Ship navigation obstacle avoidance control method and system based on artificial intelligence

Abstract

The invention provides a ship navigation obstacle avoidance control method and system based on artificial intelligence. According to the technical scheme, the method comprises the steps that a monitoring area is set according to ship data of a target ship; obstacle parameter data and obstacle type data of obstacles in the monitoring area are obtained through analysis; analyzing to obtain an initial obstacle avoidance area and a benchmarking obstacle avoidance area of the obstacle, and obtaining a total obstacle avoidance area of the monitoring area and an obstacle avoidance coefficient of the total obstacle avoidance area according to the benchmarking obstacle avoidance area corresponding to the obstacle in the monitoring area; generating a sailing route set of the target ship according to the ship initial place and the ship target place of the target ship, and analyzing to obtain a route safety coefficient and a route comprehensive coefficient of sailing routes in the sailing route set; and obtaining a target navigation path of the target ship according to the path comprehensive coefficient, and performing navigation control on the target ship according to the target navigation path. According to the invention, accurate obstacle avoidance of ship navigation is realized, and optimal control of ship navigation paths is also realized.

Description

Technical Field

[0001] The present application relates to the field of ship technology, and in particular to an artificial intelligence-based ship navigation obstacle avoidance control method and system. Background Art

[0002] At present, with the rapid development of ocean development and global trade, the global shipping network is becoming increasingly complex, the scale of ships is constantly expanding, the shipping volume is also greatly increased, and maritime traffic is becoming increasingly busy. Therefore, the issue of ship navigation safety has become particularly important. During the navigation of ships, obstacles on the route are the main factor causing ship safety problems. Therefore, accurate obstacle avoidance of obstacles during the navigation of ships through obstacle avoidance methods is of great significance to the stability and safety of ship navigation.

[0003] The ship navigation obstacle avoidance control method in the related technology is often unable to limit the safety range of the ship according to the actual situation of the ship and the obstacle, and is unable to achieve optimal control of the ship's navigation path while achieving ship navigation obstacle avoidance. There is room for improvement. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide a ship navigation obstacle avoidance control method and system based on artificial intelligence, so as to improve the problem that the ship navigation obstacle avoidance control method in the related technology is often unable to limit the safety range of the ship according to the actual situation of the ship and the obstacle, and is unable to achieve optimized control of the ship's navigation path while achieving ship navigation obstacle avoidance.

[0005] In a first aspect of the present application, there is provided a method for controlling ship navigation obstacles based on artificial intelligence, comprising the following steps:

[0006] Acquire ship data of a target ship, and set a monitoring area of ​​the target ship according to the ship data;

[0007] Acquiring regional data of the monitoring area;

[0008] Analyzing the area data to obtain obstacle parameter data and obstacle type data of obstacles in the monitoring area;

[0009] According to the ship data of the target ship and the obstacle parameter data of the obstacle, an initial obstacle avoidance area of ​​the obstacle is obtained;

[0010] According to the obstacle type data and the initial obstacle avoidance area of ​​the obstacle, the corresponding obstacle avoidance area is obtained;

[0011] And according to the benchmark obstacle avoidance area corresponding to the obstacles in the monitoring area, the total obstacle avoidance area of ​​the monitoring area and the obstacle avoidance coefficient of the total obstacle avoidance area are obtained;

[0012] Acquire the ship's initial location and ship's destination of the target ship, and generate a navigation route set of the target ship according to the ship's initial location and the ship's destination; the navigation route set includes at least one navigation route;

[0013] Obtain the interval distance between the navigation route and the total obstacle avoidance area in the monitoring area, and obtain the path safety factor corresponding to the navigation route based on the interval distance and the obstacle avoidance coefficient of the total obstacle avoidance area;

[0014] Obtaining the path distance of the navigation path, and obtaining the path comprehensive coefficient of the navigation path according to the path distance and the path safety factor analysis;

[0015] A target navigation path of the target ship is obtained according to the path comprehensive coefficient, and navigation control of the target ship is performed according to the target navigation path.

[0016] Preferably, the ship data of the target ship is obtained, and the monitoring area of ​​the target ship is set according to the ship data, specifically:

[0017] The ship data includes ship hull parameters, ship heading and ship speed of the target ship;

[0018] Setting the monitoring direction of the target ship according to the ship's heading;

[0019] Setting the monitoring distance of the target ship according to the ship hull parameters and the ship speed;

[0020] The monitoring area of ​​the target ship is obtained according to the monitoring direction and the monitoring distance.

