Bridge anti-collision early warning method and device

Abstract

The invention provides a bridge anti-collision early warning method and device, and relates to the field of bridge anti-collision early warning. The method comprises the following steps: acquiring visual perception data corresponding to a target ship, and acquiring bridge structure data corresponding to a target bridge; constructing multi-source analysis data according to the visual perception data and the bridge structure data; according to the multi-source analysis data, obtaining a ship response lag factor set and a bridge structure risk constraint set through preprocessing operation; inputting the multi-source analysis data into a ship trajectory prediction model, and outputting a predicted trajectory path corresponding to the target ship; according to the predicted trajectory path and the ship response lag factor set, constructing a dynamic safety envelope area between the target bridge and the target ship; obtaining a dynamic safety envelope area, a bridge structure risk constraint set and a target intersection in the relative movement process of the target ship and the target bridge; and based on the target intersection, calculating a risk level of collision between the target ship and the target bridge, and sending an anti-collision early warning prompt to the target ship according to the risk level. According to the method, the problems of frequent occurrence of false alarm or missing report and remarkable early warning hysteresis caused by difficulty in effective modeling of ship nonlinear trajectory offset in a strong wind current environment in a traditional bridge anti-collision technology are solved.

Description

Technical Field

[0001] This application relates to the field of bridge collision prevention warning, and particularly to a bridge collision prevention warning method and device for a large bridge. Background Art

[0002] In areas with dense waterborne traffic, large bridge structures face an increasingly severe risk of ship collisions. With the continuous development of key technologies such as intelligent perception, trajectory modeling, and multi-source fusion, the ship-bridge interaction risk perception system is evolving towards the direction of "global perception, dynamic prediction, and intelligent decision-making".

[0003] However, most traditional bridge collision prevention technologies rely on single-sensor (such as AIS or radar) data and static rule models for risk judgment. Since it is difficult to effectively model the non-linear trajectory deviation of ships in this way, it is difficult to effectively model the non-linear trajectory deviation of ships in a strong wind and current environment, resulting in frequent false alarms or missed alarms and significant warning lags.

[0004] Therefore, there is an urgent need for a bridge collision prevention warning method and device for a large bridge. Summary of the Invention

[0005] This application provides a bridge collision prevention warning method and device for a large bridge, which solves the problem that traditional bridge collision prevention technologies are difficult to effectively model the non-linear trajectory deviation of ships in a strong wind and current environment, resulting in frequent false alarms or missed alarms and significant warning lags.

[0006] In the first aspect of this application, a bridge collision prevention warning method is provided. The method includes: obtaining visual perception data corresponding to a target ship and obtaining bridge structure data corresponding to a target bridge; constructing multi-source analysis data based on the visual perception data and the bridge structure data; obtaining a set of ship response lag factors and a set of bridge structure risk constraints through preprocessing operations according to the multi-source analysis data; inputting the multi-source analysis data into a ship trajectory prediction model and outputting a predicted trajectory path corresponding to the target ship; constructing a dynamic safety envelope area between the target bridge and the target ship according to the predicted trajectory path and the set of ship response lag factors; obtaining the intersection of the dynamic safety envelope area and the set of bridge structure risk constraints during the relative movement of the target ship and the target bridge; calculating the risk level of a collision between the target ship and the target bridge based on the intersection, and sending a collision prevention warning prompt to the target ship according to the risk level.

[0007] Optionally, constructing multi-source analysis data based on the visual perception data and the bridge structure data specifically includes: registering and fusing the visual perception data and the bridge structure data in the same time domain and coordinate system to construct multi-source analysis data, which includes current motion state data, environmental disturbance factor data, bridge structure topology data, historical structure response data, and image feature extraction data.

[0008] Optionally, according to the multi-source analysis data, obtain the ship response lag factor set and the bridge structure risk constraint set through preprocessing operations, specifically including: constructing the ship response lag factor set according to the current motion state data and environmental disturbance factor data, where the ship response lag factor set includes the rudder effect lag time factor, the fluid dynamic interference delay factor, and the propulsion system inertia factor; constructing the bridge structure risk constraint set according to the bridge structure topology data, historical structure response data, and image feature extraction data, where the bridge structure risk constraint set includes the physical boundary constraint of the bridge pier, the historical high-risk point constraint, and the predefined protection area constraint.

[0009] Optionally, input the multi-source analysis data into the ship trajectory prediction model and output the predicted trajectory path corresponding to the target ship, specifically including: constructing the ship trajectory prediction model through a long short-term memory network and a probabilistic state transition diagram structure; based on the multi-source analysis data, the ship trajectory prediction model performs continuous integral prediction on the state evolution of the target ship over a future preset time window to output the predicted trajectory path corresponding to the target ship.

[0010] Optionally, perform continuous integral prediction on the state evolution of the target ship over a future preset time window to output the predicted trajectory path corresponding to the target ship, specifically including: outputting the predicted trajectory path corresponding to the target ship according to the following relational expression:

[0011]

[0012] where, represents the predicted position corresponding to the target ship at time t+τ, P(t) is the current position vector corresponding to the target ship at time t, τ is the length of the prediction time window, represents the velocity vector, represents the acceleration vector, represents, is the water flow disturbance vector, is the wind field disturbance vector, θ(s) is the course angle, is the operator for rotating and mapping the disturbance direction based on the course angle, represents the ship response lag correction vector, s is the integration variable, representing any time point during the process of advancing from the current time t to the prediction time τ, and λ1, λ2, and λ3 are all adjustment coefficients.

