A method, system, and application for generating heatmaps of ship course, speed, and collision risk based on multi-source fusion.

Abstract

This invention discloses a method, system, and application for generating collision risk heatmaps for ship course and speed based on multi-source fusion. The method first fuses AIS, radar, and electronic chart data to acquire dynamic and static obstacle information and calculates a multi-dimensional collision risk parameter set, including minimum encounter distance, minimum encounter time, distance past the bow, and time past the bow. Then, based on safety thresholds, a set of dangerous course profiles for each target is constructed and fused to obtain a total set of dangerous course profiles across all space. Next, parameters characterizing ship maneuverability are introduced to construct a dynamic risk field model based on a ring Gaussian density function. The cumulative probability across dangerous course intervals is calculated as a risk index and extended to the speed dimension to generate a set of spatiotemporal risk coefficients. Finally, an omnidirectional dynamic risk heatmap with course and speed as coordinate axes is generated. This invention provides ships with intuitive, comprehensive, and dynamic navigation risk situation assessment and decision support.

Description

Technical Field

[0001] This invention relates to the fields of traffic information engineering and control, and navigation technology and shipbuilding engineering, and particularly to a method, system and application for generating heat maps of ship course, speed and collision risk based on multi-source fusion. Background Technology

[0002] Ship collision hazard assessment is a core technology for ensuring navigational safety. Existing assessment methods have the following limitations, making it difficult to meet the urgent need for comprehensive, dynamic, and interpretable risk perception in complex environments:

[0003] (1) Traditional assessment method based on index weighting: This method calculates parameters such as minimum encounter distance (DCPA) and minimum encounter time (TCPA) and integrates them using subjective or objective weighting methods to obtain the risk index of a single target relative to the ship. Its fundamental flaw is that it is limited to "point-to-point" assessment of a single target and cannot form a unified and continuous full-space risk situation awareness in complex scenarios with multiple ships and static obstructions, which easily leads to fragmented decision-making.

[0004] (2) Improved method by introducing the concept of ship domain: This method corrects the risk by judging whether the target enters the pre-set "safety domain" around the ship. Although the concept of space is introduced, its risk assessment still relies on the weighted fusion of the aforementioned discrete indicators. It fails to fundamentally break through the framework of multi-objective global risk assessment, and the fixed nature of the "ship domain" model makes it difficult to adapt to different ship types and dynamic maneuvering intentions.

[0005] (3) Neural Network-Based Black Box Prediction Method: This method uses data-driven end-to-end risk prediction. Although the algorithm is advanced, its internal decision-making logic lacks physical interpretability and is a "black box" model. This makes it difficult for drivers to understand and trust its output results, and its generalization ability is questionable in extreme scenarios where the training data is not covered, resulting in insufficient reliability for engineering applications.

[0006] (4) Environmental density inference method based on AIS data: This method uses historical AIS data to statistically analyze the vessel density in the waterway and infer background risk. It mainly relies on a single, non-real-time AIS data source, which has problems such as data update delays and omissions in reporting some vessels (such as fishing vessels without AIS). More importantly, it completely ignores the real-time dynamic motion characteristics and maneuvering intentions of the vessel and other vessels, and cannot provide dynamic guidance for current collision avoidance decisions. In essence, it is a static, statistical macro-risk assessment.

[0007] In summary, existing technologies either limit themselves to a single objective in risk assessment, have methodological flaws, or rely on one-sided or outdated data sources. They generally fail to achieve the core objective of dynamically, intuitively, and interpretably integrating multi-source real-time sensing information, ship maneuvering characteristics, and course, speed, and risk status. Therefore, a novel solution that overcomes these shortcomings is urgently needed. Summary of the Invention

[0008] The purpose of this invention is to address the shortcomings of existing technologies, such as strong subjectivity, insufficient all-space perception, neglect of ship maneuvering characteristics, and reliance on a single data source. It provides a method, system, and application for generating ship course, speed, and collision risk heat maps based on multi-source fusion. By constructing a set of dangerous courses and speeds and using a density Gaussian function to establish a risk field model, it achieves multi-dimensional risk quantification and visualization, providing scientific support for pilots, autonomous navigation systems, and shore-based managers, and improving navigation safety and intelligence.

