Fishing boat anti-collision early warning method and system based on system safety theory

Abstract

The invention discloses a fishing boat anti-collision early warning method and system based on a system safety theory, and the method comprises the steps: obtaining the navigation monitoring information of a fishing boat, detecting other navigation ships in a current region according to the navigation monitoring information of the fishing boat, and carrying out the situation analysis, and obtaining the situation analysis information; acquiring a ship collision risk assessment instance based on data retrieval to perform ship collision risk assessment element selection, setting a collision risk assessment index and constructing a collision risk assessment model; performing collision risk assessment according to the situation analysis information to obtain collision risk assessment information, performing anti-collision early warning prompt, constructing a collision avoidance responsibility judgment rule, and performing collision avoidance responsibility judgment in combination with the collision risk assessment information; collision avoidance priority judgment is carried out according to the collision risk assessment information, a collision avoidance route making model is constructed based on an SAGA algorithm to make a collision avoidance route, and collision avoidance responsibility judgment and collision avoidance intention analysis are carried out on a next collision avoidance ship. And the anti-collision effectiveness and reliability of the fishing boat are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of anti-collision of fishing boats, and in particular to an anti-collision early warning method and system for fishing boats based on system safety theory. Background Art

[0002] In marine fisheries, fishing vessel collision accidents have always been a major problem affecting the safety of fishing vessel operations. With the increase in the number of ships operating at sea and the increasing complexity of maritime traffic, the risk of collisions between fishing vessels and between fishing vessels and other ships has increased significantly. This not only poses a threat to the safety of the ship itself, but may also cause serious casualties and property losses. Traditional collision avoidance methods mainly rely on the experience and observation of the crew, which are highly subjective and uncertain.

[0003] Therefore, how to effectively prevent fishing vessel collision accidents and improve the safety of marine operations is still an important issue that needs to be urgently addressed in the current field of marine fishery management and ship navigation technology. Summary of the invention

[0004] The present invention overcomes the defects of the prior art and provides a fishing vessel anti-collision warning method and system based on system safety theory, the important purpose of which is to improve the effectiveness and reliability of fishing vessel anti-collision.

[0005] To achieve the above-mentioned purpose, the present invention provides a first aspect of a fishing boat anti-collision warning method based on system safety theory, comprising:

[0006] Obtaining fishing vessel navigation monitoring information, detecting other sailing vessels in the current area according to the fishing vessel navigation monitoring information, and performing situation analysis to obtain situation analysis information;

[0007] Based on data retrieval, ship collision risk assessment examples are obtained to select ship collision risk assessment elements, set collision risk assessment indicators and build a collision risk assessment model;

[0008] Performing a collision risk assessment based on the situation analysis information to obtain collision risk assessment information, and performing an anti-collision warning prompt, constructing a collision avoidance responsibility determination rule, and performing a collision avoidance responsibility determination based on the collision risk assessment information;

[0009] The collision avoidance priority is determined according to the collision risk assessment information, a collision avoidance route planning model is constructed based on the SAGA algorithm to plan the collision avoidance route, and the collision avoidance responsibility and collision avoidance willingness of the next collision avoidance vessel are determined.

[0010] In this solution, the fishing vessel navigation monitoring information is obtained, other sailing vessels in the current area are detected according to the fishing vessel navigation monitoring information, and situation analysis is performed to obtain situation analysis information, which specifically includes:

[0011] Obtaining the navigation monitoring information of the fishing vessel based on the navigation monitoring equipment of the target fishing vessel, wherein the navigation monitoring information of the fishing vessel includes the navigation information of other vessels within the monitoring range of the current navigation area and the navigation information of the target fishing vessel;

[0012] Acquire fishing vessel navigation monitoring image information according to the fishing vessel navigation monitoring information, and perform image preprocessing on the fishing vessel navigation monitoring image information to obtain preprocessed image information;

[0013] Building a target detection model based on YOLOv5s, inputting the preprocessed image information into the target detection model for target detection, extracting features of the preprocessed image through a backbone network, and generating an initial feature map;

[0014] Extracting features of different scales of the preprocessed image information, introducing an attention mechanism to obtain the attention score of each feature, generating an attention feature map, constructing a feature fusion pyramid, and fusing the initial feature map with the attention feature map to obtain a fused feature map;

[0015] Generate a candidate target detection frame according to the fused feature map, use the non-maximum suppression method to screen, extract the confidence of each candidate target detection frame, judge it with a preset threshold, obtain the final target detection frame, and obtain target detection information;

[0016] According to the fishing boat navigation monitoring information, the static navigation data, dynamic navigation data and ship range data of other ships in the area are obtained, and the encounter distance between the target fishing boat and other ships is calculated in combination with the target detection information to obtain situation analysis information;

[0017] The situation analysis information includes target detection information, static navigation data, dynamic navigation data, ship range data and encounter distance of other ships sailing in the area.

[0018] In this solution, the ship collision risk assessment example is obtained based on data retrieval to select ship collision risk assessment elements, set collision risk assessment indicators and build a collision risk assessment model, specifically including:

[0019] Based on data retrieval, ship collision risk assessment instances are obtained, assessment factors of each ship collision risk assessment instance are extracted, and the Pearson correlation coefficient and mutual information value between each assessment factor and the corresponding ship collision risk assessment instance are calculated;

[0020] According to the calculated Pearson correlation coefficient and mutual information value, a Pearson correlation coefficient matrix and a mutual information matrix are respectively constructed, a Pearson correlation coefficient threshold and a mutual information threshold are preset, and threshold processing is performed on the Pearson correlation coefficient matrix and the mutual information matrix;

[0021] After the data less than the threshold is reset to zero, the updated Pearson correlation coefficient matrix and mutual information matrix are obtained, and the corresponding risk assessment factors are extracted and the union operation is performed to obtain the screened risk assessment factors;

[0022] Calculate the similarity between the risk assessment factors after screening according to the risk assessment factors after screening, simplify the risk assessment factors according to the similarity, and set the collision risk assessment index according to the simplified risk assessment factors;

[0023] Through the ship collision risk assessment example, the corresponding assessment data of each collision risk assessment index under different collision risk levels are extracted, and the assessment data range of each collision risk assessment index under different collision risk levels is analyzed;

[0024] A collision risk assessment model is constructed based on the collision risk assessment index and the assessment data interval, and a model training is performed by forming a training data set through the ship collision risk assessment examples to obtain a collision risk assessment model that meets expectations.