[0013] Optionally, a dynamic safety envelope area between the target bridge and the target ship is constructed according to the predicted trajectory path and the set of ship response lag factors, specifically including: constructing, according to the predicted trajectory path and the set of ship response lag factors, a plurality of confidence intervals corresponding to the position of the target ship at each predicted moment, one predicted moment corresponding to one confidence interval, and the confidence intervals being modeled as a set of multi-ellipsoids that dynamically evolve over time in three-dimensional space; constructing an interval union according to the plurality of confidence intervals, and using the interval union as the dynamic safety envelope area between the target bridge and the target ship.

[0014] Optionally, based on the target intersection, the risk level of collision between the target ship and the target bridge is calculated, and according to the risk level, an anti-collision warning prompt is sent to the target ship, specifically including: obtaining, through spatial overlap analysis, the intersection parameters of the dynamic safety envelope area and the bridge structure risk constraint set during the relative movement process, the intersection parameters including the intersection area parameter, the intersection duration parameter, and the intersection position parameter; constructing a multi-dimensional risk assessment vector according to the intersection parameters, and inputting the multi-dimensional risk assessment vector into a weighted fuzzy inference model; calculating the risk level index value of the collision between the target ship and the target bridge according to the weighted fuzzy inference model; if it is confirmed that the risk level index value is greater than the preset risk level threshold, it is confirmed that there is a collision risk between the target ship and the target bridge, and an anti-collision warning prompt is sent to the target ship in a preset manner.

[0015] In the second aspect of the present application, an anti-collision warning device for a bridge is provided, and the device includes an acquisition module and a processing module, wherein,

[0016] The acquisition module is used to acquire the visual perception data corresponding to the target ship, and acquire the bridge structure data corresponding to the target bridge; construct multi-source analysis data according to the visual perception data and the bridge structure data; and obtain the set of ship response lag factors and the bridge structure risk constraint set through preprocessing operations according to the multi-source analysis data.

[0017] The processing module is used to input the multi-source analysis data into a ship trajectory prediction model, and output the predicted trajectory path corresponding to the target ship; construct a dynamic safety envelope area between the target bridge and the target ship according to the predicted trajectory path and the set of ship response lag factors; obtain the target intersection of the dynamic safety envelope area and the bridge structure risk constraint set during the relative movement process between the target ship and the target bridge; calculate the risk level of collision between the target ship and the target bridge based on the target intersection, and send an anti-collision warning prompt to the target ship according to the risk level.

[0018] In a third aspect of the present application, an electronic device is provided, including a processor, a memory, a user interface, and a network interface. The memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the method of any one of the above.

[0019] In a fourth aspect of the present application, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to perform the method of any one of the above.

[0020] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0021] 1. Obtain the visual perception data corresponding to the target ship and the bridge structure data corresponding to the target bridge; construct multi-source analysis data according to the visual perception data and the bridge structure data; obtain the ship response lag factor set and the bridge structure risk constraint set through preprocessing operations according to the multi-source analysis data; input the multi-source analysis data into the ship trajectory prediction model and output the predicted trajectory path corresponding to the target ship; construct the dynamic safety envelope area between the target bridge and the target ship according to the predicted trajectory path and the ship response lag factor set; obtain the intersection of the dynamic safety envelope area and the bridge structure risk constraint set during the relative movement of the target ship and the target bridge; calculate the collision risk level between the target ship and the target bridge based on the intersection, and send a collision avoidance warning prompt to the target ship according to the risk level. By fusing visual perception and bridge structure data and constructing a multi-source dynamic analysis model, high-precision prediction and intelligent evaluation of the future state evolution of the ship and the collision risk in the bridge area are realized. Therefore, the perception accuracy of the collision avoidance warning system in a complex environment is greatly improved, and the frequency of false alarms or missed alarms is reduced.

[0022] 2. Construct a ship trajectory prediction model through a long short-term memory network and a probabilistic state transition diagram structure; based on the multi-source analysis data, the ship trajectory prediction model performs continuous integral prediction on the state evolution of the target ship within a future preset time window to output the predicted trajectory path corresponding to the target ship, so as to fully capture the non-linear motion characteristics and state change trends of the ship in a complex disturbance environment, realize high-precision dynamic fitting of the future path, and improve the robustness and credibility of trajectory prediction under non-steady conditions such as strong wind and current, sudden change of water flow, and rudder effect lag, providing a path basis with continuity, differentiability, and physical consistency for subsequent collision risk determination and dynamic safety boundary construction.

[0023] 3. Based on the predicted trajectory path and the set of ship response lag factors, construct multiple confidence intervals corresponding to the position of the target ship at each predicted moment, and construct the union of intervals based on the multiple confidence intervals. Use the union of intervals as the dynamic safety envelope area between the target bridge and the target ship, so as to comprehensively express the possible position range of the ship under the influence of factors such as speed disturbance, control lag, and environmental change during future navigation, and form a risk envelope model with time continuity, space extensibility, and response elasticity, providing a dynamic space criterion for judging whether it enters the highly sensitive area of the bridge structure. Description of the Drawings

[0024] Figure 1 is a schematic flowchart of a method for warning against bridge collisions provided by an embodiment of the present application;

[0025] Figure 2 is a schematic block diagram of a device for warning against bridge collisions provided by an embodiment of the present application;

[0026] Figure 3 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.

[0027] Description of the reference numerals: 21, acquisition module; 22, processing module; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. Detailed Embodiments

[0028] In order to enable those skilled in the art to better understand the technical solutions in this specification, the following will clearly and completely describe the technical solutions in the embodiments of this specification in conjunction with the accompanying drawings in the embodiments of this specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments.

[0029] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. As used in the specification of the present application, the singular forms "a", "one kind", "the", "above-mentioned", "this", and "this one" are also intended to include the plural forms, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to and includes any or all possible combinations of one or more of the listed items.

[0030] Hereinafter, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as implying or suggesting relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the embodiments of the present application, unless otherwise stated, the meaning of "a plurality" is two or more.