[0009] The technical solution adopted in this invention is:

[0010] A method for generating collision risk heatmaps for ships based on multi-source fusion includes the following steps:

[0011] S1: Acquire perception data of the surrounding environment of the ship to construct a dataset of all moving and static obstacles. The perception data includes dynamic information of other ships from the Automatic Identification System (AIS), target detection information from the radar system, and static obstruction information from the Electronic Chart (ENC).

[0012] S2: Calculate the collision risk parameters of each obstacle relative to the ship to construct a collision risk dataset. The collision risk parameters include at least the minimum encounter distance (DCPA), minimum encounter time (TCPA), bow distance (BCR), and bow time (BCT).

[0013] S3: Construct a hazard assessment function to determine the risk of collision on any trial course of the vessel based on a set of collision risk parameters and a preset safety threshold. Next, determine whether there is a collision risk between the vessel and each target, and construct a set of dangerous course directions for each target. ;

[0014] S4: Merge all sets This yields the total set of dangerous routes C and the total set of non-dangerous routes for the ship across all space. ;

[0015] S5: Extend the risk index to the ship's speed dimension by traversing the allowed speed range. Generate an omnidirectional hazard set in the spatiotemporal dimension Total collection of dangers ;

[0016] S6: Traverse the total set of dangers The speed of the ship was determined. Omnidirectional Danger Set Furthermore, ship maneuvering performance parameters are introduced to construct a dynamic risk field model based on a ring Gaussian density function, and the ship's speed is calculated. Omnidirectional Danger Set The cumulative probability above serves as the risk index set for the ship under the current course. Based on this, the complete set of dangers is traversed. Determine the set of speed and heading risk indices ;

[0017] S7: Set of speed and heading risk indices Generate an omnidirectional dynamic risk heat map with heading and speed as coordinate axes.

[0018] Furthermore, the implementation of obtaining the perception data of the surrounding environment of the ship in step S1 also includes: virtualizing the channel boundary and coastline, emitting rays with an interval of α centered on the ship, using the intersection of the rays with the channel boundary or coastline as virtual static obstacles, and incorporating the position information of the virtual static obstacles into the static obstruction information.

[0019] Specifically, preliminary processing is performed based on the perceived data, including using virtualization to process some obstacles, and further constructing a dataset of all moving and static obstacles;

[0020] Furthermore, in step S2, the closest encounter distance between the ship and the target ship (obstacle) The calculation formula is as follows:

[0021] ;

[0022] The most recent encounter time between this vessel and the target vessel (obstacle). The calculation formula is as follows:

[0023] ;

[0024] in, This is the true bearing of the target ship relative to our own ship. Indicates the distance between the target ship and this ship. Indicates the direction of the target's relative motion vector. This represents the relative motion vector.

[0025] The formula for calculating the distance (BCR) between the target vessel and the bow of the current vessel is as follows:

[0026] ;

[0027] The formula for calculating the time it takes for the target vessel to pass the bow of the current vessel (BCT) is as follows:

[0028] ;

[0029] in, The minimum encounter distance (DCPA) is given, θ is the angle between the bearing of the target ship and the direction of relative velocity, and β is the relative bearing of the target ship.

[0030] Furthermore, the danger determination function in step S3 for:

[0031] ;

[0032] in, For safe distance threshold, For time threshold; when =1 indicates that the course is being tested. There is a risk of collision below; The closest encounter distance between the ship and the target ship (obstacle) on the i-th heading; The closest encounter time between the ship and the target ship (obstacle) on the i-th heading; This is the closest encounter distance between the ship and the target ship (obstacle) on the i-th heading; The distance between the target ship's bow and the ship's bow at the i-th heading; Let be the time when the target ship passes the bow of our ship on the i-th heading.

