Marine collision avoidance rule-based ship collision avoidance test scene generation method and system

Through parameterization and decoupling and random generation of ship collision avoidance test scenarios, the problems of low efficiency and poor coverage of ship collision avoidance test scenarios in the prior art are solved, and efficient and diversified test scenario generation is achieved, which is particularly suitable for Ice trapped ice scene in polar seas.

CN119088684AActive Publication Date: 2024-12-06SHANGHAI SHIP & SHIPPING RES INST CO LTD +1

Patent Information

Application Number
CN202411127148.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-15
Publication Date
2024-12-06
Estimated Expiration
2044-08-15

AI Technical Summary

Technical Problem

The existing ship collision avoidance test scenario generation methods are inefficient and poorly covered, making it difficult to fully test collision avoidance algorithms, especially in polar seas where ice traps and floating ice problems exist, and traditional methods are difficult to generate comprehensive and diverse test scenarios.

Method used

By obtaining the ship's AIS data, parameterization and decoupling are performed based on the maritime collision avoidance rules, azimuth angles and distances are decomposed, the scene initial parameters are randomly generated, the relative speed vector and encounter time are calculated, and various types of ship collision avoidance test scenarios are generated.

Benefits of technology

Achieve efficient, comprehensive and diverse ship collision avoidance test scenarios, improve the credibility and efficiency of the test, and can cover all the situations stipulated by the maritime collision avoidance rules, especially in polar waters, which provide more accurate ice. Simulation of sleepy ice scene.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119088684A_ABST
    Figure CN119088684A_ABST
Patent Text Reader

Abstract

The invention provides a ship collision avoidance test scene generation method and system based on a maritime collision avoidance rule, and the method comprises the steps: obtaining AIS data of a ship and a target ship, determining a ship encounter situation based on the maritime collision avoidance rule, carrying out the parametric decoupling of the ship encounter situation, achieving the quantization of an azimuth angle and the quantization of a distance, and obtaining a ship collision avoidance test scene. The azimuth angle of a target ship relative to the ship and the distance between the ship and the target ship are quantitatively calculated, the ship encounter situation is divided to obtain a plurality of scene types, the range of each parameter for constructing a scene is set according to the scene types, then scene initial parameters are generated, the nearest encounter distance and the nearest encounter time are calculated, and the scene initial parameters are obtained. Therefore, a plurality of different types of ship collision avoidance test scenes are automatically generated, and the method has the advantages of high scene generation efficiency, controllable scene differentiation degree, high test reliability and the like.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of ship testing, and in particular to a method and system for generating a ship collision avoidance test scenario based on marine collision avoidance rules. Background Art

[0002] After more than half a century of development, the research methods of ship navigation collision avoidance technology have evolved from early classical mathematical theories to the use of artificial intelligence technology and interdisciplinary subjects. Considering the intelligent and practical needs of maritime collision avoidance rules and practical requirements, it is very necessary to test the collision avoidance algorithm.

[0003] In the current collision avoidance test scenario construction, the quantification of collision avoidance rules and complex situations is the key. When constructing collision avoidance scenarios based on the International Regulations for Preventing Collisions at Sea, 1972, since the content of these rules is mainly formulated for the crew on board, no parameterized standards are provided, and their implementation depends on the subjective judgment of the crew. Therefore, it cannot be directly used to formulate a comprehensive collision avoidance test scenario system at present, and does not include information on waterway traffic scenarios, so it is necessary to quantify and classify the collision avoidance indicators. At the same time, considering that the collision avoidance scenarios in the real navigation environment are difficult to exhaust, and each type of collision avoidance scenario is composed of many factors, and the current traditional manual setting method to generate collision avoidance scenarios is huge, very inefficient and has great limitations. It is difficult to cover all encounter situations stipulated by the maritime collision rules, and it is impossible to fully test the collision avoidance algorithm. At the same time, the probability of occurrence of dangerous scenarios and extreme scenarios is relatively small, resulting in few effective samples extracted by simulation and low test efficiency. In addition, there are also unique problems such as ice traps and floating ice in polar waters, so the test scenario parameters should be further clarified. Summary of the invention

[0004] In order to solve the problems of large workload, low efficiency, large limitations, inability to fully test the collision avoidance algorithm, ice traps and floating ice in polar waters in the current collision avoidance scene construction process, the present invention provides a ship collision avoidance test scene generation method based on maritime collision avoidance rules. Based on the various types of ship encounter situations specified in the maritime collision avoidance rules, the ship encounter situation is parameterized and decoupled into multiple scene types according to azimuth and heading angles, and a large number of different types of ship collision avoidance test scenes are quickly obtained by randomly generating determined values ​​within different ranges and using specific calculation methods, thereby effectively solving the problem of poor comprehensiveness of traditional ship collision avoidance scene generation. The present invention also relates to a ship collision avoidance test scene generation system based on maritime collision avoidance rules.