[0031] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0032] Please refer to Figure 1 , which shows a schematic flowchart of a bridge anti-collision warning method provided by an embodiment of the present application. The flowchart mainly includes the following steps: S101 to S107.

[0033] Step S101, obtain the visual perception data corresponding to the target ship and obtain the bridge structure data corresponding to the target bridge.

[0034] Specifically, the target ship is continuously monitored by non-limiting types of image acquisition devices, radar devices, laser scanning devices, and other adapted perception devices set at different positions in the bridge area to obtain visual perception data. The image acquisition devices may include, but are not limited to, visible light cameras, infrared cameras, thermal imaging devices, panoramic imaging systems, etc. The radar devices may include millimeter wave radars, lidars, hybrid band imaging radars, etc. The laser scanning device may be a static or dynamic three-dimensional laser point cloud acquisition device for extracting the ship's outer contour and the attitude change characteristics during the movement change process. The bridge structure data can be obtained in various ways, including, but not limited to, constructing a bridge structure health monitoring system, BIM model platform, geographic information system, historical structure database, and edge computing device collaboration deployed in the bridge area. The target ship is any ship approaching the bridge area, and the target bridge is any bridge structure above the ship navigation water area. In the embodiments of the present application, the size, quantity, type, navigation mode, load characteristics, propulsion mode, etc. of the target ship are not limited, and the bridge type structure, bridge deck height, number of bridge piers, arrangement form of bridge piers, bridge body material, structural complexity of the bridge area, and whether anti-collision facilities are set of the target bridge are not limited. It can be applied to large-span bridge structures in various forms such as highway bridges, railway bridges, pipeline bridges, cross-sea bridges, and combined bridges, and has good versatility and engineering adaptability.

[0035] Visual perception data is not limited to a specific structure and may specifically include, but is not limited to: ship image sequences, ship contours extracted from a single-frame image, scale, direction features, bow pointing information, image timestamps, pose estimation parameters, class labels output based on object recognition algorithms, target box position information, image optical flow vector fields, prior target behavior features generated based on deep networks, estimated values of the relative distance between ships, etc.; in some configurations, visual perception data may also include high-order semantic data such as target tracking trajectories established between multiple frames of images and structured motion event description information; Bridge structure data can be obtained through the collaboration of a bridge structure health monitoring system, BIM model platform, geographic information system, historical structure database, and edge computing devices deployed in the bridge area. The data source method is not limited and can be obtained through various methods such as sensor measurements, digital modeling, remote sensing extraction, or manual annotation. The specific data content may include, but is not limited to: the overall spatial coordinate position of the bridge, geometric parameters (height, width, centerline coordinates, etc.) of the bridge piers and bridge decks, bridge orientation, positioning data of each bridge pier in the reference coordinate system, 3D structure model of the bridge, historical impact position annotation information, structural bearing capacity assessment results, navigable cross-section data in the bridge area, wind and flow channel characteristics, historical risk area division layers, and high-frequency collected bridge dynamic response data (such as pier acceleration, vibration modes, etc.).

[0036] Step S102, construct multi-source analysis data according to the visual perception data and the bridge structure data.

[0037] Specifically, perform unified coordinate mapping, time alignment, semantic fusion, and feature recombination on the original information from different types of perception devices and data channels to generate a spatio-temporal consistency analysis data set for subsequent processing.

[0038] In a possible implementation manner, step S102 further includes: registering and fusing the visual perception data and the bridge structure data in the same time domain and coordinate system to construct multi-source analysis data, which includes current motion state data, environmental disturbance factor data, bridge structure topology data, historical structure response data, and image feature extraction data.

[0039] Specifically, register and fuse all the visual perception data and the bridge structure data in step S101 in the same time domain and coordinate system. This registration and fusion process first performs temporal alignment on various types of asynchronously collected perception data based on a unified time reference, and realizes the time consistency of the image frame sequence and the structure monitoring data through interpolation, resampling, or synchronization mechanisms; subsequently, use a spatial registration method to convert the local coordinates in the ship image coordinates, radar positions, laser point clouds, or other sensing data to the global navigation reference system or engineering local coordinate system consistent with the bridge structure data, and use object feature point recognition, known reference mark positioning, or multi-source matching algorithms to achieve spatial integrated modeling.

[0040] The multi-source analysis data includes current motion state data, environmental disturbance factor data, bridge structure topology data, historical structure response data, and image feature extraction data. Among them, the current motion state data includes dynamic parameters such as the speed, acceleration, heading angle, angular velocity, track curvature, and their time-varying change rates of the target ship under continuous frame images or radar measurements; the environmental disturbance factor data includes external influence variables such as wind speed and direction field, water flow vector distribution, wave excitation estimation, and the component of the hull's flow direction; the bridge structure topology data includes pier boundary models, deck space channel parameters, three-dimensional structure grid information, and the water-crossing section of the bridge area; the historical structure response data includes bridge vibration modes extracted from sensor historical records, impact response frequency domain characteristics, and statistical distributions of high-risk points; the image feature extraction data includes the hull geometric contour, image scale features, ship type recognition labels, motion trend optical flow field, target tracking path, and high-dimensional feature embedding representations generated by deep convolutional networks extracted from ship visual images. This fusion process forms a coupling of data from different sources in the spatio-temporal dimension through a unified structure description language and semantic label method, providing a comprehensive, quantifiable, and high-dimensional coupled information basis for subsequent prediction models.

[0041] Step S103: Obtain a set of ship response lag factors and a set of bridge structure risk constraints through preprocessing operations based on the multi-source analysis data.