[0033] Furthermore, integrate the dangerous course sets of all targets. This yields the total set of dangerous routes C and the total set of non-dangerous routes for the ship across all space. Specifically, the set of course directions that pose a risk or no risk to this vessel from the aforementioned target vessel is as follows:

[0034] ; ;

[0035] in, For the first Dangerous course set for the target vessel For this ship relative to the first The target ship's course; For the first The set of non-dangerous course directions for the target ship. Therefore, the relative total set of dangerous course directions for the ship across all space is:

[0036] ;

[0037] in, The target number of ships; This represents the total set of hazardous course directions for the ship across all space. The above set represents the hazardous course directions of obstacles at the ship's current position and speed. If the speed dimension is considered, then the total hazardous course directions for both speed and course are constructed as follows:

[0038] ;

[0039] In the formula, This indicates that the ship is assembled in all directions for danger. This indicates the minimum speed permitted for this vessel. This indicates the maximum speed permitted for this vessel. This is the total set of hazards that takes into account both the speed and heading dimensions.

[0040] Furthermore, traverse the total set of dangers. The speed of the ship was determined. Omnidirectional Danger Set By introducing ship maneuvering performance parameters, a dynamic risk field model based on a ring Gaussian density function is constructed to calculate the risk field of the ship at speeds... Omnidirectional Danger Set The cumulative probability on the course of the ship is used as the ship's trajectory. Risk index set below The density function expression is as follows:

[0041] ;

[0042] in, Indicates the course of the ships being traversed. Indicates the current number of the ship One course, ; This represents the angular span that controls the width of the distribution, indicating the angle the ship can currently turn. A ship with strong turning ability will have a larger value, and vice versa. This value is related to the ship type and length. As an optional value, it can be set to... ; This represents the phase term of the cyclic Gaussian summation.

[0043] Furthermore, the risk index By calculating the dangerous course section [ , The integral of the upper ring Gaussian density function is obtained, where , These are the total sets of dangerous headings. The left and right boundaries;

[0044] When the interval does not cross 0°, the risk index The calculation formula is:

[0045] .

[0046] When the interval crosses 0°, the risk index The calculation formula is:

[0047] ;

[0048] in, Indicates that the current course of the ship is Risk index under certain circumstances This is the angular span, which is positively correlated with the ship's turning performance. If the ship has good turning performance, this value will be larger. For the ship at speed Omnidirectional Danger Set Left boundary, For the ship at speed Omnidirectional Danger Set The right boundary. Indicates the course of the ships being traversed. This represents the phase term of the cyclic Gaussian summation; This indicates the current course of the vessel. The cumulative distribution function (CDF) of the standard normal distribution is given by the following formula:

[0049] ;

[0050] ;

[0051] In the formula, This is the error function.

[0052] The set of expressions for the risk factors at different speeds in step S6 is as follows:

[0053] Furthermore, the set of expressions for the risk factors at different speeds in step S6 is as follows:

[0054] ;

[0055] in, For the current course of the ship and speed A set of risk indices under a specific dimension.

[0056] Furthermore, based on the risk index set Generate an omnidirectional dynamic risk heat map with heading and speed as coordinate axes. That is, by using the principles of thermodynamic mapping, inputting this multi-dimensional data, a heat map of ship collision hazards is finally generated.

[0057] This invention also discloses a multi-source fusion-based system for generating heatmaps of ship course and speed collision risks, comprising:

[0058] The multi-source data fusion and processing module is used to acquire and fuse AIS, radar and electronic chart data, and calculate the spatial state parameters and collision risk parameter set of dynamic and static obstacles;

[0059] The hazard set construction module, connected to the multi-source data fusion and processing module, is used to construct a hazard course set for each target based on the hazard determination function and fuse them to generate a total hazard course set for the entire space.

[0060] The dynamic risk field calculation module, connected to the hazard set construction module, is used to receive ship maneuvering performance parameters, calculate the risk index based on the ring Gaussian density function model, and extend it to the speed dimension to generate a set of risk coefficients in the spatiotemporal dimension.

[0061] The risk heatmap generation and output module, connected to the dynamic risk field calculation module, is used to render the set of risk coefficients in the spatiotemporal dimension into an omnidirectional dynamic risk heatmap and output it.

[0062] The present invention also discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the program, it implements the steps of the method for generating a collision risk heat map of a ship's course and speed based on multi-source fusion.

[0063] The present invention also discloses a computer-readable storage medium storing a computer program thereon, characterized in that the computer program, when executed by a processor, implements the steps of the method for generating a collision risk heat map of a ship's course and speed based on multi-source fusion.