[0025] In this solution, the collision risk assessment is performed according to the situation analysis information to obtain collision risk assessment information, and an anti-collision warning prompt is performed, a collision avoidance responsibility determination rule is constructed, and a collision avoidance responsibility determination is performed in combination with the collision risk assessment information, specifically including:

[0026] Acquire situation analysis information, input the situation analysis information into a collision risk assessment model to perform collision risk assessment, and compare the input situation analysis information with the assessment data interval of each collision risk assessment indicator to obtain a risk assessment level of each collision risk assessment indicator;

[0027] According to the risk assessment level of each collision risk assessment indicator, the assessment weight corresponding to each risk assessment level is obtained, and a weighted sum is performed. The weighted sum result is judged with a preset threshold value, and the collision risk between other ships within the monitoring range of the current navigation area and the target fishing boat is assessed to obtain collision risk assessment information;

[0028] Providing anti-collision warning prompts to the target fishing vessel according to the collision risk assessment information;

[0029] The ship navigation encounter situation is divided into direct encounter, crossing encounter and overtaking encounter, and the ship navigation collision avoidance rules are obtained through data retrieval to construct the collision avoidance responsibility determination rules;

[0030] Based on the collision risk assessment information, navigation monitoring information of the vessel defined as the collision vessel is obtained, and the relative position of the collision vessel and the target fishing vessel is analyzed in combination with the navigation monitoring information of the target fishing vessel;

[0031] According to the relative positions of the colliding vessel and the target fishing boat, a collision avoidance liability determination is performed in combination with the collision avoidance liability determination rule to obtain collision avoidance liability determination information.

[0032] In this scheme, the collision avoidance priority is determined according to the collision risk assessment information, a collision avoidance route formulation model is constructed based on the SAGA algorithm to formulate a collision avoidance route, and the collision avoidance responsibility and collision avoidance willingness analysis are performed on the next collision avoidance vessel, specifically including:

[0033] Acquire collision risk assessment information, extract collision risk corresponding to the collision vessels according to the collision risk assessment information, sort the collision vessels, and determine the collision avoidance priority according to the sorting result;

[0034] Obtaining collision avoidance responsibility determination information, and obtaining collision avoidance responsibility determination results between the collision vessel with the highest collision avoidance priority and the target fishing vessel according to the collision avoidance responsibility determination information, and if the collision avoidance responsibility determination result is that the target fishing vessel is responsible for collision avoidance, formulating a collision avoidance route;

[0035] Obtaining fishing vessel navigation monitoring information, extracting navigation information of the collision avoidance vessel based on the fishing vessel navigation monitoring information, predicting the route of the collision avoidance vessel, and obtaining route prediction information;

[0036] Constructing a collision avoidance route formulation model based on the SAGA algorithm, inputting the fishing vessel navigation monitoring information and route prediction information to formulate a collision avoidance route, and obtaining collision avoidance route formulation result information;

[0037] Provide collision avoidance guidance to the target fishing vessel according to the collision avoidance route planning result information, and after completing the collision avoidance event for the highest priority collision vessel, determine the collision avoidance responsibility for the next collision avoidance vessel;

[0038] If the target fishing vessel is still responsible for collision avoidance, collision avoidance will continue according to the collision avoidance route formulation result information; if the target fishing vessel is not responsible for collision avoidance, the next collision avoidance vessel will be defined as the collision avoidance responsible vessel, and the collision avoidance intention will be determined;

[0039] The collision avoidance responsible ship is monitored in real time and the collision risk with the target fishing vessel is assessed in real time. Based on the real-time collision risk assessment results, it is determined whether the collision avoidance responsible ship has the intention to avoid collision. If there is no intention to avoid collision, an active avoidance warning is issued to the driver of the target fishing vessel.

[0040] In this solution, the input of the fishing vessel navigation monitoring information and route prediction information to formulate a collision avoidance route, and obtaining the collision avoidance route formulation result information, specifically includes:

[0041] The collision risk of the collision avoidance path is used as the optimization goal to set the objective function and constraint conditions, and the initial population is randomly generated;

[0042] The penalty function is set by taking the steering amplitude as the penalty condition. The solution that violates the constraint condition in the optimization process is marked by the penalty function, and the violation degree information is obtained. The selection probability is reduced according to the violation degree information, and the high-quality solution that meets the constraint condition is selected.

[0043] A selection threshold is preset, and the fitness value of each individual is calculated based on the generated initial population and the quality solution is selected by comparing with the selection threshold. A crossover and mutation operation is performed based on the obtained quality solution to generate a new population as the input of the SA algorithm, and whether the new solution is accepted as the current solution is determined based on the Metropolis criterion;

[0044] The life span mechanism and extinction mechanism are introduced to construct an iterative optimization strategy. The expected life span value of each individual is set according to the individual fitness value for iterative optimization. When the individual age value is greater than the corresponding expected life span value, the individual is deleted and the retained individual is obtained. Based on the retained individual, several individuals are randomly selected for replication to supplement the population.

[0045] Based on the fitness value of the target population, determine whether the target population has changed during the iteration process. If not, the target population is exterminated and the individuals in the target population are sorted according to the individual fitness values. The surviving individuals are selected according to the sorting results and a new population is constructed.

[0046] After iterative optimization converges, the final solution is output. Candidate collision avoidance paths are generated based on the final solution. The collision avoidance distance characteristics corresponding to each candidate collision avoidance path are extracted to calculate the collision avoidance consumption of each collision avoidance path, and the paths are sorted. The collision avoidance route planning result information is generated based on the sorting results.