[0042] Specifically, by analyzing and modeling the ship dynamic characteristics and bridge area structure information extracted from the multi-source analysis data, key lag characteristics and structure-sensitive areas that affect the actual response ability of the ship and the accuracy of bridge protection judgment are identified, and a data set for subsequent dynamic modeling and risk calculation is generated in a structured form.

[0043] In a possible implementation manner, step S103 further includes: constructing a set of ship response lag factors according to the current motion state data and environmental disturbance factor data, and the set of ship response lag factors includes rudder effect lag time factors, hydrodynamic interference delay factors, and propulsion system inertia factors; constructing a set of bridge structure risk constraints according to the bridge structure topology data, historical structure response data, and image feature extraction data, and the set of bridge structure risk constraints includes physical boundary constraints of bridge piers, historical high-risk point constraints, and predefined protection area constraints.

[0044] Specifically, according to the current motion state data and environmental disturbance factor data, a set of ship response lag factors is constructed: First, analyze the speed change, course deviation, and acceleration response lag characteristics of the ship within continuous time segments, and combine the influence degrees of wind and current disturbances in the lateral and longitudinal directions to identify the deviation trend between its actual response and control instructions; construct the rudder efficiency lag time factor according to the lag relationship between the ship's rudder angle change rate and the response curve, and extract the liquid flow dynamic interference delay factor by identifying the turning lag and target course deviation trajectory of the ship under the action of cross-flow; at the same time, construct the propulsion system inertia factor through the lag interval length between the propulsion power change and the speed response, and form a three-dimensional dynamic characterization model of the ship response lag factor set. In this application, the three-dimensional dynamic characterization model is used for multi-dimensional dynamic quantification and visualization modeling of the ship response lag factors. This model is not limited to a specific mathematical form or modeling method. Its core lies in constructing the key factors affecting the ship's maneuvering response - including variables such as rudder efficiency lag, propulsion inertia, and flow field disturbance delay - as time functions respectively, and describing their continuous change trends within the future prediction window in a three-dimensional domain through spatial vectors or parameter tensors.

[0045] Meanwhile, according to the bridge structure topology data, historical structure response data, and image feature extraction data, a set of bridge structure risk constraints is constructed: Analyze the pier spatial layout, outer contour boundary, component dimensions, and bridge area obstacle distribution characteristics in the bridge structure model to form the physical boundary constraints of the bridge piers; then combine the collision events, abnormal forces, and vibration response over-limit points recorded in the historical structure response data to extract the high-frequency accident areas and generate the historical high-risk point constraints; further set a buffer zone according to the bridge area protection area annotation layer and expert experience in the image recognition and deep learning results, and define the predefined collision warning range with safety significance to form the predefined protection area constraints. The bridge structure model refers to the data expression form used to describe the spatial form, geometric composition, and structural distribution relationship of the target bridge. This model is based on organizing and expressing the spatial parameters of the key structural units such as the piers, bridge decks, and bridge body connection parts of the target bridge to form a data carrier that can support spatial positioning, boundary recognition, and risk analysis. In this application, the bridge structure model is not limited to a specific data format or generation method, and can be a set of bridge geometric information extracted through sensing devices, surveying systems, engineering design drawings, or structural annotation information. This set should at least include the pier positions, quantities, spatial dimensions, relative layout relationships, and the positions and structural elements of the bridge area passageways, and is used to assist in constructing the structural risk constraint information for interaction with the ship.

[0046] Step S104: Input the multi-source analysis data into the ship trajectory prediction model and output the predicted trajectory path corresponding to the target ship.

[0047] Specifically, by introducing a modeling mechanism for the future motion trend of the ship, path prediction under complex disturbance environments is achieved. This path prediction is based on a comprehensive analysis of the ship's current state, environmental impact factors, and bridge structure constraint conditions, and outputs a continuous motion trajectory covering the future time window to support subsequent dynamic safety boundary calculation and risk intersection judgment.

[0048] In a possible implementation manner, step S104 further includes: constructing a ship trajectory prediction model through a long short-term memory network and a probabilistic state transition diagram structure; based on multi-source analysis data, the ship trajectory prediction model performs continuous integral prediction on the state evolution of the target ship based on a preset future time window to output the predicted trajectory path corresponding to the target ship.

[0049] Specifically, a ship trajectory prediction model is constructed through a long short-term memory network and a probabilistic state transition diagram structure: the long short-term memory network (LSTM) is used to capture the non-linear dynamic characteristics of the ship's motion state in the time dimension, by modeling the long-term and short-term correlations of the velocity vector, acceleration, heading angle, angular velocity, and their change trends, to enhance the forward-looking perception ability of the ship trajectory prediction model for state changes; the probabilistic state transition diagram structure is used to model the uncertain paths guided by disturbances during the transfer of the ship's state from the current state to the future state, constructs a state transition probability field by combining disturbance parameters such as water flow and wind force, and performs multi-path weighted inference on the predicted trajectory, thereby reflecting the uncertainty and variability of the trajectory.

[0050] Based on multi-source analysis data, the ship trajectory prediction model performs continuous integral prediction on the state evolution of the target ship based on a preset future time window to output the predicted trajectory path corresponding to the target ship: the ship trajectory prediction model takes the current motion state and the lag correction term as the initial boundary conditions, and performs state integral calculation for each future time step, and outputs a predicted trajectory path including a position sequence, a velocity sequence, and a path confidence interval; this path not only includes the main path direction of the ship under ideal control, but also integrates the dynamic effects of response hysteresis and external disturbances on path deviation, providing a highly fitting actual input basis for risk boundary assessment. The predicted trajectory path corresponding to the target ship can be output according to the following relational expression:

[0051]

[0052] Among them, represents the predicted position corresponding to the target ship at time t + τ, P(t) is the current position vector corresponding to the target ship at time t, τ is the length of the prediction time window, that is, the time interval range covered by the trajectory prediction from the current time t to the future, which defines the time depth of the predicted path and is used to set the calculation time span of the target ship's future trajectory. represents the velocity vector. represents the acceleration vector, represents, is the water flow disturbance vector, is the wind field disturbance vector, θ(s) is the course angle, is the operator for rotation mapping of the disturbance direction based on the course angle, represents the ship response lag correction vector, s is the integration variable, and s ∈ [t, t + τ], representing any time point during the process of advancing from the current time t to the prediction time τ, and λ1, λ2, and λ3 are all adjustment coefficients.