[0064] The present invention adopts the above technical solutions and has the following beneficial effects compared with the prior art: (1) The present invention integrates multi-source data such as radar, AIS, and electronic charts, and constructs a risk field covering the entire course and speed. It can simultaneously process the risk contribution of all dynamic and static obstacles around it and generate a continuous and unified full-space risk heat map, which fundamentally solves the problem of fragmented risk cognition in multi-target scenarios and provides the driver with a global and intuitive situational awareness. (2) The present invention abandons subjective weighting and innovatively introduces the parameters of bow distance (BCR) and bow time (BCT), and constructs a multi-dimensional feature set in combination with the classic DCPA and TCPA. The risk mapping is carried out using a unified mathematical model based on the ring Gaussian density function, which avoids the evaluation bias caused by the coupling of indicators and makes the evaluation results closer to the nautical physical intuition. The model is transparent and highly interpretable. (3) The present invention introduces the ship's maneuverability (such as turning ability) as a key parameter (σ) into the Gaussian risk field model, so that the generated risk heat map can dynamically respond to the ship's maneuverability. This means that the "safe space" assessed by the system changes dynamically with the characteristics of the ship, providing personalized risk assessments for different ship types and significantly enhancing the engineering applicability of the method. (4) The risk heat map is dynamically updated based on real-time data, identifying safe course / speed ranges and high-risk areas. This enables the driver to proactively plan collision avoidance strategies, rather than just reacting to the current danger. At the same time, the intuitive visualization output greatly reduces the difficulty of human interpretation of complex data and the probability of misjudgment, simplifying the operation process. Attached Figure Description

[0065] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments;

[0066] Figure 1 This is a schematic diagram illustrating the principle architecture of the ship course and speed collision risk heat map generation method based on multi-source fusion of the present invention.

[0067] Figure 2 This is a flowchart illustrating the method for generating collision risk heatmaps of ships based on multi-source fusion according to the present invention.

[0068] Figure 3 This is a schematic diagram of the virtual obstacle processing process of the present invention;

[0069] Figure 4 This is a schematic diagram of the environment in which the ship and the target ship are located in an embodiment of the present invention;

[0070] Figure 5 This is a schematic diagram of the collision risk heat map generated by the present invention under the ship's course and speed. Detailed Implementation

[0071] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings.

[0072] like Figures 1 to 5 As shown in any of the accompanying drawings, this invention discloses a method for generating a ship's heading and speed collision risk heat map based on multi-source fusion, comprising the following steps:

[0073] S1: Acquire perception data of the surrounding environment of the ship to construct a dataset of all moving and static obstacles. The perception data includes dynamic information of other ships from the Automatic Identification System (AIS), target detection information from the radar system, and static obstruction information from the Electronic Chart (ENC).

[0074] Furthermore, in step S1 above, this invention utilizes as much obstacle information data as possible, such as the four conventional core devices to acquire dynamic obstacle data. Various sensing devices constitute the ship's "eyes" and "ears." AIS (Automatic Identification System) can directly obtain the precise latitude and longitude, speed on land (SOG), heading on land (COG), and rate of turn of surrounding ships through message analysis. The data is accurate, but it relies on the normal operation and active activation of other ships' equipment (there is a risk of manual shutdown or equipment malfunction). Electronic Chart System (ECDIS / ECS) mainly provides static obstacle information (such as coastlines, shoals, reefs, shipwrecks, bridges, drilling platforms, etc.), which is stored in a chart database. It provides geographical reference, but chart data updates may lag behind actual geographical changes. Radar system can detect echoes regardless of whether other ships have AIS activated, obtaining the target's distance, bearing, CPA (closest point of encounter), TCPA (time to the nearest point of encounter), relative heading, and relative speed. Active detection is unaffected by cooperation from other vessels, but may be subject to clutter interference in adverse weather conditions (rain, snow, high waves), and has limited detection capability for small objects. Obstacle sensing system: This complements the above systems and typically refers to visual sensors (visible light / infrared cameras and LiDAR). It provides high-resolution local environmental information. Cameras can identify object types (such as buoys, small boats, and personnel), while LiDAR can acquire high-precision 3D point cloud data, accurately measuring the contours and distances of nearby obstacles. It compensates for the shortcomings of radar in near-field blind spots and small object detection, but is significantly affected by weather conditions (such as the impact of heavy fog on LiDAR and visible light).

