BP neural network model, and the predicted track is obtained by track fitting with the radar feature data; a current ship field and a predicted ship field are generated according to the three-dimensional dynamic ship field model, the predicted track and the channel environment influence information; whether the current ship field and the predicted ship field exist the risk of being invaded is judged according to the radar feature information, the AIS feature information and the river depth information, and risk early warning information is generated to prompt the user when the current ship field or the predicted ship field exists the risk of being invaded. The three-dimensional dynamic ship field model based on multi-factor quantization under the condition of time and space changes and the predicted track are combined, and the accuracy, rapidity and universality of the inland ship navigation risk early warning are improved.

[0076] It should be noted that in the embodiment, the AIS (Automatic Identification System) is used to acquire ship driving track, ship position and other parameters, and the AIS feature information includes current ship driving track information, driving speed, acceleration, ship position information, ship surrounding environment information, historical ship driving track, historical ship position, historical environment information and the like; the radar feature information refers to information collected by the radar and camera sensing units of the ship; the river depth information is acquired through the electronic channel chart system of the ship; and the channel environment influence information is obtained by checking various sensing devices on the ship, including wind information, water flow information and the like.

[0077] In some embodiments of the application, with reference to Figure 2 , Figure 2 The flowchart of step S101 in the inland ship navigation risk early warning method provided by the application is shown in the embodiment, and step S101 includes:

[0078] S201, quantifying the ship navigation risk factors by adopting a pull-type transformation according to a pulse signal function and a step signal function, to obtain a plurality of environment influence quantity mathematical expressions;

[0079] S202, synthesizing the plurality of environment influence quantity mathematical expressions in a matrix to obtain an environment influence element overall quantization model;

[0080] S203, constructing the three-dimensional dynamic ship field model with the scale changing with time and space based on a preset initial basic ship field model and the environment influence element overall quantization model.

[0081] It should be noted that in the ship navigation risk early warning process, the environmental factors need to be analyzed, and there are too many environmental factors. In the inland waterway traffic accident, the ship collision and grounding accident accounts for the largest proportion. Therefore, for this traffic accident, according to the characteristics of the inland ship navigation environment, the ship field environmental impact quantity of the embodiment mainly considers two natural environmental factors of wind and flow and six traffic environmental factors of channel edge, bend, bridge pier, navigation mark, navigation obstacle, underwater topography and other ships, a total of nine environmental impact factors. Other environmental impact factors can also be added according to actual conditions, and the embodiment does not limit this.

[0082] It should also be noted that in the embodiment of the application, the initial basic ship field model is a kernel static ship field model, and its mathematical expression is as follows:

[0083]

[0084] In the formula, b is the short axis in the horizontal direction, b = 1.6L, a is the long axis in the horizontal direction, a = 3L, c is the safety water depth below the waterline, and L is the ship length.

[0085] It can be understood that in the embodiment of the application, the natural environmental factors change with time, so the natural environmental factors are quantified by pulse signal functions. The influence of the traffic environmental factors is from nothing to something, which conforms to the law of step function change with time, so the traffic environmental factors are quantified by step functions, and the environmental impact quantity mathematical model is constructed. The formula of the environmental impact quantity mathematical model is substituted into the formula of the initial ship field model, and the expression of the three-dimensional dynamic ship field model is obtained, and then the three-dimensional dynamic ship field model is constructed.

[0086] In some embodiments of the application, step S201 comprises:

[0087] Based on the pulse signal function, the wind-induced drift is quantified by Laplace transform, and the wind-induced drift mathematical expression is obtained;

[0088] Based on the pulse signal function, the flow-induced drift is quantified by Laplace transform, and the flow-induced drift mathematical expression is obtained;

[0089] Based on the step signal function, the channel condition impact element is quantified by Laplace transform, and the channel condition impact mathematical expression is obtained;

[0090] Based on the step signal function, the underwater topography is quantified by Laplace transform, and the underwater topography impact mathematical expression is obtained;

[0091] Based on the step signal function, the other ship impact is quantified by Laplace transform, and the other ship impact quantity mathematical expression is obtained;

[0092] The ship navigation risk factors include wind-induced drift, current-induced drift, channel conditions, underwater topography, and the influence of other vessels. The factors affecting channel conditions include channel edges, bridge piers, navigation marks, obstructions, and curves.