[0047] The second aspect of the present invention provides a fishing boat anti-collision warning system based on system safety theory, the system comprising: a memory, a processor, the memory containing a fishing boat anti-collision warning method program based on system safety theory, the fishing boat anti-collision warning method program based on system safety theory when executed by the processor implements the following steps:

[0048] Obtaining fishing vessel navigation monitoring information, detecting other sailing vessels in the current area according to the fishing vessel navigation monitoring information, and performing situation analysis to obtain situation analysis information;

[0049] Based on data retrieval, ship collision risk assessment examples are obtained to select ship collision risk assessment elements, set collision risk assessment indicators and build a collision risk assessment model;

[0050] Performing a collision risk assessment based on the situation analysis information to obtain collision risk assessment information, and performing an anti-collision warning prompt, constructing a collision avoidance responsibility determination rule, and performing a collision avoidance responsibility determination based on the collision risk assessment information;

[0051] The collision avoidance priority is determined according to the collision risk assessment information, a collision avoidance route planning model is constructed based on the SAGA algorithm to plan the collision avoidance route, and the collision avoidance responsibility and collision avoidance willingness of the next collision avoidance vessel are determined.

[0052] The invention discloses a fishing boat anti-collision warning method and system based on system safety theory, including: obtaining fishing boat navigation monitoring information, detecting other sailing ships in the current area according to the fishing boat navigation monitoring information, and performing situation analysis to obtain situation analysis information; obtaining ship collision risk assessment examples based on data retrieval to select ship collision risk assessment elements, setting collision risk assessment indicators and building a collision risk assessment model; performing collision risk assessment according to situation analysis information to obtain collision risk assessment information, and performing anti-collision warning prompts, building collision avoidance responsibility determination rules, and performing collision avoidance responsibility determination in combination with the collision risk assessment information; performing collision avoidance priority determination according to the collision risk assessment information, constructing a collision avoidance route formulation model based on the SAGA algorithm to formulate a collision avoidance route, and performing collision avoidance responsibility determination and collision avoidance willingness analysis on the next collision avoidance ship. The effectiveness and reliability of fishing boat anti-collision are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the embodiments or exemplary embodiments of the present invention, the drawings required for use in the embodiments or exemplary descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained according to the drawings without paying creative work.

[0054] Figure 1 A flow chart of a fishing boat anti-collision warning method based on system safety theory provided in one embodiment of the present invention;

[0055] Figure 2 A flowchart of a fishing vessel collision avoidance analysis based on system safety theory provided by an embodiment of the present invention;

[0056] Figure 3 A block diagram of a fishing vessel anti-collision warning system based on system safety theory provided by an embodiment of the present invention;

[0057] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0058] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0059] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0060] Figure 1 A flow chart of a fishing boat anti-collision warning method based on system safety theory provided in one embodiment of the present invention;

[0061] like Figure 1 As shown, the present invention provides a flow chart of a fishing vessel anti-collision warning method based on system safety theory, including:

[0062] S102, obtaining fishing vessel navigation monitoring information, detecting other sailing vessels in the current area according to the fishing vessel navigation monitoring information, and performing situation analysis to obtain situation analysis information;

[0063] S104, obtaining a ship collision risk assessment instance based on data retrieval to select ship collision risk assessment elements, set collision risk assessment indicators and build a collision risk assessment model;

[0064] S106, performing a collision risk assessment based on the situation analysis information to obtain collision risk assessment information, and performing an anti-collision warning prompt, constructing a collision avoidance responsibility determination rule, and performing a collision avoidance responsibility determination based on the collision risk assessment information;

[0065] S110, performing collision avoidance priority judgment based on the collision risk assessment information, constructing a collision avoidance route formulation model based on the SAGA algorithm to formulate a collision avoidance route, and performing collision avoidance responsibility judgment and collision avoidance willingness analysis on the next collision avoidance vessel.

[0066] It should be noted that the present invention provides a method and system for anti-collision warning of fishing vessels based on system safety theory, obtains navigation monitoring information of fishing vessels based on the navigation monitoring equipment of the target fishing vessel, uses YOLOv5s to build a target detection model for target detection, and then identifies other vessels in the current navigation area. Further situation analysis is performed on the target fishing vessel, the encounter distance between other vessels and the target fishing vessel is calculated, and situation analysis information is formed in combination with the navigation monitoring information. By constructing a collision risk assessment model, the collision risk between other vessels and the target fishing vessel is assessed, and anti-collision warning prompts are given. At the same time, a collision avoidance responsibility determination rule is constructed to determine the collision avoidance responsibility. After the collision avoidance responsibility is determined, the collision avoidance priority is determined, and the collision avoidance route formulation model is used to formulate the collision avoidance route, thereby providing a collision avoidance reference. The collision avoidance responsibility determination and collision avoidance willingness analysis are performed on the next collision avoidance vessel, so as to avoid the collision avoidance willingness of the next vessel when multiple vessels are avoiding collision, avoid accidents, and effectively improve the safety and efficiency of maritime navigation.

[0067] Further, in a preferred embodiment of the present invention, the obtaining of fishing vessel navigation monitoring information, detecting other sailing vessels in the current area according to the fishing vessel navigation monitoring information, and performing situation analysis to obtain situation analysis information specifically includes:

[0068] Obtaining the navigation monitoring information of the fishing vessel based on the navigation monitoring equipment of the target fishing vessel, wherein the navigation monitoring information of the fishing vessel includes the navigation information of other vessels within the monitoring range of the current navigation area and the navigation information of the target fishing vessel;

[0069] Acquire fishing vessel navigation monitoring image information according to the fishing vessel navigation monitoring information, and perform image preprocessing on the fishing vessel navigation monitoring image information to obtain preprocessed image information;

[0070] Building a target detection model based on YOLOv5s, inputting the preprocessed image information into the target detection model for target detection, extracting features of the preprocessed image through a backbone network, and generating an initial feature map;

[0071] Extracting features of different scales of the preprocessed image information, introducing an attention mechanism to obtain the attention score of each feature, generating an attention feature map, constructing a feature fusion pyramid, and fusing the initial feature map with the attention feature map to obtain a fused feature map;

[0072] Generate a candidate target detection frame according to the fused feature map, use the non-maximum suppression method to screen, extract the confidence of each candidate target detection frame, judge it with a preset threshold, obtain the final target detection frame, and obtain target detection information;

[0073] According to the fishing boat navigation monitoring information, the static navigation data, dynamic navigation data and ship range data of other ships in the area are obtained, and the encounter distance between the target fishing boat and other ships is calculated in combination with the target detection information to obtain situation analysis information;

[0074] The situation analysis information includes target detection information, static navigation data, dynamic navigation data, ship range data and encounter distance of other ships sailing in the area.