[0075] Furthermore, as an optional improved implementation, the acquisition of perception data of the ship's surrounding environment in step S1 of the present invention further includes: virtualizing the channel boundary and coastline, emitting rays with an interval of α centered on the ship, using the intersection of the rays with the channel boundary or coastline as virtual static obstacles, and incorporating the position information of the virtual static obstacles into the static obstruction information. Specifically, using the ship as the center, finding the vertical point from the ship to the channel or coastline, such as... Figure 3 The zero point is shown, then forward and backward at intervals. Angle-emitting rays are emitted, and the intersections of the rays with the channel boundary line or coastline are -2, -1, 1, 2... These points are virtual static obstacles, which are added to the static obstacle list, and then spatial state parameters are calculated.

[0076] S2: Calculate the collision risk parameters of each obstacle relative to the ship to construct a collision risk dataset. The collision risk parameters include at least the minimum encounter distance (DCPA), minimum encounter time (TCPA), bow distance (BCR), and bow time (BCT).

[0077] Specifically, in step S2 above, the acquired obstacle position, speed, and heading data are further calculated to obtain key relative spatial parameters. Among these key relative spatial parameters is the closest encounter distance between the ship and the target ship (obstacle). The calculation formula is as follows:

[0078] ;

[0079] The most recent encounter time between this vessel and the target vessel (obstacle). The calculation formula is as follows:

[0080] ;

[0081] in, This is the true bearing of the target ship relative to our own ship. Indicates the distance between the target ship and this ship. Indicates the direction of the target's relative motion vector. This represents the relative motion vector.

[0082] Therefore, once a ship's position is relatively fixed, and the speed and direction of other ships are known, the nearest encounter distance is... and the most recent meeting time Since the distance between the target ship's bow and the target ship's bow (BCR) is determined based on the ship's speed and course, the formula for calculating this distance is as follows:

[0083] ;

[0084] The formula for calculating the time it takes for the target vessel to pass the bow of the current vessel (BCT) is as follows:

[0085] ;

[0086] in, The minimum encounter distance (DCPA) is given, θ is the angle between the bearing of the target ship and the direction of relative velocity, and β is the relative bearing of the target ship.

[0087] S3: Construct a hazard assessment function to determine the risk of collision on any trial course of the vessel based on a set of collision risk parameters and a preset safety threshold. Next, determine whether there is a collision risk between the vessel and each target, and construct a set of dangerous course directions for each target. ;

[0088] Furthermore, in step S3 above, the collision risk between ships is determined based on the dynamic parameters of static and dynamic obstacles. Therefore, the risk determination function... for:

[0089] ;

[0090] in, For safe distance threshold, For time threshold; when =1 indicates that the course is being tested. There is a risk of collision below; The closest encounter distance between the ship and the target ship (obstacle) on the i-th heading; The closest encounter time between the ship and the target ship (obstacle) on the i-th heading; This is the closest encounter distance between the ship and the target ship (obstacle) on the i-th heading; The distance between the target ship's bow and the ship's bow at the i-th heading; The time when the target ship on the i-th heading passes the bow of this ship;

[0091] S4: Combine dangerous course sets for all targets This yields the total set of dangerous routes C and the total set of non-dangerous routes for the ship across all space. ;

[0092] Specifically, the set of course directions that pose a risk or no risk to this vessel from the aforementioned target vessel is as follows:

[0093] ;

[0094] ;

[0095] In the formula, For the first Dangerous course set for the target vessel For this ship relative to the first The target ship's course; For the first The set of non-dangerous course routes for the target vessel.

[0096] Therefore, the total set of dangerous course directions for the entire space of this ship is as follows:

[0097] ;

[0098] The total set of non-dangerous course directions for this vessel is:

[0099] ;

[0100] in, The target number of ships; This is the total set of dangerous course directions for the entire space of the ship. This is the total set of non-dangerous course directions for this vessel.

[0101] The above set represents the danger set of obstacles under different headings at the ship's current position and speed. If the speed dimension is considered, then the total danger set under both speed and heading dimensions is constructed as follows:

[0102] ;

[0103] In the formula, This indicates that the ship is assembled in all directions for danger. This indicates the minimum speed permitted for this vessel. This indicates the maximum speed permitted for this vessel. This is the total set of hazards that takes into account both the speed and heading dimensions.