[0005] The technical solution of the present invention is as follows:

[0006] A method for generating a ship collision avoidance test scenario based on marine collision avoidance rules, characterized in that it comprises the following steps:

[0007] Data acquisition and parameter decoupling steps: acquiring AIS data of the own ship and the target ship, wherein the AIS data includes the latitude and longitude position information, heading angle, own ship speed and target ship speed of the own ship and the target ship, determining the ship encounter situation based on the acquired AIS data and the maritime collision avoidance rules, and then performing parameter decoupling on the ship encounter situation to achieve azimuth quantification and distance quantification, and respectively quantifying and calculating the azimuth of the target ship relative to the own ship, and the distance between the own ship and the target ship;

[0008] Scenario type determination and parameter range setting steps: dividing the ship encounter situation according to the heading angle and the quantified azimuth of the target ship relative to the own ship to obtain multiple scenario types, setting the heading angle range of the own ship and the target ship, and the azimuth range of the target ship relative to the own ship according to the scenario type, and setting the distance range between the own ship and the target ship according to the quantified distance between the own ship and the target ship and based on the maritime collision avoidance rules;

[0009] The step of generating the initial parameters of the scene: the heading angle determination values ​​of the own ship and the target ship, the azimuth determination value of the target ship relative to the own ship, and the distance determination value between the own ship and the target ship are randomly generated by the random number generator within the respective set ranges, and the randomly generated determination values ​​are all used as the initial parameters of the scene;

[0010] Relative speed vector calculation steps: calculate the own ship's speed vector according to the determined value of the own ship's heading angle and the own ship's speed, calculate the target ship's speed vector according to the determined value of the target ship's heading angle and the target ship's speed, and calculate the target ship's relative speed vector relative to the own ship according to the own ship's speed vector and the target ship's speed vector;

[0011] The steps of calculating the closest encounter distance and the closest encounter time are as follows: the closest encounter distance between the target ship and the own ship is calculated according to the determined value of the distance between the own ship and the target ship, the relative speed vector, the determined value of the azimuth of the target ship relative to the own ship, and the angle formed by the relative speed vector and the line connecting the own ship's position, and the closest encounter time between the target ship and the own ship is calculated according to the determined value of the distance between the own ship and the target ship and the angle formed by the relative speed vector and the line connecting the own ship's position;

[0012] Ship collision avoidance test scenario generation step: automatically generate a plurality of different types of ship collision avoidance test scenarios based on the generated scenario initial parameters, the speeds of the own ship and the target ship, the closest encounter distance, and the closest encounter time.

[0013] Preferably, in the data acquisition and parameter decoupling step, the ship encounter situation includes an overtaking scenario, a head-on encounter scenario and a cross-encounter scenario.

[0014] Preferably, in the scene type determination and parameter range setting steps, dividing the ship encounter situation according to the heading angle and the quantified azimuth of the target ship relative to the own ship includes: dividing the ship encounter situation into port and starboard according to the azimuth of the target ship relative to the own ship, and dividing the intersection encounter scene into small-angle intersection encounter and large-angle intersection encounter according to the heading angle difference between the own ship and the target ship.

[0015] Preferably, in the scene type determination and parameter range setting steps, the initial encounter situation of the two ships is also classified according to the distance range between the own ship and the target ship. If the distance between the own ship and the target ship is greater than a first preset distance threshold and is less than or equal to an urgent danger distance threshold, the initial encounter situation is regarded as an urgent danger situation; if the distance between the own ship and the target ship is greater than the urgent danger distance threshold and is less than or equal to a third preset distance threshold, the initial encounter situation is regarded as an urgent situation; if the distance between the own ship and the target ship is greater than the third preset distance threshold and is less than or equal to a fourth preset distance threshold, the initial encounter situation is regarded as a general encounter situation.

[0016] Preferably, the imminent danger distance threshold is the ship's emergency braking distance plus a 50% safety margin distance, and it is determined whether the ship is sailing in the polar region. If there is floating ice during the ship's polar navigation, the imminent danger distance threshold is the ship's emergency braking distance plus a 100% safety margin distance.

[0017] A ship collision avoidance test scenario generation system based on maritime collision avoidance rules, characterized in that it comprises a data acquisition and parameter decoupling module, a scenario type determination and parameter range setting module, a scenario initial parameter generation module, a relative speed vector calculation module, a closest encounter distance and closest encounter time calculation module and a ship collision avoidance test scenario generation module which are connected in sequence.

[0018] The data acquisition and parameter decoupling module acquires AIS data of the own ship and the target ship, wherein the AIS data includes the latitude and longitude position information, heading angle, own ship speed and target ship speed of the own ship and the target ship, determines the ship encounter situation based on the acquired AIS data and the maritime collision avoidance rules, and then performs parameter decoupling on the ship encounter situation to achieve azimuth quantification and distance quantification, and respectively quantifies and calculates the azimuth of the target ship relative to the own ship, and the distance between the own ship and the target ship;

[0019] The scenario type determination and parameter range setting module divides the ship encounter situation according to the heading angle and the quantified azimuth of the target ship relative to the own ship to obtain multiple scenario types, sets the heading angle range of the own ship and the target ship, and the azimuth range of the target ship relative to the own ship according to the scenario type, and sets the distance range between the own ship and the target ship according to the quantified distance between the own ship and the target ship and based on the maritime collision avoidance rules;

[0020] The scene initial parameter generation module randomly generates the heading angle determination values ​​of the own ship and the target ship, the azimuth determination value of the target ship relative to the own ship, and the distance determination value between the own ship and the target ship through a random number generator within the respective set ranges, and each randomly generated determination value is used as the scene initial parameter;

[0021] The relative speed vector calculation module calculates the speed vector of the own ship according to the determined value of the own ship's heading angle and the speed of the own ship, calculates the speed vector of the target ship according to the determined value of the heading angle of the target ship and the speed of the target ship, and calculates the relative speed vector of the target ship relative to the own ship according to the speed vector of the own ship and the speed vector of the target ship;