[0021] Preferably, the regional data of the monitoring area is acquired; and the regional data is analyzed to obtain obstacle parameter data and obstacle type data of obstacles in the monitoring area, specifically:

[0022] Setting a sensor device, and acquiring regional data of a monitoring area according to the sensor device, wherein the regional data includes sensor data of obstacles in the monitoring area;

[0023] Set up obstacle recognition model;

[0024] The obstacle recognition model is provided with a model input terminal, a model analysis terminal and a model output terminal; the model analysis terminal is preset with an obstacle recognition algorithm;

[0025] The model input terminal is used to input sensor data of obstacles in the monitoring area;

[0026] The model analysis end is used to perform algorithm training on the obstacle recognition algorithm preset by the model analysis end according to the historical sensing data and the historical obstacle recognition results, so as to obtain the target obstacle recognition algorithm of the obstacle recognition model;

[0027] and performing obstacle recognition on the sensor data of obstacles in the monitoring area according to the target obstacle recognition algorithm, and obtaining an obstacle recognition result, wherein the obstacle recognition result includes obstacle parameter data and obstacle type data;

[0028] The model output end is used to output obstacle parameter data and obstacle type data of obstacles in the monitoring area.

[0029] Preferably, the initial obstacle avoidance area of ​​the obstacle is obtained according to the ship data of the target ship and the obstacle parameter data of the obstacle, specifically:

[0030] The obstacle parameter data includes obstacle shape data, obstacle position data and obstacle movement data;

[0031] Classifying obstacles into static obstacles and dynamic obstacles according to the obstacle movement data;

[0032] Obtaining an initial obstacle avoidance area for the static obstacle based on obstacle shape data and obstacle position data of the static obstacle;

[0033] An initial obstacle avoidance area of ​​the dynamic obstacle is obtained according to the obstacle shape data, the obstacle position data and the obstacle movement data of the dynamic obstacle.

[0034] Preferably, according to the obstacle type data of the obstacle and the initial obstacle avoidance area, the corresponding obstacle avoidance area is obtained, specifically:

[0035] Setting the obstacle avoidance weight corresponding to the obstacle according to the obstacle type data of the obstacle;

[0036] Obtaining the impact area of ​​the obstacle according to the obstacle avoidance weight;

[0037] And according to the initial obstacle avoidance area and the influence area range of the obstacle, the corresponding benchmark obstacle avoidance area of ​​the obstacle is obtained.

[0038] Preferably, the total obstacle avoidance area of ​​the monitoring area and the obstacle avoidance coefficient of the total obstacle avoidance area are obtained according to the benchmark obstacle avoidance area corresponding to the obstacle in the monitoring area, specifically:

[0039] Mark all the obstacle avoidance areas corresponding to all obstacles in the monitoring area, and record the marked areas in the monitoring area as the total obstacle avoidance area of ​​the monitoring area;

[0040] Obtaining a regional distribution map of each area in the total obstacle avoidance area, and dividing the total obstacle avoidance area into at least one sub-obstacle avoidance area according to the regional distribution map;

[0041] Obtaining the number of area markings of each sub-obstacle avoidance area in the total obstacle avoidance area;

[0042] The obstacle avoidance coefficient corresponding to the sub-obstacle avoidance area in the total obstacle avoidance area is set according to the number of area markings.

[0043] Preferably, the ship initial location and the ship destination of the target ship are obtained, and a navigation route set of the target ship is generated according to the ship initial location and the ship destination; the navigation route set includes at least one navigation route, specifically:

[0044] Setting a path generation model, the path generation model comprising a model input end, a model analysis end and a model output end; the model analysis end is provided with a path generation algorithm;

[0045] The model input end is used to input the ship's initial location and ship's destination of the target ship;

[0046] The model analysis end is used to train the path generation algorithm preset by the model analysis end according to the ship's initial location, the ship's destination and the historical navigation route, so as to obtain the target path generation algorithm of the path generation model;

[0047] and identifying the ship's initial location and the ship's target location of the target ship according to the target path generation algorithm, and generating a navigation path between the ship's initial location and the ship's target location;

[0048] The model output end is used to output a navigation path set consisting of the navigation path between the ship's initial location and the ship's destination.