[0104] S5: Extend the risk index to the ship's speed dimension by traversing the allowed speed range. Generate an omnidirectional danger set in the spatiotemporal dimension Total collection of dangers The speed of the ship was determined. Omnidirectional Danger Set .

[0105] Furthermore, ship maneuvering performance parameters are introduced to construct a dynamic risk field model based on a ring Gaussian density function, and the model is used to calculate the risk field of the ship at speeds... Omnidirectional Danger Set The cumulative probability on the course of the ship is used as the ship's trajectory. Risk index set below ;

[0106] Furthermore, the collision risk index is represented by the area of ​​a ring-shaped Gaussian density function, the expression of which is as follows:

[0107] ;

[0108] in, Indicates the course of the ships being traversed. Indicates the current number of the ship One course, ; This represents the angular span that controls the width of the distribution, indicating the angle the ship can currently turn. A ship with strong turning ability will have a larger value, and vice versa. This value is related to the ship type and length. As an optional value, it can be set to... ; This represents the phase term of the cyclic Gaussian summation.

[0109] Furthermore, the risk index By calculating the dangerous course section [ , The integral of the upper ring Gaussian density function is obtained, where , Speeds Omnidirectional Danger Set The left and right boundaries;

[0110] When the interval does not cross 0°, the risk index The calculation formula is:

[0111] .

[0112] When the interval crosses 0°, the risk index The calculation formula is:

[0113] ;

[0114] in, Indicates that the current course of the ship is Risk index under certain circumstances This is a concept related to angular span, which is positively correlated with a ship's turning performance. A ship with good turning performance will have a larger value. For the ship at speed Omnidirectional Danger Set Left boundary, for The right boundary. Indicates the course of the ships being traversed. This represents the phase term of the cyclic Gaussian summation; This indicates the current course of the vessel. The cumulative distribution function (CDF) of the standard normal distribution is given by the following formula:

[0115] ;

[0116] ;

[0117] In the formula, This is the error function.

[0118] S6: After step S5, the entire set of hazards has been traversed. Then, the set of risk factors at different speeds and under different headings is obtained. The expression is as follows:

[0119] ;

[0120] in, For the current course of the ship and speed The set of risk indices under the space-time dimension; because speed takes time into account and heading takes space into account, this set is also called the danger set under the space-time dimension.

[0121] S7: Based on a set of risk indices Generate an omnidirectional dynamic risk heat map with heading and speed as coordinate axes. That is, by using the principles of thermodynamic mapping, inputting this multi-dimensional data, a heat map of ship collision hazards is finally generated.

[0122] This invention also discloses a multi-source fusion-based system for generating heatmaps of ship course and speed collision risks, comprising:

[0123] The multi-source data fusion and processing module is used to acquire and fuse AIS, radar and electronic chart data, and calculate the spatial state parameters and collision risk parameter set of dynamic and static obstacles;

[0124] The hazard set construction module, connected to the multi-source data fusion and processing module, is used to construct a hazard course set for each target based on the hazard determination function and fuse them to generate a total hazard course set for the entire space.

[0125] The dynamic risk field calculation module, connected to the hazard set construction module, is used to receive ship maneuvering performance parameters, calculate the risk index based on the ring Gaussian density function model, and extend it to the speed dimension to generate a set of risk coefficients in the spatiotemporal dimension.

[0126] The risk heatmap generation and output module, connected to the dynamic risk field calculation module, is used to render the set of risk coefficients in the spatiotemporal dimension into an omnidirectional dynamic risk heatmap and output it.

[0127] The present invention also discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the program, it implements the steps of the method for generating a collision risk heat map of a ship's course and speed based on multi-source fusion.

[0128] The present invention also discloses a computer-readable storage medium storing a computer program thereon, characterized in that the computer program, when executed by a processor, implements the steps of the method for generating a collision risk heat map of a ship's course and speed based on multi-source fusion.