[0053] The above relational expression generates a continuous evolution path of the target ship trajectory through dynamic integration of various influencing factors within the future time interval, has the ability to model non - linear disturbances and dynamic response lags, and improves the physical consistency and engineering applicability of the path prediction results by embedding state - transfer uncertainties.

[0054] Step S105: Construct a dynamic safety envelope area between the target bridge and the target ship according to the predicted trajectory path and the set of ship response lag factors.

[0055] Specifically, based on the future trajectory of the target ship and its possible path deviation behavior during the prediction process, a dynamic spatial boundary that can cover its actual possible reach range is established for subsequent determination of collision risks.

[0056] In a possible implementation manner, step S105 further includes: constructing multiple confidence intervals corresponding to the position of the target ship at each prediction moment according to the predicted trajectory path and the set of ship response lag factors, one prediction moment corresponding to one confidence interval, and the confidence intervals are modeled as a set of multi - ellipsoids that evolve dynamically with time in three - dimensional space; constructing an interval union according to the multiple confidence intervals, and taking the interval union as the dynamic safety envelope area between the target bridge and the target ship.

[0057] Specifically, the construction of the confidence interval uses the predicted trajectory path as the center line, comprehensively considers the influence of key factors such as rudder effect lag time, propulsion system response inertia, and fluid flow disturbance delay on the possible deviation of the actual position of the ship, quantifies the uncertain range of the ship position at each prediction moment, and determines the confidence interval size using the probability tolerance boundary. The confidence interval is modeled as a set of multi-ellipsoids that dynamically evolve over time in three-dimensional space, that is, a three-dimensional ellipsoid centered on the predicted position is constructed at each prediction time point. The lengths of the ellipsoid axes are determined by the distribution characteristics of the current velocity direction, the estimated offset in the direction of control response hysteresis, and the degree of influence of the external disturbance field on the attitude stability, so that the shapes and directions of the ellipsoids at different time points can reflect the dynamic changes in the actual response ability. An interval union is constructed based on multiple confidence intervals to form a temporally continuous spatially enclosed area. This union covers all the path deviation ranges that the target ship may cause due to control deviation, environmental disturbance, and lag effect within the entire prediction time window. Finally, this interval union is used as the dynamic safety envelope area between the target bridge and the target ship, which is used to represent all the dynamic risk space boundaries where the ship may approach or contact the bridge structure in the future, providing a complete, continuous, and dynamically evolvable spatial expression basis for subsequent intersection calculations with the bridge structure risk constraint set.

[0058] Step S106, obtain the intersection of the dynamic safety envelope area and the bridge structure risk constraint set during the relative movement of the target ship and the target bridge.

[0059] Specifically, through a spatial interaction analysis algorithm, a spatio-temporal matching operation is performed on the safety envelope area that dynamically evolves in the time series and the bridge structure risk constraint set to identify the spatial overlap of the potentially high-risk areas where the target ship may touch the bridge structure during its future movement. The dynamic safety envelope area is composed of multiple three-dimensional confidence ellipsoids at different times, and its overall can be represented as a closed volume path within the prediction time window; the bridge structure risk constraint set includes physical boundary constraints of bridge piers, historical high-risk point constraints, and predefined protection area constraints. Each type of constraint object is represented by a three-dimensional geometric model and has spatial attributes that can be projected onto a unified coordinate system. During the calculation process, the dynamic safety envelope area is decomposed into multiple static sub-blocks according to time slices, and a Boolean intersection operation is performed with each bridge risk boundary element to extract the spatial intersection at each prediction moment; at the same time, the geometric overlap volume of each intersection, the shortest distance between the intersection center and the pier center, the intersection duration, and the spatial weight position of the intersection position in the bridge area structure (such as whether it is in a historical high-risk area) are extracted to form a complete set of target intersection data, which is used to characterize the dynamic evolution trend of the target ship that may cause structural contact or risk approximation in the future period.

[0060] Step S107: Calculate the risk level of a collision between the target ship and the target bridge based on the target intersection, and send a collision prevention warning prompt to the target ship according to the risk level.

[0061] Specifically, quantify the potential collision trend reflected by the dynamic space intersection into an identifiable risk level, and drive the early warning trigger mechanism accordingly to achieve the intelligent response of the system.

[0062] In a possible implementation manner, step S107 further includes: obtaining the intersection parameters of the dynamic safety envelope area and the bridge structure risk constraint set during the relative movement through spatial overlap analysis, where the intersection parameters include the intersection area parameter, the intersection duration parameter, and the intersection position parameter; constructing a multi-dimensional risk assessment vector according to the intersection parameters, and inputting the multi-dimensional risk assessment vector into a weighted fuzzy inference model; calculating the risk level index value of a collision between the target ship and the target bridge according to the weighted fuzzy inference model; if it is confirmed that the risk level index value is greater than the preset risk level threshold, it is confirmed that there is a collision risk between the target ship and the target bridge, and a collision prevention warning prompt is sent to the target ship in a preset manner.