[0022] The module for calculating the closest encounter distance and the closest encounter time calculates the closest encounter distance between the target ship and the own ship according to the determined value of the distance between the own ship and the target ship, the relative speed vector, the determined value of the azimuth of the target ship relative to the own ship, and the angle formed by the relative speed vector and the line connecting the own ship's position, and calculates the closest encounter time between the target ship and the own ship according to the determined value of the distance between the own ship and the target ship and the angle formed by the relative speed vector and the line connecting the own ship's position;

[0023] The ship collision avoidance test scenario generation module automatically generates a plurality of different types of ship collision avoidance test scenarios based on the generated scenario initial parameters, the speeds of the own ship and the target ship, the closest approach distance, and the closest approach time.

[0024] Preferably, the ship encounter situation includes an overtaking scenario, a head-on encounter scenario and a cross-encounter scenario.

[0025] Preferably, in the scenario type determination and parameter range setting module, dividing the ship encounter situation according to the heading angle and the quantified azimuth of the target ship relative to the own ship includes: dividing the ship encounter situation into port and starboard according to the azimuth of the target ship relative to the own ship, and dividing the intersection encounter scenario into small-angle intersection encounter and large-angle intersection encounter according to the heading angle difference between the own ship and the target ship.

[0026] Preferably, in the scene type determination and parameter range setting module, the initial encounter situation of the two ships is also classified according to the distance range between the own ship and the target ship. If the distance between the own ship and the target ship is greater than the first preset distance threshold and is less than or equal to the urgent danger distance threshold, the initial encounter situation is regarded as an urgent danger situation; if the distance between the own ship and the target ship is greater than the urgent danger distance threshold and is less than or equal to the third preset distance threshold, the initial encounter situation is regarded as an urgent situation; if the distance between the own ship and the target ship is greater than the third preset distance threshold and is less than or equal to the fourth preset distance threshold, the initial encounter situation is regarded as a general encounter situation.

[0027] Preferably, the imminent danger distance threshold is the ship's emergency braking distance plus a 50% safety margin distance, and it is determined whether the ship is sailing in the polar region. If there is floating ice during the ship's polar navigation, the imminent danger distance threshold is the ship's emergency braking distance plus a 100% safety margin distance.

[0028] The beneficial effects of the present invention are:

[0029] The invention provides a method for generating a ship collision avoidance test scenario based on the rules for avoiding collision at sea. The method comprises the following steps: firstly obtaining AIS data of a own ship and a target ship, determining a ship encounter situation based on the AIS data and the international rules for avoiding collision at sea, then performing parameter decoupling on the ship encounter situation to realize azimuth quantification and distance quantification, then dividing the ship encounter situation based on the heading angle and the quantified azimuth of the target ship relative to the own ship to obtain a plurality of scenario types, setting the heading angle range of the own ship and the target ship and the azimuth range of the target ship relative to the own ship according to the scenario type, and setting the distance range between the own ship and the target ship based on the international rules for avoiding collision at sea, then randomly taking values ​​in different ranges by a random number generator as the initial parameters of the scenario, and adopting a specific calculation method to quickly obtain a large number of different types of ship collision avoidance test scenarios, that is, the invention divides the ship encounter situation into a plurality of levels according to the quantized azimuth and distance based on the various types of encounter situations specified in the rules for avoiding collision at sea, and randomly taking values ​​in different ranges, so as to quickly obtain a large number of different types of ship collision avoidance test scenarios, ensure the diversity and comprehensiveness of the test scenarios, and effectively solve the problem of poor comprehensiveness of traditional ship collision avoidance scenario generation. The present invention quantitatively represents concepts such as azimuth and distance related to collision avoidance based on actual marine collision avoidance test data. By quantitatively representing important concepts related to marine collision avoidance rules, the present invention can parameterize and decouple complex encounter situations according to the azimuth and distance of a target ship, and superimpose the speed and heading data of the own ship and the target ship to form a parameterized collision avoidance test scenario. Therefore, a large number of different types of ship collision avoidance test scenarios can be automatically acquired. In addition, the ship collision avoidance test scenarios are generated in a process-based manner and can be generated in parallel, so the scenario generation efficiency is high, the degree of scenario differentiation is controllable, and the test credibility is high.

[0030] The present invention also relates to a ship collision avoidance test scenario generation system based on maritime collision avoidance rules. The system corresponds to the above-mentioned ship collision avoidance test scenario generation method based on maritime collision avoidance rules, and can be understood as a system for realizing the above-mentioned ship collision avoidance test scenario generation method based on maritime collision avoidance rules, comprising a data acquisition and parameter decoupling module, a scenario type determination and parameter range setting module, a scenario initial parameter generation module, a relative speed vector calculation module, a closest encounter distance and closest encounter time calculation module, and a ship collision avoidance test scenario generation module, which are sequentially connected. The modules work in coordination with each other. Under AIS data, based on various types of ship encounter situations specified in maritime collision avoidance rules, the ship encounter situation is parameterized and decoupled to realize azimuth quantization and distance quantization, and the ship encounter situation is divided into multiple scenario types according to azimuth and heading angle, and the ship encounter situation is divided into multiple levels according to distance, and a large number of different types of ship collision avoidance test scenarios are quickly acquired by randomly taking values ​​in different ranges and adopting a specific calculation method, thereby effectively solving the problem of poor comprehensiveness of traditional ship collision avoidance scenario generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a flow chart of a method for generating a ship collision avoidance test scenario based on maritime collision avoidance rules of the present invention.