[0093] It is understood that, in the embodiments of the present invention, by introducing pulse signal function, step signal function and linear extrema, and then using inverse Laplace transform, the mathematical functions corresponding to the changes of each environmental influencing factor over time can be obtained.

[0094] It should be understood that, in practical implementation, the expression for wind-induced drift is:

[0095]

[0096] In the formula k w (t) represents the impact of wind-induced drift on the ship's domain, with K typically ranging from 0.038 to 0.041. a (m2) represents the area exposed to wind at the upper part of the waterline, B w (m2) represents the windward area below the waterline, α f (°) represents the angle between the wind and the course, S(m) represents the length of the major axis of the ship, U(m / s) represents the speed of the water flow, β(°) represents the angle of pressure, and t represents time.

[0097] The expression for flow-induced drift is:

[0098]

[0099] In the formula, f c (t) represents the effect of flow-induced drift on the ship's domain, V represents the ship's speed, and the rest are the same as above.

[0100] It should be noted that the factors affecting waterway conditions include the waterway edge, the distance between the vessel and the bridge pier, the safe distance between the vessel's outer hull and navigational aids, the safe distance between the vessel's outer hull and obstructions, and curves. In specific implementation, the expression for the safe distance d1 between the vessel's outer hull and the waterway edge is:

[0101] d1=γ1×(B s +L·sinβ)(0.34<γ1<0.40)

[0102] In the formula, (B s +L·sinβ) is the track bandwidth, and γ1 is the channel edge multiplication factor. The safe distance from the outer side of the cargo ship to the channel edge can be taken as 0.34 to 0.40 times the track width.

[0103] The expression for the safe width distance d2 between the ship and the bridge pier is:

[0104] d2=γ2×(Bs +L*sinβ)(0.5<γ2<0.6)

[0105] Wherein, the intersection angle between the normal direction of the axis of the river-crossing building on the natural and channelized river and the flow direction of the water flow is not greater than 5°, the safety width between the ship and the two side piers can be 0.6 times the width of the track belt in the first to fifth grade channel, and 0.5 times the width of the track belt in the sixth and seventh grade channel, and γ2 is the pier multiplication coefficient.

[0106] The expression of the safety distance d3 between the outer side of the ship and the navigation mark is:

[0107] d3=1.0×B s

[0108] The expression of the safety distance d4 between the outer side of the ship and the navigation mark is:

[0109] d4=1.5×B s

[0110] It can be understood that in the specific implementation, the influence quantity f hi (t) of the channel other than the bend is:

[0111]

[0112] It should be noted that when the ship is sailing in the bend, the ship needs to be constantly manipulated to adjust the direction, and when the curvature radius of the bend is less than 3 times the length of the ship, the width of the ship field needs to be increased to ensure that the ship and the curved channel maintain an additional safety distance, and the expression of the width increase value is:

[0113]

[0114] In the formula, ΔB(m) is the bend widening value, R(m) is the corresponding curvature radius of the bend, B(m) is the corresponding standard width of the channel, and L(m) is the length of the ship. When the curvature radius of the bend is greater than 3 times the length of the ship and less than 6 times the length of the ship, it can be determined according to the specific sailing situation whether the width of the ship field needs to be increased.

[0115] It can be understood that the underwater topography will affect the size of the ship field in the vertical horizontal direction, and the safety water depth calculation formula of the ship sailing is:

[0116] U i =U0+U1+U 2i +U3+U4(i=1,2,3,4)

[0117] In the formula, U0 represents the draft of the ship, U1 represents the sinking amount of the ship body, U 2iU represents the clearance depth under the keel of the ship, and U4 represents the clearance depth for trim, which is generally taken as 0.15m when the ship type is an oil tanker or bulk carrier; when i is 1, U... i This represents a relatively loose, muddy bottom; while when i is 2, U i This represents sandy soil; also, when i is 3, U i This represents the hard rock stratum; and when i is 4, U i This indicates a hard substrate.

[0118] underwater topography influence f ui The expression for (t) is:

[0119]

[0120] refer to Figure 3 , Figure 3 This is a diagram illustrating the impact of other vessels on the vessel's territorial domain. Figure 3 It can be seen that when another ship approaches, the changes in the major and minor axes of the ellipse in the ship's domain conform to a two-dimensional normal distribution. The standard deviation σ of the ship's position offset along the major and minor axes of the ellipse accumulates over time, and its impact on the horizontal scale of the ship's domain is denoted as σ(t). ω is the product factor of the major and minor semi-axes of the uncertain ellipse under the corresponding confidence coefficient. Figure 4 In this context, ωσ1(t) represents the increment of the long half-axis in the ship domain, and ωσ2(t) represents the increment of the short half-axis.