[0049] Preferably, the interval distance between the navigation route and the total obstacle avoidance area in the monitoring area is obtained, and the path safety factor corresponding to the navigation route is obtained according to the interval distance and the obstacle avoidance coefficient of the total obstacle avoidance area, specifically:

[0050] Get the interval distance between the navigation path and the intermediate interval area in the total obstacle avoidance area;

[0051] The navigation path and the interval distance between the sub-interval area in the total obstacle avoidance area are taken and marked to obtain the interval distance JL; the obstacle avoidance coefficient of the sub-interval area in the total obstacle avoidance area is taken and marked to obtain the obstacle avoidance coefficient BX;

[0052] By first calculating the function The path safety factor AX corresponding to the navigation path is calculated, where i is the number of the sub-obstacle avoidance area in the total obstacle avoidance area, m is the number of the sub-obstacle avoidance areas in the total obstacle avoidance area, and a1 and a2 are preset proportional factors.

[0053] Preferably, the path distance of the navigation path is obtained, and the path comprehensive coefficient of the navigation path is obtained according to the path distance and the path safety factor analysis, specifically:

[0054] The path distance of the navigation path is obtained by taking values ​​and marking the path distance HJ;

[0055] Through the second calculation function The comprehensive coefficient ZX of the navigation route is calculated, where b1 and b2 are preset proportional factors;

[0056] The target navigation path of the target ship is obtained according to the path comprehensive coefficient, and the navigation of the target ship is controlled according to the target navigation path, specifically:

[0057] The comprehensive coefficients of the target ship's navigation routes are compared, and the navigation route with the largest comprehensive coefficient is recorded as the target navigation route.

[0058] In a second aspect of the present application, a ship navigation obstacle avoidance control system based on artificial intelligence is provided, which is applied to a ship navigation obstacle avoidance control method based on artificial intelligence, and the system comprises:

[0059] The first acquisition module is used to acquire the ship data of the target ship and set the monitoring area of ​​the target ship according to the ship data;

[0060] The second acquisition module is used to acquire the area data of the monitoring area;

[0061] Analyzing the area data to obtain obstacle parameter data and obstacle type data of obstacles in the monitoring area;

[0062] The first analysis module: obtains the initial obstacle avoidance area of ​​the obstacle according to the ship data of the target ship and the obstacle parameter data of the obstacle;

[0063] According to the obstacle type data and the initial obstacle avoidance area of ​​the obstacle, the corresponding obstacle avoidance area is obtained;

[0064] And according to the benchmark obstacle avoidance area corresponding to the obstacles in the monitoring area, the total obstacle avoidance area of ​​the monitoring area and the obstacle avoidance coefficient of the total obstacle avoidance area are obtained;

[0065] The second analysis module: obtaining the ship's initial location and ship's destination of the target ship, and generating a navigation route set of the target ship according to the ship's initial location and ship's destination; the navigation route set includes at least one navigation route;

[0066] The third analysis module: obtains the interval distance between the navigation route and the total obstacle avoidance area in the monitoring area, and obtains the path safety factor corresponding to the navigation route according to the interval distance and the obstacle avoidance coefficient of the total obstacle avoidance area;

[0067] Obtaining the path distance of the navigation path, and obtaining the path comprehensive coefficient of the navigation path according to the path distance and the path safety factor analysis;

[0068] Obstacle avoidance control module: obtains the target navigation path of the target ship according to the path comprehensive coefficient, and performs navigation control on the target ship according to the target navigation path.

[0069] In summary, the beneficial effects of the present application are as follows: the present application sets a monitoring area for a target ship according to the ship data of the target ship, and analyzes the area data to obtain obstacle parameter data and obstacle type data of obstacles in the monitoring area; obtains an initial obstacle avoidance area for the obstacle according to the ship data of the target ship and the obstacle parameter data of the obstacle, obtains a reference obstacle avoidance area corresponding to the obstacle according to the obstacle type data and the initial obstacle avoidance area of ​​the obstacle, and obtains a total obstacle avoidance area for the monitoring area and an obstacle avoidance coefficient for the total obstacle avoidance area according to the reference obstacle avoidance area corresponding to the obstacles in the monitoring area; in addition, according to the initial obstacle avoidance area of ​​the target ship, A navigation path set of the target ship is generated based on the ship's target location, and the interval distance between the navigation path and the total obstacle avoidance area in the monitoring area is obtained. The path safety factor corresponding to the navigation path is obtained based on the interval distance and the obstacle avoidance coefficient of the total obstacle avoidance area. The path comprehensive coefficient of the navigation path is obtained based on the path distance and path safety factor of the navigation path. Finally, the target navigation path of the target ship is obtained based on the path comprehensive coefficient, and the navigation of the target ship is controlled based on the target navigation path, so as to limit the safety range of the ship based on the actual situation of the ship and the obstacle, and realize the optimization control of the ship's navigation path while realizing the ship's navigation and obstacle avoidance. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] In order to more clearly illustrate the technical solutions of the embodiments of the present application, some of the drawings in the embodiments of the present application will be briefly described below. It should be understood that the following drawings only show some embodiments of the present application and therefore should not be considered as limiting the scope of the present application.