[0063] Specifically, the intersection parameters include an intersection area parameter, which is used to characterize the degree of spatial overlap between the envelope area and the structure boundary in three-dimensional space; an intersection duration parameter, which is used to describe the duration of the spatial overlap within the predicted time window; and an intersection position parameter, which is used to identify the specific spatial position of the intersection point on the bridge structure and label whether it is in a high-risk area, a key structural part, or a sensitive protection boundary. Based on the above intersection parameters, a multi-dimensional risk assessment vector is constructed. The vector includes, but is not limited to, the intersection area ratio, the intersection duration, the shortest distance from the intersection point to the nearest point of the pier, the structural risk weight value at the center position of the intersection, the angle between the target heading and the intersection direction, etc., which are used to comprehensively reflect the spatial characteristics and dynamic evolution characteristics of the possible future collision trend. The multi-dimensional risk assessment vector is input into a weighted fuzzy inference model. The model pre-establishes a multi-input fuzzy rule base and outputs a continuous risk level index value according to the fuzzy relationship and empirical weight between the input parameters. Calculate the risk level index value of the collision between the target ship and the target bridge according to the weighted fuzzy inference model, and compare it with the preset risk level threshold. If it is confirmed that the risk level index value is greater than the threshold, it is determined that the target ship has an actual risk of physical contact with the target bridge or crossing a highly sensitive area within the predicted time window, and the system then confirms the existence of a collision risk. Under the condition of confirming the collision risk, the system sends an anti-collision warning prompt to the target ship in a preset manner. The warning prompt may include real-time instruction push based on a communication link (such as VHF, AIS, or cellular network), voice broadcast, sound and light signals, or graphical path deviation warnings; at the same time, the warning information may include recommended control actions (such as deceleration, course change, stop propulsion) or risk level explanation text, and is sent to the bridge area management platform synchronously to drive linkage response measures. This implementation method enhances the flexibility and forward-looking of risk judgment through a fuzzy logic model based on the intersection spatial characteristics, and effectively improves the intelligent level and response adaptability of the warning system.

[0064] By adopting the above method, this application obtains the visual perception data corresponding to the target ship and the bridge structure data corresponding to the target bridge; constructs multi-source analysis data based on the visual perception data and the bridge structure data; obtains the ship response lag factor set and the bridge structure risk constraint set through preprocessing operations according to the multi-source analysis data; inputs the multi-source analysis data into the ship trajectory prediction model and outputs the predicted trajectory path corresponding to the target ship; constructs the dynamic safety envelope area between the target bridge and the target ship according to the predicted trajectory path and the ship response lag factor set; obtains the target intersection of the dynamic safety envelope area and the bridge structure risk constraint set during the relative movement of the target ship and the target bridge; calculates the risk level of collision between the target ship and the target bridge based on the target intersection, and sends a collision prevention warning prompt to the target ship according to the risk level. By integrating visual perception and bridge structure data and constructing a multi-source dynamic analysis model, high-precision prediction and intelligent evaluation of the future state evolution of the ship and the collision risk in the bridge area are realized, so the perception accuracy of the collision prevention warning system in a complex environment is greatly improved, and the frequency of false alarms or missed alarms is reduced.

[0065] Please refer to Figure 2 , which shows a module schematic diagram of a bridge anti-collision warning device provided by an embodiment of this application. The device includes an acquisition module 21 and a processing module 22. Among them,

[0066] The acquisition module 21 is used to obtain the visual perception data corresponding to the target ship and the bridge structure data corresponding to the target bridge; construct multi-source analysis data based on the visual perception data and the bridge structure data; obtain the ship response lag factor set and the bridge structure risk constraint set through preprocessing operations according to the multi-source analysis data.

[0067] The processing module 22 is used to input the multi-source analysis data into the ship trajectory prediction model and output the predicted trajectory path corresponding to the target ship; construct the dynamic safety envelope area between the target bridge and the target ship according to the predicted trajectory path and the ship response lag factor set; obtain the target intersection of the dynamic safety envelope area and the bridge structure risk constraint set during the relative movement of the target ship and the target bridge; calculate the risk level of collision between the target ship and the target bridge based on the target intersection, and send a collision prevention warning prompt to the target ship according to the risk level.

[0068] In a possible implementation manner, the acquisition module 21 is used to construct multi-source analysis data according to the visual perception data and the bridge structure data, specifically including: registering and fusing the visual perception data and the bridge structure data in the same time domain and coordinate system to construct multi-source analysis data, and the multi-source analysis data includes current motion state data, environmental disturbance factor data, bridge structure topology data, historical structure response data, and image feature extraction data.

[0069] In a possible implementation, the acquisition module 21 is configured to obtain a set of ship response lag factors and a set of bridge structure risk constraints through preprocessing operations based on multi-source analysis data, specifically including: constructing a set of ship response lag factors according to the current motion state data and environmental disturbance factor data, where the set of ship response lag factors includes the rudder effect lag time factor, the hydrodynamic interference delay factor, and the propulsion system inertia factor; constructing a set of bridge structure risk constraints according to the bridge structure topology data, historical structure response data, and image feature extraction data, where the set of bridge structure risk constraints includes the physical boundary constraints of bridge piers, the constraints of historical high-risk points, and the predefined protection area constraints.

[0070] In a possible implementation, the processing module 22 is configured to input the multi-source analysis data into a ship trajectory prediction model and output a predicted trajectory path corresponding to the target ship, specifically including: constructing a ship trajectory prediction model through a long short-term memory network and a probabilistic state transition diagram structure; based on the multi-source analysis data, the ship trajectory prediction model performs continuous integral prediction on the state evolution of the target ship over a future preset time window to output a predicted trajectory path corresponding to the target ship.