[0121] In some embodiments of the present invention, reference is made to... Figure 4 , Figure 4 This is a flowchart illustrating an embodiment of step S202 of the inland waterway vessel navigation risk early warning method provided by the present invention. Figure 4 It can be seen that step S202 includes:

[0122] S401. Based on the x-axis, y-axis, and z-axis corresponding to the shipbuilding field, construct multiple diagonal matrices corresponding to the mathematical expressions of the multiple environmental impact quantities;

[0123] S402. Sum the multiple diagonal matrices to obtain the expression for the total environmental impact;

[0124] S403. Construct an overall quantitative model of the environmental impact elements based on the total environmental impact expression.

[0125] It is understandable that this invention constructs a three-dimensional dynamic ship domain model, which needs to consider the impact of environmental influences on the ship domain from a three-dimensional perspective. In specific implementation, the x-axis, y-axis, and z-axis of the ship domain are selected according to the setting of the diagonal matrix. Then, the unit impulse response matrix of wind-induced drift at time t is:

[0126]

[0127]

[0128] wherein, corresponding to the x-axis, corresponding to the y-axis, corresponding to the z-axis.

[0129] Similarly, the unit impulse response matrix of the flow-induced drift at time t is:

[0130]

[0131] The vertical distances between the ship and the edges of the channel, the piers, the navigation marks and the navigation obstructions are D1, D2, D3, D4, D i = min(D1, D2, D3, D4), the unit step response matrix of the channel at time t is:

[0132]

[0133] The unit step response matrix of the bend at time t is:

[0134]

[0135]

[0136] The unit step response matrix of the underwater terrain at time t is:

[0137]

[0138] The corresponding matrix of the influence of the other ship on the subject ship at time t is:

[0139]

[0140] The total amount of environmental influence A i The expression is:

[0141]

[0142] In some embodiments of the present application, the mathematical expression of the three-dimensional dynamic ship field model is:

[0143]

[0144] In the formula, L is the length of the ship; a is the short semi-axis of the static ship field; b is the long semi-axis of the static ship field; z is the draft of the ship; f t (x, y, z) is the size of the three-dimensional dynamic ship field; k w (t) is the influence of the wind-induced drift on the ship field; f c (t) is the influence of the flow-induced drift on the ship field; for the influence of a bend on the ship domain; ωs (t) for the influence of another ship on the ship domain of the target; for the influence of underwater topography on the ship domain; for the influence of an above-water obstruction on the ship domain; for the influence of a channel edge, bridge pier, navigation mark and horizontal plane obstruction on the ship domain.

[0145] In some embodiments of the present application, step S102 comprises:

[0146] inputting the AIS feature information corresponding to the own ship into the BP neural network prediction model to obtain a first predicted track;

[0147] obtaining a second predicted track of the own ship based on the radar feature information;

[0148] performing piecewise fitting on the first predicted track and the second predicted track by using the least square method to obtain the predicted track of the own ship.

[0149] It should be noted that the BP neural network refers to an error back propagation neural network, which has two parts, the first part is the forward transmission of information, and the second part is the backward transmission of error; it has a three-layer structure, the first layer is the input layer, which transmits the information as input to the hidden layer; the second layer structure is the hidden layer, which transmits the input signal to the output layer after processing; the third layer is the output layer, when the expected output signal is inconsistent with the actual output signal, the output layer will transmit the error to the input layer through the hidden layer, so as to calculate and correct the weight of each unit, and the weight correction is an uninterrupted repeated process, through continuous forward transmission and backward correction, the accuracy of the output signal is improved.