[0032] Figure 2 It is a schematic diagram of the encounter situation of ships of the present invention. DETAILED DESCRIPTION

[0033] The present invention will be described below in conjunction with the accompanying drawings.

[0034] The present invention relates to a method for generating a ship collision avoidance test scenario based on maritime collision avoidance rules, which parametrically decouples a complex encounter situation according to the target ship azimuth and target ship distance, superimposes the initial speed and heading data of the own ship and the target ship, and thus forms a parametric collision avoidance test scenario. The flowchart of the method is as follows Figure 1 As shown, the following steps are included in sequence:

[0035] 1. Data acquisition and parameter decoupling steps: Obtain the AIS data of the own ship and the target ship. The AIS data includes the latitude and longitude position information, heading angle, own ship speed and target ship speed of the own ship and the target ship. According to the acquired AIS data and based on the rules for avoiding collisions at sea, determine the ship encounter situation, that is, according to Articles 13, 14 and 15 of the rules for avoiding collisions at sea, divide the ship encounter situation into overtaking, meeting and crossing; then perform parameter decoupling on the ship encounter situation to achieve azimuth quantification and distance quantification, and quantify and calculate the azimuth of the target ship relative to the own ship, that is, Figure 1 The azimuth of the target ship shown in the figure is quantified, that is, the azimuth of the target ship relative to the own ship is calculated according to the longitude and latitude of the own ship and the longitude and latitude of the target ship, and the distance between the own ship and the target ship is quantitatively calculated, that is, Figure 1 Quantification of target ship distance shown.

[0036] Specifically, the AIS data of the own ship and the target ship are first queried from the database. The AIS data includes the initial longitude and latitude position information of the own ship and the target ship, the heading angle θ of the own ship, and the O 、Target ship heading angle θ T 、Own ship speed V O and the target ship speed V T , and according to the longitude and latitude of the ship (lon O ,lat O ) and the target ship's longitude and latitude (lon T ,lat T ) respectively calculate the azimuth angle θ of the target ship relative to the own ship d And the distance d0 between the own ship and the target ship. Among them, the distance between two points on the earth's surface is calculated according to the Haversine formula:

[0037]

[0038] In the above formula, a is an intermediate variable calculated using spherical trigonometric functions, which represents the projection ratio of the straight-line distance between two points on the sphere, c is the length of the arc between the two points on the earth's surface, and R is the radius of the earth.

[0039] 2. Scenario type determination and parameter range setting step, which can also be called the ship encounter situation quantification and parameter range setting step: the ship encounter situation is divided according to the heading angle and the quantified azimuth of the target ship relative to the own ship to obtain multiple scenario types, and the heading angle range of the own ship and the target ship, as well as the azimuth range of the target ship relative to the own ship are set according to the scenario type, and the distance range between the own ship and the target ship is set based on the International Regulations for Preventing Collisions at Sea.

[0040] Specifically, Figure 2 As shown in FIG1 , the ship encounter situation is first divided into port and starboard according to the quantified azimuth of the target ship relative to the own ship, and the crossing encounter scene is divided into small-angle crossing encounter and large-angle crossing encounter according to the heading angle difference between the own ship and the target ship, thereby obtaining multiple scene types, that is, the ship encounter situation is quantified to be subdivided into multiple levels, and then the heading angle range of the own ship and the target ship, as well as the azimuth range of the target ship relative to the own ship are set according to the scene type. Among them, the ship encounter situation quantification is shown in Table 1.

[0041] Table 1

[0042]

[0043]

[0044] After quantifying the ship encounter situation, the distance range between the own ship and the target ship is set according to the quantified distance between the own ship and the target ship and based on the International Regulations for Preventing Collisions at Sea, and the initial encounter situation of the two ships is classified according to the distance range between the own ship and the target ship, that is, the initial encounter situation is quantified.

[0045] Specifically, the current rules for avoiding collisions at sea do not give precise definitions for many important concepts, such as urgent situations and urgent danger situations, which reflect the changing process of the relationship between ships during the navigation process. Therefore, the present invention quantitatively characterizes the concepts related to collision avoidance, such as direction and distance, based on actual collision avoidance test data at sea. The present invention is based on the international rules for avoiding collisions at sea and classifies the initial encounter situation of the two ships according to the distance range between the own ship and the target ship. If the distance between the own ship and the target ship is greater than the first preset distance threshold and less than or equal to the urgent danger distance threshold D S, that is, 0<distance≤D S , the initial encounter situation is regarded as an urgent danger situation; if the distance between the own ship and the target ship is greater than the urgent danger distance threshold and less than the third preset distance threshold, that is, D S If the distance between the own ship and the target ship is greater than the third preset distance threshold and less than the fourth preset distance threshold, that is, 1 < distance ≤ 6, the initial encounter situation is regarded as a general encounter situation. That is, the initial encounter situation is classified according to the distance between the own ship and the target ship, and is divided into a general encounter situation, an urgent situation, and an urgent and dangerous situation in the avoidance stage, which can also be called the quantification of the initial encounter situation, as shown in Table 2.