[0071] In a possible implementation, the processing module 22 is configured to perform continuous integral prediction on the state evolution of the target ship over a future preset time window to output a predicted trajectory path corresponding to the target ship, specifically including: outputting a predicted trajectory path corresponding to the target ship according to the following relational expression:

[0072]

[0073] where represents the predicted position of the target ship at time t + τ, P(t) is the current position vector of the target ship at time t, τ is the length of the prediction time window, represents the velocity vector, represents the acceleration vector, represents, is the water flow disturbance vector, is the wind field disturbance vector, θ(s) is the heading angle, is the operator for rotating and mapping the disturbance direction based on the heading angle, represents the ship response lag correction vector, s is the integration variable, representing any time point during the process of advancing from the current time t to the prediction time τ, and λ1, λ2, and λ3 are all adjustment coefficients.

[0074] In a possible implementation, the processing module 22 is configured to construct a dynamic safety envelope area between the target bridge and the target ship according to the predicted trajectory path and the set of ship response lag factors, specifically including: constructing, according to the predicted trajectory path and the set of ship response lag factors, a plurality of confidence intervals corresponding to the position of the target ship at each predicted moment, one predicted moment corresponding to one confidence interval, and the confidence interval being modeled as a set of multi-ellipsoids that dynamically evolve over time in three-dimensional space; constructing an interval union according to the plurality of confidence intervals, and using the interval union as the dynamic safety envelope area between the target bridge and the target ship.

[0075] In a possible implementation, the processing module 22 is configured to calculate the risk level of a collision between the target ship and the target bridge based on the target intersection, and send a collision avoidance warning prompt to the target ship according to the risk level, specifically including: obtaining, through spatial overlap analysis, the intersection parameters of the dynamic safety envelope area and the bridge structure risk constraint set during the relative movement, the intersection parameters including the intersection area parameter, the intersection duration parameter, and the intersection position parameter; constructing a multi-dimensional risk assessment vector according to the intersection parameters, and inputting the multi-dimensional risk assessment vector into a weighted fuzzy inference model; calculating the risk level index value of a collision between the target ship and the target bridge according to the weighted fuzzy inference model; if it is confirmed that the risk level index value is greater than the preset risk level threshold, it is confirmed that there is a collision risk between the target ship and the target bridge, and a collision avoidance warning prompt is sent to the target ship according to the preset method.

[0076] It should be noted that: when the device provided in the above embodiment realizes its functions, only the above-mentioned division of each functional module is used for illustration. In actual applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process is detailed in the method embodiments, which will not be repeated here.

[0077] This application also provides an electronic device. Refer to Figure 3 , Figure 3 is a schematic structural diagram of an electronic device provided in an embodiment of this application. The electronic device may include: at least one processor 301, at least one communication bus 302, a user interface 303, at least one network interface 304, and a memory 305.

[0078] Among them, the communication bus 302 is used to realize the connection and communication between these components.

[0079] Among them, the user interface 303 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 303 may further include a standard wired interface and a wireless interface.

[0080] Among them, the network interface 304 may optionally include standard wired interfaces, wireless interfaces (such as WI-FI interfaces).

[0081] Among them, the processor 301 may include one or more processing cores. The processor 301 uses various interfaces and lines to connect various parts within the entire server. By running or executing instructions, programs, code sets, or instruction sets stored in the memory 305, and by calling the data stored in the memory 305, it executes various functions of the server and processes data. Optionally, the processor 301 may be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), or programmable logic array (PLA). The processor 301 may integrate one or a combination of several of a central processing unit (CPU), a graphics processing unit (GPU), and a modem, etc. Among them, the CPU mainly processes the operating system, user interface, application programs, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communication. It can be understood that the above-mentioned modem may not be integrated into the processor 301 and may be implemented separately by a single chip.

[0082] Among them, the memory 305 may include random access memory (RAM) and may also include read-only memory. Optionally, the memory 305 includes a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 305 may include a program storage area and a data storage area. Among them, the program storage area may store instructions for implementing the operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store the data involved in the above-mentioned various method embodiments. The memory 305 may optionally also be at least one storage device located far from the aforementioned processor 301. Refer to Figure 3 , the memory 305, as a computer storage medium, may include an operating system, a network communication module, a user interface module, and a bridge anti-collision warning application program.

[0083] In Figure 3 In the electronic device shown, the user interface 303 is mainly used to provide an interface for the user to input and obtain the data input by the user. The processor 301 can be used to call the anti-collision warning application program for large bridges stored in the memory 305. When executed by one or more processors 301, the electronic device is caused to execute one or more of the methods as described in the above embodiments. It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that this application is not limited by the described action sequence, because according to this application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0084] This application also provides a computer-readable storage medium storing instructions. When executed by one or more processors, the electronic device is caused to execute one or more of the methods as described in the above embodiments.

[0085] In the above embodiments, the descriptions of the various embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0086] In several implementation manners provided by this application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is only a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some service interfaces. The indirect couplings or communication connections of the devices or units can be in electrical or other forms.

[0087] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0088] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0089] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present application. The aforementioned memory includes: various media such as USB flash drives, external hard drives, magnetic disks, or optical discs that can store program codes.

[0090] The above are only exemplary embodiments disclosed in the present application and should not be used to limit the scope of the disclosure of the present application. That is, any equivalent changes and modifications made in accordance with the teachings disclosed in the present application still fall within the scope covered by the disclosure of the present application.

[0091] The present application aims to cover any variations, uses, or adaptive changes of the present application, which follow the general principles disclosed in the present application and include common general knowledge or conventional technical means in the technical field not recorded in the present application.