[0150] It can be understood that, in the embodiment of the present application, the dynamic information and static information of the ship are acquired in real time by the AIS device, and other ship-transmitted information is also received and historical data is updated, the position and sailing speed of the ship are determined by the radar, the BP neural network prediction model is trained by using the historical information of the ship, the trained BP neural network prediction model is obtained, the information required for the track prediction is input into the BP neural network prediction model from the AIS device, and the first predicted track is obtained. However, the current AIS information transmission frequency of the inland river is generally 15-30s, and the update rate of the AIS data directly affects the ship track prediction accuracy, so, in the embodiment of the present application, the ship information obtained by the radar is input into the BP neural network prediction model, the second predicted track is obtained, and the least square method is used to fit the first predicted track and the second predicted track in three segments, so that the final predicted track is obtained, and the accuracy of the track prediction is improved. The number of fitting segments is not limited in the embodiment of the present application.

[0151] In some embodiments of the present application, step S104 comprises:

[0152] Obtaining corresponding sailing environment influence information of a plurality of predicted sailing points in the predicted track and the current position of the ship;

[0153] Inputting the corresponding sailing environment influence information into the three-dimensional dynamic ship field model to obtain a current ship field of the ship and a plurality of predicted ship fields corresponding to the plurality of predicted sailing points.

[0154] It can be understood that, after the predicted track is obtained, the coordinates of the ship in a future period of time can be known, the ship field environment influence quantity corresponding to the coordinates is obtained, and the ship field environment influence quantity is input into the three-dimensional dynamic ship field model, so that the ship field corresponding to the coordinate point position is obtained. The ship field is the predicted ship field.

[0155] In some embodiments of the present application, step S105 comprises:

[0156] Calculating the coordinates of the target relative to the predicted sailing point according to the radar feature information to obtain coordinate information of the target;

[0157] Calculating the coordinate range of the predicted ship field relative to the predicted sailing point to obtain coordinate range information;

[0158] Judging whether the target is located in the predicted ship field according to the coordinate information of the target and the coordinate range information of the predicted ship field;

[0159] When the target is located in the predicted ship field, the predicted ship field has a risk of being invaded;

[0160] When the target is located outside the predicted ship field, the predicted ship field is not at risk of being invaded.

[0161] It should be understood that in the specific implementation, the target includes a ship, an obstacle, and the like, the position coordinates of the ship to be reached at time t are determined according to a predicted track, and the position coordinates of the waterway influence quantity at time t are determined, whether the position coordinates of the waterway influence quantity are in the predicted ship field corresponding to the position coordinates of the ship at time t is determined, if yes, the predicted ship field is invaded, indicating that there is a risk of collision at time t of the ship, a risk warning is given in advance, and a water traffic accident is avoided.

[0162] It should be further explained that when the target is a ship, the BP neural network model is used to predict the track of the ship, and whether the ship will invade the corresponding predicted ship field at a future time is determined. For example, after the track of the ship is predicted, the position information of the ship at time t is obtained, whether the ship at time t is located in the ship field of the ship at time t is determined through the position information, and whether the ship field of the ship at time t is at risk of being invaded by the ship at time t (the position of the ship at time t is the predicted navigation point, and the ship field of the ship at time t is the predicted ship field) is determined.

[0163] In some embodiments of the present application, step S105 further comprises:

[0164] The draft information of the ship is obtained from the AIS feature information;

[0165] The corresponding river water depth information of the plurality of predicted navigation points is obtained based on an electronic navigation chart system;

[0166] The draft information and the corresponding river water depth information are compared to determine the size of the draft of the ship and the river water depth corresponding to the plurality of navigation points;

[0167] When the draft is smaller than the river water depth, the predicted ship field is not at risk of being invaded;

[0168] When the draft is greater than or equal to the river water depth, the predicted ship field is at risk of being invaded.

[0169] It should be noted that the three-dimensional ship field model warning includes collision warning in the horizontal dimension and grounding warning in the vertical dimension. The ship generally has an electronic navigation chart system, which contains river water depth information. By comparing the river water depth value with the predicted navigation point corresponding to the ship draft value, it can be determined whether the predicted ship field corresponding to the predicted navigation point is at risk of being invaded (i.e., whether the ship is at risk of grounding).

[0170] Reference Figure 5The embodiment also provides an inland ship navigation risk early warning device.

[0171] The model construction unit 501 is configured to quantize ship navigation risk factors based on a signal function and a pull-type transformation, and construct a three-dimensional dynamic ship field model.

[0172] The information acquisition unit 502 is configured to acquire radar feature information, AIS feature information, river water depth information and channel environment influence information of the ship.

[0173] The track prediction unit 503 is configured to predict a track of the ship based on a preset BP neural network model according to the AIS feature information, and perform track fitting with the radar feature data to obtain a predicted track.