[0046] Table 2:

[0047] Type Name Minimum distance (nmile) Maximum distance (nmile) Usually encounter situations 1 6** Urgent situation <![CDATA[D S ]]> 1 Imminent danger situation 0 <![CDATA[D S *]]>

[0048] Among them, the general situation refers to avoiding the formation of an urgent situation under any visibility. An urgent situation means that it is impossible to pass at a safe distance by only one ship. An urgent and dangerous situation means that a collision cannot be avoided by only one ship.

[0049] It should be noted that the imminent danger distance D S It refers to the ship's emergency braking distance plus a 50% safety margin. During polar navigation, if there is floating ice, the imminent danger distance needs to be further increased to the ship's emergency braking distance plus a 100% safety margin. Among them, 6 nautical miles is the 6 nautical miles required by the maritime collision prevention regulations to find the target ship, so only the situation within 6 nautical miles is considered.

[0050] 3. Steps for generating initial parameters of the scene: Within the respective set ranges, the random number generator is used to randomly generate the determined values ​​of the heading angles of the host ship and the target ship, the determined value of the azimuth angle of the target ship relative to the host ship, and the determined value of the distance between the host ship and the target ship. The randomly generated determined values ​​are all used as the initial parameters of the scene.

[0051] That is to say, after the value ranges of the initial heading angles, azimuth angles and distances of the own ship and the target ship are determined, the present invention randomly generates a certain value within each value range through a computer random value selection method as the initial parameter of the scene, and then generates a parameterized collision avoidance test scene in subsequent steps.

[0052] 4. Relative speed vector calculation steps: Calculate the speed vector of own ship according to the determined value of own ship's heading angle and own ship's speed, and calculate the speed vector of target ship according to the determined value of target ship's heading angle and target ship's speed, and calculate the relative speed vector of target ship relative to own ship according to own ship's speed vector and target ship's speed vector.

[0053] Specifically, let the ship's speed be VO , the determined value of the ship's heading angle is θ O , the target ship speed is V T , the heading angle of the target ship is determined as θ T , determine the value θ according to the ship's heading angle O and own ship's speed V O Calculate the ship's speed vector V Ox and V Oy , and determine the value θ according to the heading angle of the target ship T and the target ship speed V T Calculate the target ship speed vector V Tx and V Ty , that is, convert the heading angle and speed of the own ship and the target ship into the component V of the velocity vector Ox 、V Oy 、V Tx 、V Ty (Assume that the east direction is the positive direction of the x-axis and the north direction is the positive direction of the y-axis), recorded as:

[0054]

[0055] Then, the relative speed vector V of the target ship relative to the own ship is calculated based on the own ship's speed vector and the target ship's speed vector. R , then:

[0056]

[0057] Among them, V R is the magnitude of the relative velocity vector, θ R is the direction of the relative velocity vector (angle relative to the x-axis).

[0058] 5. Calculation steps for the closest encounter distance and the closest encounter time: Calculate the closest encounter distance between the target ship and own ship based on the determined value of the distance between own ship and target ship, the relative speed vector, the determined value of the azimuth of the target ship relative to own ship, and the angle formed by the relative speed vector and the line connecting the position of own ship. Calculate the closest encounter time between the target ship and own ship based on the determined value of the distance between own ship and target ship and the angle formed by the relative speed vector and the line connecting the position of own ship.

[0059] Specifically, the value d0 and the relative speed vector d0 are determined according to the distance between the own ship and the target ship. The azimuth angle of the target ship relative to the own ship is determined as θ d The angle α formed by the relative velocity vector and the line connecting the own ship's position is used to calculate the closest encounter distance DCPA between the target ship and the own ship. The collision point of the two ships can be determined based on the calculated closest encounter distance DCPA; and the value d0, the relative velocity vector, and the distance between the own ship and the target ship are determined based on the distance d0, the relative velocity vector, and the distance between the own ship and the target ship. The angle α formed by the line connecting the target ship's position is used to calculate the closest encounter time TCPA between the target ship and the ship. The calculation formula is as follows:

[0060]

[0061] Wherein, D is the closest encounter distance DCPA between the own ship and the target ship, and T is the closest encounter time TCPA between the own ship and the target ship.

[0062] 6. Ship collision avoidance test scenario generation step: Based on the generated initial parameters of the scenario (i.e., the heading angles of the own ship and the target ship are determined as θ O ,θ T 、The azimuth angle of the target ship relative to the own ship θ d , and the distance between the own ship and the target ship is determined by d0), the speed V of the own ship and the target ship O 、V T , closest encounter distance D and closest encounter time T, and automatically generate multiple different types of ship collision avoidance test scenarios.

[0063] It should be noted that in order to ensure that the ship must take necessary avoidance actions, the process of automatically generating scenarios requires that the DCPA value be as small as possible (generally 0), while ensuring that the TCPA value is greater than zero (when the TCPA value is less than zero, the DCPA point is missed and the plan is invalid).

[0064] Example:

[0065] A detailed explanation is given by taking the generation of an intersection collision avoidance scenario as an example.

[0066] 1) First, determine the initial parameter range of the own ship and the target ship.

[0067] The course angle range of own ship: Assume that the course angle of own ship is between 60 degrees and 120 degrees; the azimuth range of target ship relative to own ship: Assume that the target ship is between 0 degrees and 90 degrees to the starboard side of own ship; the distance range between own ship and target ship: Assume that the distance is between 1 nautical mile and 3 nautical miles.