[0075] It should be noted that the navigation monitoring equipment installed on the target fishing boat is used to collect the navigation data of the fishing boat in real time, including the navigation data of the target fishing boat itself, such as position, speed, heading, etc., and also covers the navigation data of other ships within the monitoring range of the current navigation area. Next, the positions of other ships and obstacles in the current navigation area are determined, and the target detection model is constructed using YOLOv5s for target detection, so as to further improve the accuracy of the detection results of the navigation monitoring equipment. Finally, according to the navigation monitoring information of the fishing boat and the target detection information, the encounter distance between the target fishing boat and other ships is calculated, and the situation analysis information is generated by combining the navigation static data, navigation dynamic data and ship voyage data of other sailing ships in the area. Among them, the encounter distance refers to the distance between the closest points predicted by the two ships based on their respective speeds and headings during the navigation process. The navigation static data includes the type of ship, basic information of the ship, etc., the navigation dynamic data includes longitude, latitude, bow direction, ship track direction, ship speed, etc., and the navigation voyage data includes ship status, ship draft depth, etc. It fully reflects the ship information in the navigation area.

[0076] Further, in a preferred embodiment of the present invention, the process of obtaining a ship collision risk assessment instance based on data retrieval to select ship collision risk assessment elements, setting collision risk assessment indicators and constructing a collision risk assessment model specifically includes:

[0077] Based on data retrieval, ship collision risk assessment instances are obtained, assessment factors of each ship collision risk assessment instance are extracted, and the Pearson correlation coefficient and mutual information value between each assessment factor and the corresponding ship collision risk assessment instance are calculated;

[0078] According to the calculated Pearson correlation coefficient and mutual information value, a Pearson correlation coefficient matrix and a mutual information matrix are respectively constructed, a Pearson correlation coefficient threshold and a mutual information threshold are preset, and threshold processing is performed on the Pearson correlation coefficient matrix and the mutual information matrix;

[0079] After the data less than the threshold is reset to zero, the updated Pearson correlation coefficient matrix and mutual information matrix are obtained, and the corresponding risk assessment factors are extracted and the union operation is performed to obtain the screened risk assessment factors;

[0080] Calculate the similarity between the risk assessment factors after screening according to the risk assessment factors after screening, simplify the risk assessment factors according to the similarity, and set the collision risk assessment index according to the simplified risk assessment factors;

[0081] Through the ship collision risk assessment example, the corresponding assessment data of each collision risk assessment index under different collision risk levels are extracted, and the assessment data range of each collision risk assessment index under different collision risk levels is analyzed;

[0082] A collision risk assessment model is constructed based on the collision risk assessment index and the assessment data interval, and a model training is performed by forming a training data set through the ship collision risk assessment examples to obtain a collision risk assessment model that meets expectations.

[0083] It should be noted that based on data retrieval, ship collision risk assessment instances are obtained, and assessment elements of each ship collision risk assessment instance are extracted, including the ship's speed, heading, position, environmental conditions, etc. Next, the Pearson correlation coefficient and mutual information value of each assessment element and the corresponding ship collision risk assessment instance are calculated to measure the correlation between the assessment elements. According to the calculated Pearson correlation coefficient and mutual information value, the Pearson correlation coefficient matrix and the mutual information matrix are constructed respectively. The Pearson correlation coefficient threshold and the mutual information threshold are preset, and the mutual information threshold can be calculated by the method of predicting the minimum description length, and the Pearson correlation coefficient threshold can be calculated by the breakpoint detection method. The Pearson correlation coefficient matrix and the mutual information matrix are threshold processed, and the data less than the threshold is set to zero to obtain the updated Pearson correlation coefficient matrix and the mutual information matrix. After obtaining the updated matrix, the corresponding risk assessment elements are extracted for union operation to obtain the screened risk assessment elements, and the assessment elements that have a significant impact on the collision risk are retained. Next, the similarity between the elements is calculated based on the screened risk assessment elements, and the risk assessment elements are simplified by the similarity, thereby reducing redundant information. The collision risk assessment index is set according to the streamlined risk assessment factors. Then, by analyzing the ship collision risk assessment examples, the assessment data corresponding to each collision risk assessment index at different collision risk levels are extracted, and the distribution and interval of these data at different risk levels are further analyzed. Based on the collision risk assessment index and the assessment data interval, a collision risk assessment model is constructed. The ship collision risk assessment examples are used to form a training data set, and the model is trained to ensure that the model can accurately reflect the actual situation, and finally a collision risk assessment model that meets the expectations is obtained.

[0084] Further, in a preferred embodiment of the present invention, the collision risk assessment is performed according to the situation analysis information to obtain collision risk assessment information, and an anti-collision warning prompt is performed, and a collision avoidance responsibility determination rule is constructed, and the collision avoidance responsibility determination is performed in combination with the collision risk assessment information, specifically including:

[0085] Acquire situation analysis information, input the situation analysis information into a collision risk assessment model to perform collision risk assessment, and compare the input situation analysis information with the assessment data interval of each collision risk assessment indicator to obtain a risk assessment level of each collision risk assessment indicator;

[0086] According to the risk assessment level of each collision risk assessment indicator, the assessment weight corresponding to each risk assessment level is obtained, and a weighted sum is performed. The weighted sum result is judged with a preset threshold value, and the collision risk between other ships within the monitoring range of the current navigation area and the target fishing boat is assessed to obtain collision risk assessment information;

[0087] Providing anti-collision warning prompts to the target fishing vessel according to the collision risk assessment information;

[0088] The ship navigation encounter situation is divided into direct encounter, crossing encounter and overtaking encounter, and the ship navigation collision avoidance rules are obtained through data retrieval to construct the collision avoidance responsibility determination rules;

[0089] Based on the collision risk assessment information, navigation monitoring information of the vessel defined as the collision vessel is obtained, and the relative position of the collision vessel and the target fishing vessel is analyzed in combination with the navigation monitoring information of the target fishing vessel;

[0090] According to the relative positions of the colliding vessel and the target fishing boat, a collision avoidance liability determination is performed in combination with the collision avoidance liability determination rule to obtain collision avoidance liability determination information.