[0174] The ship field prediction unit 504 is configured to generate a current ship field and a predicted ship field of the ship according to the three-dimensional dynamic ship field model, the predicted track and the channel environment influence information.

[0175] The risk judgment unit 505 is configured to judge whether the current ship field and the predicted ship field of the ship have a risk of being invaded according to the radar feature information, the AIS feature information and the river water depth information.

[0176] The risk early warning unit 506 is configured to generate risk early warning information to give a risk early warning prompt to a user when the current ship field or the predicted ship field of the ship has a risk of being invaded.

[0177] The above-mentioned embodiment provides an inland ship navigation risk early warning device, which can realize the scheme described in the above-mentioned inland ship navigation risk early warning method embodiment, and the specific principles of the above-mentioned units can refer to the inland ship navigation risk early warning method embodiment, which will not be described here.

[0178] Those skilled in the art can understand that all or part of the processes of the above-mentioned embodiments can be completed by a computer program instructing related hardware, and the program can be stored in a computer readable storage medium, such as a magnetic disk, an optical disk, a read-only memory or a random access memory.

[0179] The above-mentioned is only a preferred specific embodiment of the present application, but the protection scope of the present application is not limited to this. Any changes or replacements within the technical range disclosed by the present application can be easily thought by those skilled in the art, which should be covered in the protection scope of the present application.

Claims

1. A method for early warning of navigation risks for inland waterway vessels, characterized in that, include: Based on signal functions and Laplace transforms, a three-dimensional dynamic ship domain model is constructed by quantifying ship navigation risk factors and including: quantifying the ship navigation risk factors using Laplace transforms based on pulse signal functions and step signal functions to obtain multiple mathematical expressions for environmental impact quantities; synthesizing the multiple mathematical expressions for environmental impact quantities into a matrix to obtain an overall quantitative model of environmental impact elements; and constructing the three-dimensional dynamic ship domain model with spatiotemporal scale variations based on a preset initial basic ship domain model and the overall quantitative model of environmental impact elements. Acquire the ship's radar signature, AIS signature, river depth information, and navigation environment impact information; Based on the AIS feature information, the ship's trajectory is predicted using a preset BP neural network model, and the trajectory is fitted with the radar feature data to obtain the predicted trajectory. The current and predicted ship domains of the ship are generated based on the three-dimensional dynamic ship domain model, the predicted trajectory, and the navigation environment impact information. Based on the radar feature information, the AIS feature information, and the river water depth information, determine whether there is a risk of intrusion into the current ship area and the predicted ship area of ​​this vessel; When there is a risk of intrusion into the current or predicted area of ​​the ship, a risk warning message is generated to alert the user. The mathematical expression for the three-dimensional dynamic ship domain model is: In the formula, L is the length of the ship; a is the short half-axis in the static ship domain; b is the long half-axis in the static ship domain; and z is the ship's draft. The size of the three-dimensional dynamic ship domain; The impact of wind-induced drift on the shipbuilding industry; The impact of flow-induced drift on the shipbuilding industry; The impact of curves on the shipping industry; The impact of other vessels on the ship's maritime domain; The impact of underwater topography on the shipping industry; The impact of obstructions above the waterline on the shipping industry; The impact of obstructions to navigation, such as those along the edge of the waterway, bridge piers, navigation marks, and the horizontal plane, on the shipping industry.

2. The inland waterway vessel navigation risk early warning method according to claim 1, characterized in that, The method involves quantifying the ship navigation risk factors using a Laplace transform based on the pulse signal function and the step signal function, resulting in multiple mathematical expressions for environmental impact quantities, including: Based on the pulse signal function, the wind-induced drift is quantized using the Laplace transform to obtain the mathematical expression of the wind-induced drift. Based on the pulse signal function, the flow-induced drift is quantized using the Laplace transform to obtain the mathematical expression for the flow-induced drift; Based on the step signal function, the Laplace transform is used to quantify the factors affecting waterway conditions, and the mathematical expression of the influence of waterway conditions is obtained. Based on the step signal function, the underwater topography is quantified by Laplace transform to obtain the mathematical expression of the underwater topography influence; Based on the step signal function, the influence of other ships is quantified by Laplace transform to obtain a mathematical expression for the influence of other ships. The ship navigation risk factors include wind-induced drift, current-induced drift, channel conditions, underwater topography, and the influence of other vessels. The factors affecting channel conditions include channel edges, bridge piers, navigation marks, obstructions, and curves.