[0129] The present invention adopts the above technical solutions and has the following beneficial effects compared with the prior art: (1) The present invention integrates multi-source data such as radar, AIS, and electronic charts, and constructs a risk field covering the entire course and speed. It can simultaneously process the risk contribution of all dynamic and static obstacles around it and generate a continuous and unified full-space risk heat map, which fundamentally solves the problem of fragmented risk cognition in multi-target scenarios and provides the driver with a global and intuitive situational awareness. (2) The present invention abandons subjective weighting and innovatively introduces the parameters of bow distance (BCR) and bow time (BCT), and constructs a multi-dimensional feature set in combination with the classic DCPA and TCPA. The risk mapping is carried out using a unified mathematical model based on the ring Gaussian density function, which avoids the evaluation bias caused by the coupling of indicators and makes the evaluation results closer to the nautical physical intuition. The model is transparent and highly interpretable. (3) The present invention introduces the ship's maneuverability (such as turning ability) as a key parameter (σ) into the Gaussian risk field model, so that the generated risk heat map can dynamically respond to the ship's maneuverability. This means that the "safe space" assessed by the system changes dynamically with the characteristics of the ship, providing personalized risk assessments for different ship types and significantly enhancing the engineering applicability of the method. (4) The risk heat map is dynamically updated based on real-time data, identifying safe course / speed ranges and high-risk areas. This enables the driver to proactively plan collision avoidance strategies, rather than just reacting to the current danger. At the same time, the intuitive visualization output greatly reduces the difficulty of human interpretation of complex data and the probability of misjudgment, simplifying the operation process.

[0130] Obviously, the described embodiments are only a portion, not all, of the embodiments of this application. Without conflict, the embodiments and features described and illustrated herein can be combined with each other. The components of the embodiments of this application generally described and illustrated in the accompanying drawings can be arranged and designed in various different configurations. Therefore, the detailed description of the embodiments of this application is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

Claims

1. A method for generating heatmaps of ship course, speed, and collision risk based on multi-source fusion, characterized in that, It includes the following steps: S1: Acquire sensing data of the surrounding environment of the ship; virtualize the channel boundary and coastline in the sensing data, emit rays with an interval of α centered on the ship, and use the intersection of the rays with the channel boundary or coastline as virtual static obstacles, and incorporate the position information of the virtual static obstacles into the static navigation obstacle information list to construct a dataset of all dynamic and static obstacles; the sensing data includes dynamic information of other ships from the automatic identification system, target detection information from the radar system, and static navigation obstacle information from electronic charts; S2: Calculate the collision risk parameters of each obstacle relative to the ship to construct a collision risk dataset. The collision risk parameters include at least the minimum encounter distance, minimum encounter time, distance past the bow, and time past the bow. S3: Construct a hazard assessment function to determine the risk of collision on any trial course of the vessel based on a set of collision risk parameters and a preset safety threshold. Next, determine whether there is a collision risk between the vessel and each target, and construct a set of dangerous course directions for each target. ; S4: Combine dangerous course sets for all targets This yields the total set of dangerous routes C and the total set of non-dangerous routes for the ship across all space. ; S5: Extend the risk index to the ship's speed dimension by traversing the allowed speed range. Generate an omnidirectional hazard set in the spatiotemporal dimension Total collection of dangers ;in, This indicates the minimum speed permitted for this vessel. Indicates the maximum speed permitted for this vessel; S6: Traverse the total set of dangers derive the speed Omnidirectional Danger Set Furthermore, ship maneuvering performance parameters are introduced to construct a dynamic risk field model based on a ring Gaussian density function, and the ship's speed is calculated. Omnidirectional Danger Set The cumulative probability above serves as the risk index set for the ship under the current course. Based on this, traverse the complete set of dangers. The set of speed and heading risk indices is derived. ;in, For the current course of the ship , The current speed of the ship; the density function expression is as follows: ； in, Indicates the course of the ships being traversed. Indicates the current number of the ship One course, ; The angular span representing the turning performance of a ship; This represents the phase term of the cyclic Gaussian summation; S7: Based on a set of speed and heading risk indices Generate an omnidirectional dynamic risk heat map with heading and speed as coordinate axes.