[0091] It should be noted that based on the situation analysis information, the collision risk assessment is performed through the constructed collision risk assessment model, and the preset collision risk assessment indicators and their corresponding assessment data intervals are used to divide the risk level of each assessment indicator. Next, according to the risk assessment level obtained by each collision risk assessment indicator, the corresponding assessment weight is found and applied. Through the weighted summation method, a comprehensive collision risk score is obtained, which is compared with the preset threshold value to assess the risk level of the target fishing vessel colliding with other ships within the monitoring range in the current navigation area, and generate corresponding collision risk assessment information. Based on the collision risk assessment information, an anti-collision warning prompt is immediately issued to the target fishing vessel to remind the fishing vessel operator to pay attention to potential dangers and take corresponding preventive measures. At the same time, the navigation encounter situation of the ship is subdivided into three situations: direct encounter, crossing encounter and overtaking encounter, and the corresponding ship navigation collision avoidance rules are obtained with the help of data retrieval to construct collision avoidance responsibility determination rules, such as the International Regulations for Preventing Collisions at Sea. At the same time, the target fishing boat's forward direction is used as a reference to divide the encounter situation into four areas, namely area A (355°, 5°), area B (5°, 112.5°), area C (112.5°, 247.5°), and area D (247.5°, 355°). Area A corresponds to direct encounters, areas B and D correspond to cross encounters, and area C corresponds to overtaking encounters. When a possible collision hazard is detected, the navigation monitoring information of the vessel defined as the collision vessel is further obtained and compared with the navigation monitoring information of the target fishing vessel to determine the relative position relationship between the two. Finally, according to the relative position of the collision vessel and the target fishing vessel, combined with the constructed collision avoidance responsibility determination rules, the collision avoidance responsibility is determined, and the collision avoidance responsibility determination information is generated. This ensures the safety and order of navigation avoidance of fishing vessels.

[0092] Furthermore, in a preferred embodiment of the present invention, the collision avoidance priority is determined according to the collision risk assessment information, a collision avoidance route formulation model is constructed based on the SAGA algorithm to formulate a collision avoidance route, and a collision avoidance responsibility determination and collision avoidance willingness analysis are performed on the next collision avoidance vessel, specifically including:

[0093] Acquire collision risk assessment information, extract collision risk corresponding to the collision vessels according to the collision risk assessment information, sort the collision vessels, and determine the collision avoidance priority according to the sorting result;

[0094] Obtaining collision avoidance responsibility determination information, and obtaining a collision avoidance responsibility determination result between the collision vessel with the highest collision avoidance priority and the target fishing vessel according to the collision avoidance responsibility determination information, and if the collision avoidance responsibility determination result is that the target fishing vessel is responsible for collision avoidance, formulating a collision avoidance route;

[0095] Obtaining fishing vessel navigation monitoring information, extracting navigation information of the collision avoidance vessel based on the fishing vessel navigation monitoring information, predicting the route of the collision avoidance vessel, and obtaining route prediction information;

[0096] Constructing a collision avoidance route formulation model based on the SAGA algorithm, inputting the fishing vessel navigation monitoring information and route prediction information to formulate a collision avoidance route, and obtaining collision avoidance route formulation result information;

[0097] Provide collision avoidance guidance to the target fishing vessel according to the collision avoidance route planning result information, and after completing the collision avoidance event for the highest priority collision vessel, determine the collision avoidance responsibility for the next collision avoidance vessel;

[0098] If the target fishing vessel is still responsible for collision avoidance, collision avoidance will continue according to the collision avoidance route formulation result information; if the target fishing vessel is not responsible for collision avoidance, the next collision avoidance vessel will be defined as the collision avoidance responsible vessel, and the collision avoidance intention will be determined;

[0099] The collision avoidance responsible ship is monitored in real time and the collision risk with the target fishing vessel is assessed in real time. Based on the real-time collision risk assessment results, it is determined whether the collision avoidance responsible ship has the intention to avoid collision. If there is no intention to avoid collision, an active avoidance warning is issued to the driver of the target fishing vessel.

[0100] It should be noted that in the navigation collision avoidance scenario, there are not only single-ship collision avoidance, but also multi-ship collision avoidance scenarios. For single-ship collision avoidance, it is only necessary to execute the avoidance path after evaluating the collision risk and avoidance responsibility. For multi-ship collision avoidance, it is necessary to first consider the risk between the collision avoidance ship and the collision avoidance ship, so as to analyze the collision avoidance priority, that is, give priority to ships with high collision risk. At this time, the multi-ship collision avoidance event is analyzed according to the priority of the collision avoidance ship. After analyzing the ship with the highest collision avoidance priority, the end point of the previous collision avoidance path is used as the starting point of the next collision avoidance ship for collision avoidance path analysis. At this time, when it is confirmed that our ship is still a collision avoidance ship, our ship faces two choices, namely, re-voyage to avoid collision or not re-voyage to avoid collision. At this time, two types of collision avoidance paths are generated, and the optimal collision avoidance path is selected based on the principles of feasibility and economy. After the collision avoidance of the ship with the highest collision avoidance priority is completed, the collision avoidance responsibility corresponding to the next collision avoidance event is not our ship. At this time, a collision avoidance willingness analysis is performed to analyze whether it has the willingness to give way. If not, active collision avoidance is required to avoid safety accidents.