3. The inland waterway vessel navigation risk early warning method according to claim 1, characterized in that, The step of synthesizing the mathematical expressions of the multiple environmental impact quantities into a matrix to obtain an overall quantitative model of environmental impact factors includes: Based on the x-axis, y-axis, and z-axis corresponding to the shipbuilding field, construct multiple diagonal matrices corresponding to the mathematical expressions of the multiple environmental impact quantities; By summing the multiple diagonal matrices, the total environmental impact expression is obtained; Based on the expression for the total environmental impact, construct an overall quantitative model for the environmental impact elements.

4. The inland waterway vessel navigation risk early warning method according to claim 1, characterized in that, Based on the AIS feature information, the ship's trajectory is predicted using a preset BP neural network model, and then fitted with the radar feature data to obtain the predicted trajectory, including: The AIS feature information corresponding to this ship is input into the BP neural network prediction model to obtain the first predicted trajectory; The second predicted trajectory of the ship is obtained based on the radar feature information; The first and second predicted trajectories are piecewise fitted using the least squares method to obtain the predicted trajectories of the ship.

5. The inland waterway vessel navigation risk early warning method according to claim 1, characterized in that, The process of generating the current and predicted ship domains based on the three-dimensional dynamic ship domain model, the predicted trajectory, and the navigation environment impact information includes: Obtain the corresponding navigation environment impact information of multiple predicted navigation points and the ship's current position in the predicted trajectory; The corresponding navigation environment impact information is input into the three-dimensional dynamic ship domain model to obtain the current ship domain of the ship and the multiple predicted ship domains corresponding to the multiple predicted navigation points.

6. The inland waterway vessel navigation risk early warning method according to claim 5, characterized in that, The step of determining whether there is a risk of intrusion into the current and predicted ship domains of the vessel based on the radar feature information, the AIS feature information, and the river depth information includes: The coordinates of the target relative to the predicted navigation point are calculated based on the radar feature information to obtain the target's coordinate information; The coordinate range of the predicted vessel area relative to the predicted navigation point is calculated to obtain coordinate range information; Based on the coordinate information of the object and the coordinate range information of the predicted ship area, it is determined whether the object is located within the predicted ship area; When the target is located within the predicted vessel area, the predicted vessel area is at risk of being compromised. When the target is located outside the predicted vessel area, there is no risk of the predicted vessel area being invaded.

7. The inland waterway vessel navigation risk early warning method according to claim 5, characterized in that, The method for determining whether there is a risk of intrusion into the current and predicted vessel domains based on the radar signature information, AIS signature information, and river depth information also includes: Obtain the ship's draft depth information from the AIS feature information; The corresponding river depth information of the multiple predicted navigation points is obtained based on the electronic navigation chart system; Compare the ship's draft with the corresponding river depth information based on the draft information and the corresponding river depth information at the multiple navigation points. When the draft is less than the river depth, there is no risk of intrusion into the predicted vessel area; When the draft is greater than or equal to the river depth, the predicted vessel area is at risk of intrusion.

8. An inland waterway vessel navigation risk early warning device, used to execute the inland waterway vessel navigation risk early warning method as described in any one of claims 1-7, characterized in that, include: The model building unit is used to quantify ship navigation risk factors based on signal functions and Laplace transforms, and to build a three-dimensional dynamic ship domain model. The information acquisition unit is used to acquire the ship's radar signature information, AIS signature information, river water depth information, and waterway environmental impact information. The trajectory prediction unit is used to predict the trajectory of the ship based on the AIS feature information and a preset BP neural network model, and to fit the trajectory with the radar feature data to obtain the predicted trajectory. The ship domain prediction unit is used to generate the current ship domain and the predicted ship domain of the ship based on the three-dimensional dynamic ship domain model, the predicted trajectory and the waterway environmental impact information. The risk assessment unit is used to determine whether there is a risk of intrusion into the current ship area and the predicted ship area of ​​the vessel based on the radar feature information, the AIS feature information and the river water depth information. The risk warning unit is used to generate risk warning information and provide risk warning prompts to users when there is a risk of intrusion into the current or predicted ship area of ​​the ship.

Images