2. The method for generating ship course, speed, and collision risk heatmaps based on multi-source fusion according to claim 1, characterized in that, The closest encounter distance between our ship and the target ship in step S2 The calculation formula is as follows: ； Time of nearest encounter between this vessel and the target vessel The calculation formula is as follows: ； in, This is the true bearing of the target ship relative to our own ship. Indicates the distance between the target ship and this ship. Indicates the direction of the target's relative motion vector. Represents the relative motion vector; The formula for calculating the distance between the target ship and the bow of this ship is as follows: ； The formula for calculating the time it takes for the target vessel to pass the bow of this vessel is as follows: ； in, The minimum encounter distance is θ, where θ is the angle between the bearing of the target ship and the direction of relative velocity, and β is the relative bearing of the target ship.

3. The method for generating ship course, speed, and collision risk heatmaps based on multi-source fusion according to claim 1 or 2, characterized in that, Danger determination function in step S3 for: ； in, For safe distance threshold, For time threshold; when =1 indicates that the course is being tested. There is a risk of collision below; The closest encounter distance between the current vessel and the target vessel is at the i-th heading. The closest encounter time between the current vessel and the target vessel on the i-th heading; This is the closest encounter distance between the current vessel and the target vessel on the i-th heading; The distance between the target ship's bow and the ship's bow at the i-th heading; Let be the time when the target ship passes the bow of our ship on the i-th heading.

4. The method for generating ship course, speed, and collision risk heatmaps based on multi-source fusion according to claim 1, characterized in that, In step S4, the target vessel constitutes a dangerous course set for this vessel. Non-dangerous heading set The expression is: ； ； in, For the first Dangerous course set for the target vessel For this ship relative to the first The target ship's course; For the first The set of non-dangerous course routes for the target vessel; For this ship relative to the first The target ship's Danger determination function for heading; The relative total set of dangerous course directions for this ship across all space is as follows: ； in, The target number of ships; This is the total set of dangerous course directions for the entire space of the ship.

5. The method for generating ship course, speed, and collision risk heatmaps based on multi-source fusion according to claim 1 or 4, characterized in that, Step S5: Total Hazard Set in Moderate and Heading Dimensions The expression is: ； in, This indicates an all-directional hazard assembly for the vessel.

6. The method for generating a ship course-speed collision risk heat map based on multi-source fusion according to claim 5, wherein the risk index in step S5... By calculating the dangerous course section [ , The integral of the upper ring Gaussian density function is obtained as follows: When the interval does not cross 0°, the risk index The calculation formula is: ； When the interval crosses 0°, the risk index The calculation formula is: ； in, Indicates that the current course of the ship is Risk index under certain circumstances The angular span representing the turning performance of a ship; For the ship at speed Omnidirectional Danger Set Left boundary, For the ship at speed Omnidirectional Danger Set Right boundary; Indicates the course of the ships being traversed. This represents the phase term of the cyclic Gaussian summation; This indicates the current course of the vessel. The cumulative distribution function of the standard normal distribution; The expression is as follows: ； ； in, This is the error function.

7. The method for generating ship course, speed, and collision risk heatmaps based on multi-source fusion according to claim 6, characterized in that, The set of expressions for the risk factors at different speeds in step S6 are as follows: ； in, For the current course of the ship and speed A dangerous set in a given dimension.

8. A system for generating heatmaps of ship course and speed collision risks based on multi-source fusion, employing the method for generating heatmaps of ship course and speed collision risks based on multi-source fusion as described in any one of claims 1 to 7, characterized in that, The system includes: The multi-source data fusion and processing module is used to acquire and fuse AIS, radar and electronic chart data, and calculate the spatial state parameters and collision risk parameter set of dynamic and static obstacles; The hazard set construction module, connected to the multi-source data fusion and processing module, is used to construct a hazard course set for each target based on the hazard determination function and fuse them to generate a total hazard course set for the entire space. The dynamic risk field calculation module, connected to the hazard set construction module, is used to receive ship maneuvering performance parameters, calculate the risk index based on the ring Gaussian density function model, and extend it to the speed dimension to generate a set of risk coefficients in the spatiotemporal dimension. The risk heatmap generation and output module, connected to the dynamic risk field calculation module, is used to render the set of risk coefficients in the spatiotemporal dimension into an omnidirectional dynamic risk heatmap and output it.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the program, it implements the steps of the method for generating a ship course, speed, and collision risk heat map based on multi-source fusion as described in any one of claims 1 to 7.

Images