[0101] Further, in a preferred embodiment of the present invention, the input of the fishing vessel navigation monitoring information and the route prediction information to formulate a collision avoidance route, and obtaining the collision avoidance route formulation result information, specifically includes:

[0102] The collision risk of the collision avoidance path is used as the optimization goal to set the objective function and constraint conditions, and the initial population is randomly generated;

[0103] The penalty function is set by taking the steering amplitude as the penalty condition. The solution that violates the constraint condition in the optimization process is marked by the penalty function, and the violation degree information is obtained. The selection probability is reduced according to the violation degree information, and the high-quality solution that meets the constraint condition is selected.

[0104] A selection threshold is preset, and the fitness value of each individual is calculated based on the generated initial population and the quality solution is selected by comparing with the selection threshold. A crossover and mutation operation is performed based on the obtained quality solution to generate a new population as the input of the SA algorithm, and whether the new solution is accepted as the current solution is determined based on the Metropolis criterion;

[0105] The life span mechanism and extinction mechanism are introduced to construct an iterative optimization strategy. The expected life span value of each individual is set according to the individual fitness value for iterative optimization. When the individual age value is greater than the corresponding expected life span value, the individual is deleted and the retained individual is obtained. Based on the retained individual, several individuals are randomly selected for replication to supplement the population.

[0106] Based on the fitness value of the target population, determine whether the target population has changed during the iteration process. If not, the target population is exterminated and the individuals in the target population are sorted according to the individual fitness values. The surviving individuals are selected according to the sorting results and a new population is constructed.

[0107] After iterative optimization converges, the final solution is output. Candidate collision avoidance paths are generated based on the final solution. The collision avoidance distance characteristics corresponding to each candidate collision avoidance path are extracted to calculate the collision avoidance consumption of each collision avoidance path, and the paths are sorted. The collision avoidance route planning result information is generated based on the sorting results.

[0108] It should be noted that the collision avoidance route is formulated by the SAGA algorithm (simulated annealing genetic algorithm), which combines the global search capability of the genetic algorithm and the local search capability of the simulated annealing algorithm to achieve an effective balance in the search space. The genetic algorithm is responsible for extensively searching in the solution space to find potential high-quality areas; while the simulated annealing algorithm conducts a fine search in these areas to find the optimal solution or a near-optimal solution. In the ship avoidance scenario, the turning amplitude represents the feasibility of the avoidance operation. Based on the statistical analysis of the actual ship avoidance operation, when avoiding, the crew generally does not take a turning amplitude of more than 50° to avoid. Therefore, the turning amplitude in the open sea can be limited to between plus or minus 50°. At the same time, it can be set to between plus or minus 30° in narrow waterways, so that the penalty function is population-restricted, so that the formulation of the collision avoidance route meets the actual feasibility. For the iterative optimization strategy constructed with the lifespan mechanism and the extinction mechanism, first, the lifespan mechanism sets the corresponding age for the individuals in the initialization population when the population is initialized, and sets the expected lifespan value according to the corresponding fitness value. In the process of iterative optimization, the age and the expected lifespan value are judged to analyze whether there are "aging individuals", so as to delete individuals and then iterate based on the best. At the same time, based on the existence of the lifespan mechanism, there may be a problem that the fitness of the population does not change during the iteration process, which means that the corresponding population may be trapped in a local extreme value. At this time, the extinction mechanism is used to "extinct" the population, and the individuals with better fitness are retained based on the individual fitness values ​​in the target population to construct a new population, retaining the diversity and randomness of the population. Through the interaction of the lifespan mechanism and the extinction mechanism, the accuracy and effectiveness of collision avoidance route planning are improved.

[0109] Figure 2 A flowchart of a fishing vessel collision avoidance analysis based on system safety theory provided by an embodiment of the present invention;

[0110] like Figure 2 As shown, the present invention provides a fishing vessel collision avoidance analysis flow chart based on system safety theory, including:

[0111] S202, obtaining fishing vessel navigation monitoring information based on the navigation monitoring equipment of the target fishing vessel, performing situation analysis on the target navigation area, and obtaining situation analysis information;

[0112] S204, performing a collision risk assessment according to the situation analysis information to obtain collision risk assessment information, performing an anti-collision warning prompt, and performing a collision avoidance responsibility determination according to a collision avoidance responsibility determination rule;

[0113] S206, determining the avoidance priority according to the collision risk assessment information, formulating a collision avoidance route, and determining the collision avoidance responsibility and collision avoidance willingness of the next collision avoidance vessel;

[0114] S208, judging whether the next collision avoidance vessel has the avoidance intention based on the collision avoidance intention analysis result, and if not, issuing an active avoidance warning and formulating an avoidance route.

[0115] In addition, in a fishing boat anti-collision warning method based on system safety theory provided by the present invention, the following steps are also included:

[0116] Acquiring real-time ship maneuvering information when performing a ship avoidance scenario, and performing real-time maneuvering route prediction based on the real-time ship maneuvering information to obtain real-time maneuvering route prediction information;

[0117] Obtaining fishing vessel navigation monitoring information, and predicting a navigation route of the vessel to be avoided based on the fishing vessel navigation monitoring information to obtain first prediction information;

[0118] Performing a collision risk assessment in combination with the real-time operation route prediction information and the first prediction information to obtain real-time collision risk assessment information, and comparing the real-time collision risk assessment information with a preset threshold to analyze whether there is a collision risk;

[0119] If there is a risk of collision, the real-time collision avoidance route information is obtained, and a difference analysis is performed with the real-time maneuvering route prediction information. Correction parameters are generated based on the difference analysis results to provide collision warning prompts and maneuvering correction assistance to the crew.

[0120] It should be noted that when ships are avoiding collisions, there are objective and subjective factors that cause the ship's collision avoidance route to differ from the preset route. At this time, real-time ship route prediction is performed to determine whether there is still a risk of collision, and correction parameters are generated based on the difference between the preset collision avoidance route and the actual route to assist the operator in maneuvering and avoid safety accidents.