[0068] 2) Generate random initial parameters.

[0069] Randomly generate a determined value of the ship's heading angle: for example, the randomly generated ship's heading angle is 90 degrees; randomly generate a determined value of the target ship's azimuth relative to the ship: for example, the randomly generated target ship's azimuth relative to the ship is 30 degrees; randomly generate a determined value of the distance between the ship and the target ship: for example, the randomly generated distance is 2 nautical miles.

[0070] 3) Determine the type of encounter: According to the azimuth of the target ship relative to the own ship, it is determined that this is a crossing encounter with the target ship located on the starboard side of the own ship, because the azimuth of the target ship relative to the own ship is 30 degrees, which is within the range of 0 to 90 degrees.

[0071] 4) Calculate the distance to closest approach (DCPA) and time to closest approach (TCPA).

[0072] 5) Automatically generate a cross-encounter ship collision avoidance test scenario based on the determined values ​​of the heading angles of own ship and target ship, the speeds of own ship and target ship, the determined values ​​of the azimuth of the target ship relative to own ship, the determined values ​​of the distance between own ship and target ship, the closest encounter distance and the closest encounter time.

[0073] The present invention also relates to a ship collision avoidance test scenario generation system based on maritime collision avoidance rules. The system corresponds to the above-mentioned ship collision avoidance test scenario generation method based on maritime collision avoidance rules, and can be understood as a system for implementing the above-mentioned method. The system includes a data acquisition and parameter decoupling module, a scenario type determination and parameter range setting module, a scenario initial parameter generation module, a relative speed vector calculation module, a closest encounter distance and closest encounter time calculation module, and a ship collision avoidance test scenario generation module, which are connected in sequence. Specifically,

[0074] The data acquisition and parameter decoupling module acquires AIS data of the own ship and the target ship, wherein the AIS data includes the latitude and longitude position information, heading angle, own ship speed and target ship speed of the own ship and the target ship, determines the ship encounter situation based on the acquired AIS data and the maritime collision avoidance rules, and then performs parameter decoupling on the ship encounter situation to achieve azimuth quantification and distance quantification, and respectively quantifies and calculates the azimuth of the target ship relative to the own ship, and the distance between the own ship and the target ship;

[0075] The scenario type determination and parameter range setting module divides the ship encounter situation according to the heading angle and the quantified azimuth of the target ship relative to the own ship to obtain multiple scenario types, sets the heading angle range of the own ship and the target ship, and the azimuth range of the target ship relative to the own ship according to the scenario type, and sets the distance range between the own ship and the target ship according to the quantified distance between the own ship and the target ship and based on the maritime collision avoidance rules;

[0076] The scene initial parameter generation module randomly generates the heading angle determination values ​​of the own ship and the target ship, the azimuth determination value of the target ship relative to the own ship, and the distance determination value between the own ship and the target ship through a random number generator within the respective set ranges, and each randomly generated determination value is used as the scene initial parameter;

[0077] The relative speed vector calculation module calculates the speed vector of the own ship according to the determined value of the own ship's heading angle and the speed of the own ship, calculates the speed vector of the target ship according to the determined value of the heading angle of the target ship and the speed of the target ship, and calculates the relative speed vector of the target ship relative to the own ship according to the speed vector of the own ship and the speed vector of the target ship;

[0078] The module for calculating the closest encounter distance and the closest encounter time calculates the closest encounter distance between the target ship and the own ship according to the determined value of the distance between the own ship and the target ship, the relative speed vector, the determined value of the azimuth of the target ship relative to the own ship, and the angle formed by the relative speed vector and the line connecting the own ship's position, and calculates the closest encounter time between the target ship and the own ship according to the determined value of the distance between the own ship and the target ship and the angle formed by the relative speed vector and the line connecting the own ship's position;

[0079] The ship collision avoidance test scenario generation module automatically generates a plurality of different types of ship collision avoidance test scenarios based on the generated scenario initial parameters, the speeds of the own ship and the target ship, the closest approach distance, and the closest approach time.

[0080] Preferably, the ship encounter situation includes an overtaking scenario, a head-on encounter scenario and a cross-encounter scenario.

[0081] Preferably, in the scene type determination and parameter range setting module, the ship encounter situation is divided according to the heading angle and the quantified azimuth of the target ship relative to the own ship, including: dividing the ship encounter situation into port and starboard according to the azimuth of the target ship relative to the own ship, and dividing the intersection encounter scene into small-angle intersection encounter and large-angle intersection encounter according to the heading angle difference between the own ship and the target ship.

[0082] Preferably, in the scene type determination and parameter range setting module, the initial encounter situation of the two ships is also classified according to the distance range between the own ship and the target ship. If the distance between the own ship and the target ship is greater than the first preset distance threshold and is less than or equal to the urgent danger distance threshold, the initial encounter situation is regarded as an urgent danger situation; if the distance between the own ship and the target ship is greater than the urgent danger distance threshold and is less than or equal to the third preset distance threshold, the initial encounter situation is regarded as an urgent situation; if the distance between the own ship and the target ship is greater than the third preset distance threshold and is less than or equal to the fourth preset distance threshold, the initial encounter situation is regarded as a general encounter situation.