Claims

1. A method for anti-collision warning of a large bridge, characterized in that, The method includes: Obtaining visual perception data corresponding to a target ship and obtaining bridge structure data corresponding to a target bridge; Constructing multi-source analysis data according to the visual perception data and the bridge structure data; According to the multi-source analysis data, obtaining a set of ship response lag factors and a set of bridge structure risk constraints through preprocessing operations; Inputting the multi-source analysis data into a ship trajectory prediction model and outputting a predicted trajectory path corresponding to the target ship; Constructing a dynamic safety envelope area between the target bridge and the target ship according to the predicted trajectory path and the set of ship response lag factors; Obtaining an intersection of the dynamic safety envelope area and the set of bridge structure risk constraints during the relative movement between the target ship and the target bridge; Based on the intersection, calculating a risk level of a collision between the target ship and the target bridge, and sending an anti-collision warning prompt to the target ship according to the risk level.

2. The method according to claim 1, wherein The constructing of the multi-source analysis data according to the visual perception data and the bridge structure data specifically includes: Registering and fusing the visual perception data and the bridge structure data in the same time domain and coordinate system to construct the multi-source analysis data, where the multi-source analysis data includes current motion state data, environmental disturbance factor data, bridge structure topology data, historical structure response data, and image feature extraction data.

3. The method according to claim 2, wherein The obtaining of the set of ship response lag factors and the set of bridge structure risk constraints through preprocessing operations according to the multi-source analysis data specifically includes: Constructing the set of ship response lag factors according to the current motion state data and the environmental disturbance factor data, where the set of ship response lag factors includes a rudder effect lag time factor, a fluid dynamic interference delay factor, and a propulsion system inertia factor; Constructing the set of bridge structure risk constraints according to the bridge structure topology data, the historical structure response data, and the image feature extraction data, where the set of bridge structure risk constraints includes a physical boundary constraint of a bridge pier, a historical high-risk point constraint, and a predefined protection area constraint.

4. The method according to claim 1, wherein The inputting of the multi-source analysis data into the ship trajectory prediction model and the outputting of the predicted trajectory path corresponding to the target ship specifically includes: Constructing the ship trajectory prediction model through a long short-term memory network and a probabilistic state transition diagram structure; According to the multi-source analysis data, the ship trajectory prediction model performs continuous integral prediction on the state evolution of the target ship based on a future preset time window to output the predicted trajectory path corresponding to the target ship.

5. The method according to claim 4, characterized in that, The performing of continuous integral prediction on the state evolution of the target ship based on a future preset time window to output the predicted trajectory path corresponding to the target ship specifically includes: Outputting the predicted trajectory path corresponding to the target ship according to the following relational expression: Among them, represents the predicted position corresponding to the target ship at time \(t + \tau\), \(P(t)\) is the current position vector corresponding to the target ship at time \(t\), and \(\tau\) is the length of the prediction time window. represents the velocity vector, represents the acceleration vector, represents, is the water flow disturbance vector, is the wind field disturbance vector, \(\theta(s)\) is the heading angle, is the operator for rotating and mapping the disturbance direction based on the heading angle, represents the ship response lag correction vector, \(s\) is the integration variable, representing any time point during the process of advancing from the current time \(t\) to the prediction time \(\tau\), and \(\lambda_1\), \(\lambda_2\), and \(\lambda_3\) are all adjustment coefficients.

6. The method according to claim 1, wherein The constructing of the dynamic safety envelope area between the target bridge and the target ship according to the predicted trajectory path and the set of ship response lag factors specifically includes: Based on the predicted trajectory path and the set of ship response lag factors, multiple confidence intervals corresponding to the position of the target ship at each prediction moment are constructed. One prediction moment corresponds to one confidence interval, and the confidence interval is modeled as a set of multi-ellipsoids that dynamically evolve over time in three-dimensional space; An interval union is constructed based on the multiple confidence intervals, and the interval union is used as the dynamic safety envelope area between the target bridge and the target ship.

7. The method according to claim 1, characterized in that, Based on the target intersection, calculating the risk level of a collision between the target ship and the target bridge, and sending a collision avoidance warning prompt to the target ship according to the risk level, specifically including: Through spatial overlap analysis, obtaining the intersection parameters of the dynamic safety envelope area and the bridge structure risk constraint set during the relative movement process, where the intersection parameters include intersection area parameters, intersection duration parameters, and intersection position parameters; Constructing a multi-dimensional risk assessment vector according to the intersection parameters, and inputting the multi-dimensional risk assessment vector into a weighted fuzzy inference model; Calculating the risk level index value of a collision between the target ship and the target bridge according to the weighted fuzzy inference model; If it is confirmed that the risk level index value is greater than a preset risk level threshold, it is confirmed that there is a collision risk between the target ship and the target bridge, and the collision avoidance warning prompt is sent to the target ship in a preset manner.

8. An anti-collision warning device for a large bridge, characterized in that, The device includes an acquisition module and a processing module, where, The acquisition module is used to acquire visual perception data corresponding to the target ship, and acquire bridge structure data corresponding to the target bridge; construct multi-source analysis data according to the visual perception data and the bridge structure data; and obtain the set of ship response lag factors and the bridge structure risk constraint set through preprocessing operations according to the multi-source analysis data; The processing module is used to input the multi-source analysis data into a ship trajectory prediction model and output the predicted trajectory path corresponding to the target ship; construct the dynamic safety envelope area between the target bridge and the target ship according to the predicted trajectory path and the set of ship response lag factors; obtain the target intersection of the dynamic safety envelope area and the bridge structure risk constraint set during the relative movement process between the target ship and the target bridge; calculate the risk level of a collision between the target ship and the target bridge based on the target intersection, and send a collision avoidance warning prompt to the target ship according to the risk level.

9. An electronic device, characterized in that, It includes a processor, a communication bus, a user interface, a network interface, and a memory. The memory is used to store instructions. The user interface and the network interface are used to communicate with other devices. The processor is used to execute the instructions stored in the memory so that the electronic device executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 7 is executed.

Images