[0121] Figure 3 A fishing boat anti-collision warning system 3 based on system safety theory is provided in one embodiment of the present invention. The system comprises: a memory 31 and a processor 32. The memory 31 contains a fishing boat anti-collision warning method program based on system safety theory. When the fishing boat anti-collision warning method program based on system safety theory is executed by the processor 32, the following steps are implemented:

[0122] Obtaining fishing vessel navigation monitoring information, detecting other sailing vessels in the current area according to the fishing vessel navigation monitoring information, and performing situation analysis to obtain situation analysis information;

[0123] Based on data retrieval, ship collision risk assessment examples are obtained to select ship collision risk assessment elements, set collision risk assessment indicators and build a collision risk assessment model;

[0124] Performing a collision risk assessment based on the situation analysis information to obtain collision risk assessment information, and performing an anti-collision warning prompt, constructing a collision avoidance responsibility determination rule, and performing a collision avoidance responsibility determination based on the collision risk assessment information;

[0125] The collision avoidance priority is determined according to the collision risk assessment information, a collision avoidance route planning model is constructed based on the SAGA algorithm to plan the collision avoidance route, and the collision avoidance responsibility and collision avoidance willingness of the next collision avoidance vessel are determined.

[0126] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0127] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0128] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0129] Those skilled in the art can understand that: all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), disks or optical disks, and other media that can store program codes.

[0130] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0131] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A fishing vessel anti-collision warning method based on system safety theory, characterized in that: include: Obtaining fishing vessel navigation monitoring information, detecting other sailing vessels in the current area according to the fishing vessel navigation monitoring information, and performing situation analysis to obtain situation analysis information; Based on data retrieval, ship collision risk assessment examples are obtained to select ship collision risk assessment elements, set collision risk assessment indicators and build a collision risk assessment model; Performing a collision risk assessment based on the situation analysis information to obtain collision risk assessment information, and performing an anti-collision warning prompt, constructing a collision avoidance responsibility determination rule, and performing a collision avoidance responsibility determination based on the collision risk assessment information; The collision avoidance priority is determined according to the collision risk assessment information, a collision avoidance route planning model is constructed based on the SAGA algorithm to plan the collision avoidance route, and the collision avoidance responsibility and collision avoidance willingness of the next collision avoidance vessel are determined.

2. A fishing vessel anti-collision warning method based on system safety theory according to claim 1, characterized in that: The obtaining of the fishing vessel navigation monitoring information, detecting other sailing vessels in the current area according to the fishing vessel navigation monitoring information, and performing situation analysis to obtain situation analysis information specifically includes: Obtaining the navigation monitoring information of the fishing vessel based on the navigation monitoring equipment of the target fishing vessel, wherein the navigation monitoring information of the fishing vessel includes the navigation information of other vessels within the monitoring range of the current navigation area and the navigation information of the target fishing vessel; Acquire fishing vessel navigation monitoring image information according to the fishing vessel navigation monitoring information, and perform image preprocessing on the fishing vessel navigation monitoring image information to obtain preprocessed image information; Building a target detection model based on YOLOv5s, inputting the preprocessed image information into the target detection model for target detection, extracting features of the preprocessed image through a backbone network, and generating an initial feature map; Extracting features of different scales of the preprocessed image information, introducing an attention mechanism to obtain the attention score of each feature, generating an attention feature map, constructing a feature fusion pyramid, and fusing the initial feature map with the attention feature map to obtain a fused feature map; Generate a candidate target detection frame according to the fused feature map, use the non-maximum suppression method to screen, extract the confidence of each candidate target detection frame, judge it with a preset threshold, obtain the final target detection frame, and obtain target detection information; According to the fishing boat navigation monitoring information, the static navigation data, dynamic navigation data and ship range data of other ships in the area are obtained, and the encounter distance between the target fishing boat and other ships is calculated in combination with the target detection information to obtain situation analysis information; The situation analysis information includes target detection information, static navigation data, dynamic navigation data, ship range data and encounter distance of other ships sailing in the area.

3. A fishing vessel anti-collision warning method based on system safety theory according to claim 1, characterized in that: The method of obtaining a ship collision risk assessment instance based on data retrieval to select ship collision risk assessment elements, setting collision risk assessment indicators and constructing a collision risk assessment model specifically includes: Based on data retrieval, ship collision risk assessment instances are obtained, assessment factors of each ship collision risk assessment instance are extracted, and the Pearson correlation coefficient and mutual information value between each assessment factor and the corresponding ship collision risk assessment instance are calculated; According to the calculated Pearson correlation coefficient and mutual information value, a Pearson correlation coefficient matrix and a mutual information matrix are respectively constructed, a Pearson correlation coefficient threshold and a mutual information threshold are preset, and threshold processing is performed on the Pearson correlation coefficient matrix and the mutual information matrix; After the data less than the threshold is reset to zero, the updated Pearson correlation coefficient matrix and mutual information matrix are obtained, and the corresponding risk assessment factors are extracted and the union operation is performed to obtain the screened risk assessment factors; Calculate the similarity between the risk assessment factors after screening according to the risk assessment factors after screening, simplify the risk assessment factors according to the similarity, and set the collision risk assessment index according to the simplified risk assessment factors; Through the ship collision risk assessment example, the corresponding assessment data of each collision risk assessment index under different collision risk levels are extracted, and the assessment data range of each collision risk assessment index under different collision risk levels is analyzed; A collision risk assessment model is constructed based on the collision risk assessment index and the assessment data interval, and a model training is performed by forming a training data set through the ship collision risk assessment examples to obtain a collision risk assessment model that meets expectations.