[0083] Preferably, the imminent danger distance threshold is the ship's emergency braking distance plus a 50% safety margin distance, and it is determined whether the ship is sailing in the polar region. If there is floating ice during the ship's polar navigation, the imminent danger distance threshold is the ship's emergency braking distance plus a 100% safety margin distance.

[0084] The present invention provides an objective and scientific method and system for generating ship collision avoidance test scenarios based on maritime collision avoidance rules. Based on various types of ship encounter situations specified in maritime collision avoidance rules, the ship encounter situations are divided into multiple scenario types according to azimuth and heading angles, and the ship encounter situations are divided into multiple levels according to the distance between the two ships. A large number of different types of ship collision avoidance test scenarios are quickly acquired by randomly taking values ​​within different ranges, thereby effectively solving the problem of poor comprehensiveness of traditional ship collision avoidance scenario generation.

[0085] It should be noted that the above-described specific implementations can enable those skilled in the art to more fully understand the invention, but do not limit the invention in any way. Therefore, although this specification has described the invention in detail with reference to the drawings and embodiments, those skilled in the art should understand that the invention can still be modified or replaced by equivalents. In short, all technical solutions and improvements that do not deviate from the spirit and scope of the invention should be included in the protection scope of the patent for the invention.

Claims

1. A method for generating a ship collision avoidance test scenario based on marine collision avoidance rules, characterized in that: The following steps are involved: Data acquisition and parameter decoupling steps: acquiring AIS data of the own ship and the target ship, wherein the AIS data includes the latitude and longitude position information, heading angle, own ship speed and target ship speed of the own ship and the target ship, determining the ship encounter situation based on the acquired AIS data and the maritime collision avoidance rules, and then performing parameter decoupling on the ship encounter situation to achieve azimuth quantification and distance quantification, and respectively quantifying and calculating the azimuth of the target ship relative to the own ship, and the distance between the own ship and the target ship; Scenario type determination and parameter range setting steps: dividing the ship encounter situation according to the heading angle and the quantified azimuth of the target ship relative to the own ship to obtain multiple scenario types, setting the heading angle range of the own ship and the target ship, and the azimuth range of the target ship relative to the own ship according to the scenario type, and setting the distance range between the own ship and the target ship according to the quantified distance between the own ship and the target ship and based on the maritime collision avoidance rules; The step of generating the initial parameters of the scene: the heading angle determination values ​​of the own ship and the target ship, the azimuth determination value of the target ship relative to the own ship, and the distance determination value between the own ship and the target ship are randomly generated by the random number generator within the respective set ranges, and the randomly generated determination values ​​are all used as the initial parameters of the scene; Relative speed vector calculation steps: calculate the own ship's speed vector according to the determined value of the own ship's heading angle and the own ship's speed, calculate the target ship's speed vector according to the determined value of the target ship's heading angle and the target ship's speed, and calculate the target ship's relative speed vector relative to the own ship according to the own ship's speed vector and the target ship's speed vector; The steps of calculating the closest encounter distance and the closest encounter time are as follows: the closest encounter distance between the target ship and the own ship is calculated according to the determined value of the distance between the own ship and the target ship, the relative speed vector, the determined value of the azimuth of the target ship relative to the own ship, and the angle formed by the relative speed vector and the line connecting the own ship's position, and the closest encounter time between the target ship and the own ship is calculated according to the determined value of the distance between the own ship and the target ship and the angle formed by the relative speed vector and the line connecting the own ship's position; Ship collision avoidance test scenario generation step: automatically generate a plurality of different types of ship collision avoidance test scenarios based on the generated scenario initial parameters, the speeds of the own ship and the target ship, the closest encounter distance, and the closest encounter time.

2. The method for generating a ship collision avoidance test scenario based on maritime collision avoidance rules according to claim 1, characterized in that: In the data acquisition and parameter decoupling step, the ship encounter situation includes an overtaking scenario, a head-on encounter scenario and a crossing encounter scenario.

3. The method for generating a ship collision avoidance test scenario based on marine collision avoidance rules according to claim 2, characterized in that: In the step of determining the scene type and setting the parameter range, dividing the ship encounter situation according to the heading angle and the quantified azimuth of the target ship relative to the own ship includes: dividing the ship encounter situation into port and starboard according to the azimuth of the target ship relative to the own ship, and dividing the intersection encounter scene into small-angle intersection encounter and large-angle intersection encounter according to the heading angle difference between the own ship and the target ship.

4. The method for generating a ship collision avoidance test scenario based on marine collision avoidance rules according to claim 1, characterized in that: In the step of determining the scene type and setting the parameter range, the initial encounter situation of the two ships is also classified according to the distance range between the own ship and the target ship. If the distance between the own ship and the target ship is greater than the first preset distance threshold and less than or equal to the urgent danger distance threshold, the initial encounter situation is regarded as an urgent danger situation; if the distance between the own ship and the target ship is greater than the urgent danger distance threshold and less than or equal to the third preset distance threshold, the initial encounter situation is regarded as an urgent situation; if the distance between the own ship and the target ship is greater than the third preset distance threshold and less than or equal to the fourth preset distance threshold, the initial encounter situation is regarded as a general encounter situation.

5. The method for generating a ship collision avoidance test scenario based on marine collision avoidance rules according to claim 4, characterized in that: The imminent danger distance threshold is the ship's emergency braking distance plus a 50% safety margin distance, and it is determined whether the ship is sailing in the polar region. If there is floating ice during the ship's polar navigation, the imminent danger distance threshold is the ship's emergency braking distance plus a 100% safety margin distance.

6. A ship collision avoidance test scenario generation system based on marine collision avoidance rules, characterized in that: It includes a data acquisition and parameter decoupling module, a scene type determination and parameter range setting module, a scene initial parameter generation module, a relative speed vector calculation module, a closest encounter distance and closest encounter time calculation module, and a ship collision avoidance test scene generation module. The data acquisition and parameter decoupling module acquires AIS data of the own ship and the target ship, wherein the AIS data includes the latitude and longitude position information, heading angle, own ship speed and target ship speed of the own ship and the target ship, determines the ship encounter situation based on the acquired AIS data and the maritime collision avoidance rules, and then performs parameter decoupling on the ship encounter situation to achieve azimuth quantification and distance quantification, and respectively quantifies and calculates the azimuth of the target ship relative to the own ship, and the distance between the own ship and the target ship; The scenario type determination and parameter range setting module divides the ship encounter situation according to the heading angle and the quantified azimuth of the target ship relative to the own ship to obtain multiple scenario types, sets the heading angle range of the own ship and the target ship, and the azimuth range of the target ship relative to the own ship according to the scenario type, and sets the distance range between the own ship and the target ship according to the quantified distance between the own ship and the target ship and based on the maritime collision avoidance rules; The scene initial parameter generation module randomly generates the heading angle determination values ​​of the own ship and the target ship, the azimuth determination value of the target ship relative to the own ship, and the distance determination value between the own ship and the target ship through a random number generator within the respective set ranges, and each randomly generated determination value is used as the scene initial parameter; The relative speed vector calculation module calculates the speed vector of the own ship according to the determined value of the own ship's heading angle and the speed of the own ship, calculates the speed vector of the target ship according to the determined value of the heading angle of the target ship and the speed of the target ship, and calculates the relative speed vector of the target ship relative to the own ship according to the speed vector of the own ship and the speed vector of the target ship; The module for calculating the closest encounter distance and the closest encounter time calculates the closest encounter distance between the target ship and the own ship according to the determined value of the distance between the own ship and the target ship, the relative speed vector, the determined value of the azimuth of the target ship relative to the own ship, and the angle formed by the relative speed vector and the line connecting the own ship's position, and calculates the closest encounter time between the target ship and the own ship according to the determined value of the distance between the own ship and the target ship and the angle formed by the relative speed vector and the line connecting the own ship's position; The ship collision avoidance test scenario generation module automatically generates a plurality of different types of ship collision avoidance test scenarios based on the generated scenario initial parameters, the speeds of the own ship and the target ship, the closest approach distance, and the closest approach time.

7. The ship collision avoidance test scenario generation system based on marine collision avoidance rules according to claim 6, characterized in that: The ship encounter situation includes overtaking scenario, head-on encounter scenario and crossing encounter scenario.

8. The ship collision avoidance test scenario generation system based on marine collision avoidance rules according to claim 7, characterized in that: In the scenario type determination and parameter range setting module, dividing the ship encounter situation according to the heading angle and the quantified azimuth of the target ship relative to the own ship includes: dividing the ship encounter situation into port and starboard according to the azimuth of the target ship relative to the own ship, and dividing the intersection encounter scenario into small-angle intersection encounter and large-angle intersection encounter according to the heading angle difference between the own ship and the target ship.

9. The ship collision avoidance test scenario generation system based on marine collision avoidance rules according to claim 6, characterized in that: In the scenario type determination and parameter range setting module, the initial encounter situation of the two ships is also classified according to the distance range between the own ship and the target ship. If the distance between the own ship and the target ship is greater than the first preset distance threshold and less than or equal to the urgent danger distance threshold, the initial encounter situation is regarded as an urgent danger situation; if the distance between the own ship and the target ship is greater than the urgent danger distance threshold and less than or equal to the third preset distance threshold, the initial encounter situation is regarded as an urgent situation; if the distance between the own ship and the target ship is greater than the third preset distance threshold and less than or equal to the fourth preset distance threshold, the initial encounter situation is regarded as a general encounter situation.

10. The ship collision avoidance test scenario generation system based on marine collision avoidance rules according to claim 9, characterized in that: The imminent danger distance threshold is the ship's emergency braking distance plus a 50% safety margin distance, and it is determined whether the ship is sailing in the polar region. If there is floating ice during the ship's polar navigation, the imminent danger distance threshold is the ship's emergency braking distance plus a 100% safety margin distance.

Citation Information

Patent Citations

  • Obstacle collision avoidance method based on dynamic navigation ship domain and collision avoidance rules

    CN112965475A

  • Method for autonomous collision avoidance by evaluating ship navigation collision avoidance compliance

    CN113012475A

  • Self-adaptive generation method of intelligent ship collision avoidance algorithm test scene

    CN113971316A

  • Intelligent ship collision avoidance method and device fusing COLREGs scene knowledge

    CN114550499A

  • Ship collision avoidance test scene generation method and system and storage medium

    CN117057223A

Cited By

  • Group ship collision avoidance route automatic planning method and system based on AIS

    CN119879932A

  • Ship dynamic risk group identification method and system based on spectral clustering

    CN120296608A

  • End-to-end multi-modal large model autonomous decision-making method and system for ship-aided navigation

    CN121209280A