4. A fishing vessel anti-collision warning method based on system safety theory according to claim 1, characterized in that: The performing of collision risk assessment according to the situation analysis information to obtain collision risk assessment information, and performing anti-collision warning prompts, constructing collision avoidance responsibility determination rules, and performing collision avoidance responsibility determination in combination with the collision risk assessment information specifically includes: Acquire situation analysis information, input the situation analysis information into a collision risk assessment model to perform collision risk assessment, and compare the input situation analysis information with the assessment data interval of each collision risk assessment indicator to obtain a risk assessment level of each collision risk assessment indicator; According to the risk assessment level of each collision risk assessment indicator, the assessment weight corresponding to each risk assessment level is obtained, and a weighted sum is performed. The weighted sum result is judged with a preset threshold value, and the collision risk between other ships within the monitoring range of the current navigation area and the target fishing boat is assessed to obtain collision risk assessment information; Providing anti-collision warning prompts to the target fishing vessel according to the collision risk assessment information; The ship navigation encounter situation is divided into direct encounter, crossing encounter and overtaking encounter, and the ship navigation collision avoidance rules are obtained through data retrieval to construct the collision avoidance responsibility determination rules; Based on the collision risk assessment information, navigation monitoring information of the vessel defined as the collision vessel is obtained, and the relative position of the collision vessel and the target fishing vessel is analyzed in combination with the navigation monitoring information of the target fishing vessel; According to the relative positions of the colliding vessel and the target fishing boat, a collision avoidance liability determination is performed in combination with the collision avoidance liability determination rule to obtain collision avoidance liability determination information.

5. The fishing vessel anti-collision warning method based on system safety theory according to claim 1 is characterized in that: The collision avoidance priority is determined according to the collision risk assessment information, a collision avoidance route formulation model is constructed based on the SAGA algorithm to formulate a collision avoidance route, and a collision avoidance responsibility determination and collision avoidance willingness analysis are performed on the next collision avoidance vessel, specifically including: Acquire collision risk assessment information, extract collision risk corresponding to the collision vessels according to the collision risk assessment information, sort the collision vessels, and determine the collision avoidance priority according to the sorting result; Obtaining collision avoidance responsibility determination information, and obtaining a collision avoidance responsibility determination result between the collision vessel with the highest collision avoidance priority and the target fishing vessel according to the collision avoidance responsibility determination information, and if the collision avoidance responsibility determination result is that the target fishing vessel is responsible for collision avoidance, formulating a collision avoidance route; Obtaining fishing vessel navigation monitoring information, extracting navigation information of the collision avoidance vessel based on the fishing vessel navigation monitoring information, predicting the route of the collision avoidance vessel, and obtaining route prediction information; Constructing a collision avoidance route formulation model based on the SAGA algorithm, inputting the fishing vessel navigation monitoring information and route prediction information to formulate a collision avoidance route, and obtaining collision avoidance route formulation result information; Provide collision avoidance guidance to the target fishing vessel according to the collision avoidance route planning result information, and after completing the collision avoidance event for the highest priority collision vessel, determine the collision avoidance responsibility for the next collision avoidance vessel; If the target fishing vessel is still responsible for collision avoidance, collision avoidance will continue according to the collision avoidance route planning result information; if the target fishing vessel is not responsible for collision avoidance, the next collision avoidance vessel will be defined as the collision avoidance responsible vessel, and collision avoidance intention will be determined; The collision avoidance responsible ship is monitored in real time and the collision risk with the target fishing vessel is assessed in real time. Based on the real-time collision risk assessment results, it is determined whether the collision avoidance responsible ship has the intention to avoid collision. If there is no intention to avoid collision, an active avoidance warning is issued to the driver of the target fishing vessel.

6. A fishing boat anti-collision warning method based on system safety theory according to claim 5, characterized in that: The inputting of the fishing vessel navigation monitoring information and the route prediction information to formulate a collision avoidance route, and obtaining the collision avoidance route formulation result information specifically includes: The collision risk of the collision avoidance path is used as the optimization goal to set the objective function and constraint conditions, and the initial population is randomly generated; The penalty function is set by taking the steering amplitude as the penalty condition. The solution that violates the constraint condition in the optimization process is marked by the penalty function, and the violation degree information is obtained. The selection probability is reduced according to the violation degree information, and the high-quality solution that meets the constraint condition is selected. A selection threshold is preset, and the fitness value of each individual is calculated based on the generated initial population and the quality solution is selected by comparing with the selection threshold. A crossover and mutation operation is performed based on the obtained quality solution to generate a new population as the input of the SA algorithm, and whether the new solution is accepted as the current solution is determined based on the Metropolis criterion; The life span mechanism and extinction mechanism are introduced to construct an iterative optimization strategy. The expected life span value of each individual is set according to the individual fitness value for iterative optimization. When the individual age value is greater than the corresponding expected life span value, the individual is deleted and the retained individual is obtained. Based on the retained individual, several individuals are randomly selected for replication to supplement the population. Based on the fitness value of the target population, determine whether the target population has changed during the iteration process. If not, the target population is exterminated and the individuals in the target population are sorted according to the individual fitness values. The surviving individuals are selected according to the sorting results and a new population is constructed. After iterative optimization converges, the final solution is output. Candidate collision avoidance paths are generated based on the final solution. The collision avoidance distance characteristics corresponding to each candidate collision avoidance path are extracted to calculate the collision avoidance consumption of each collision avoidance path, and the paths are sorted. The collision avoidance route planning result information is generated based on the sorting results.

7. A fishing vessel anti-collision warning system based on system safety theory, characterized in that: The system comprises: a memory and a processor, wherein the memory contains a fishing vessel anti-collision warning method program based on system safety theory, and when the fishing vessel anti-collision warning method program based on system safety theory is executed by the processor, the following steps are implemented: Obtaining fishing vessel navigation monitoring information, detecting other sailing vessels in the current area according to the fishing vessel navigation monitoring information, and performing situation analysis to obtain situation analysis information; Based on data retrieval, ship collision risk assessment examples are obtained to select ship collision risk assessment elements, set collision risk assessment indicators and build a collision risk assessment model; Performing a collision risk assessment based on the situation analysis information to obtain collision risk assessment information, and performing an anti-collision warning prompt, constructing a collision avoidance responsibility determination rule, and performing a collision avoidance responsibility determination based on the collision risk assessment information; The collision avoidance priority is determined according to the collision risk assessment information, a collision avoidance route planning model is constructed based on the SAGA algorithm to plan the collision avoidance route, and the collision avoidance responsibility and collision avoidance willingness of the next collision avoidance vessel are determined.

Images