[0071] Figure 1 A flowchart of a ship navigation obstacle avoidance control method based on artificial intelligence provided in an embodiment of the present application;

[0072] Figure 2 A schematic diagram of the structure of an artificial intelligence-based ship navigation obstacle avoidance control system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0073] Below is a combination of the embodiments and Figure 1 and Figure 2 The present application is further described in detail, but the implementation methods of the present application are not limited thereto.

[0074] Reference Figure 1 As shown, it is a flow chart of an artificial intelligence-based ship navigation obstacle avoidance control method provided in an embodiment of the present application.

[0075] A ship navigation obstacle avoidance control method based on artificial intelligence comprises the following steps:

[0076] Obtain the ship data of the target ship and set the monitoring area of ​​the target ship according to the ship data;

[0077] Get the regional data of the monitoring area;

[0078] Analyze the regional data to obtain obstacle parameter data and obstacle type data of obstacles in the monitoring area;

[0079] According to the ship data of the target ship and the obstacle parameter data of the obstacle, an initial obstacle avoidance area of ​​the obstacle is obtained;

[0080] According to the obstacle type data and the initial obstacle avoidance area of ​​the obstacle, the corresponding obstacle avoidance area is obtained;

[0081] And according to the benchmark obstacle avoidance area corresponding to the obstacles in the monitoring area, the total obstacle avoidance area of ​​the monitoring area and the obstacle avoidance coefficient of the total obstacle avoidance area are obtained;

[0082] Acquire the ship's initial location and ship's destination of the target ship, and generate a navigation route set of the target ship according to the ship's initial location and the ship's destination; the navigation route set includes at least one navigation route;

[0083] Obtain the interval distance between the navigation route and the total obstacle avoidance area in the monitoring area, and obtain the path safety factor corresponding to the navigation route based on the interval distance and the obstacle avoidance coefficient of the total obstacle avoidance area;

[0084] Obtain the path distance of the navigation path, and obtain the path comprehensive coefficient of the navigation path based on the path distance and path safety factor analysis;

[0085] The target navigation path of the target ship is obtained according to the path comprehensive coefficient, and the navigation of the target ship is controlled according to the target navigation path.

[0086] Get the ship data of the target ship and set the monitoring area of ​​the target ship according to the ship data, specifically:

[0087] The ship data includes the ship hull parameters, ship heading and ship speed of the target ship;

[0088] Set the monitoring direction of the target ship according to the ship's heading;

[0089] Set the monitoring distance of the target ship according to the ship's hull parameters and ship speed;

[0090] The monitoring area of ​​the target ship is obtained according to the monitoring direction and monitoring distance.

[0091] In some embodiments, the monitoring distance of the target ship can be obtained based on the ship's hull parameters and the ship's speed-monitoring distance comparison table, wherein the ship's hull parameters and the ship's speed-monitoring distance comparison table are obtained based on the ship's historical navigation data.

[0092] Obtain regional data of the monitoring area; analyze the regional data to obtain obstacle parameter data and obstacle type data of obstacles in the monitoring area, specifically:

[0093] Setting a sensor device, and obtaining regional data of the monitoring area according to the sensor device, wherein the regional data includes sensor data of obstacles in the monitoring area;

[0094] Set up obstacle recognition model;

[0095] The obstacle recognition model is provided with a model input terminal, a model analysis terminal and a model output terminal; the model analysis terminal is preset with an obstacle recognition algorithm;

[0096] The model input is used to input sensor data of obstacles in the monitoring area;

[0097] The model analysis end is used to train the obstacle recognition algorithm preset by the model analysis end according to the historical sensor data and the historical obstacle recognition results, so as to obtain the target obstacle recognition algorithm of the obstacle recognition model;

[0098] And according to the target obstacle recognition algorithm, the sensor data of obstacles in the monitoring area are used for obstacle recognition, and an obstacle recognition result is obtained, wherein the obstacle recognition result includes obstacle parameter data and obstacle type data;

[0099] The model output end is used to output obstacle parameter data and obstacle type data of obstacles in the monitoring area.

[0100] In some embodiments, the sensing device may be configured as a radar sensor or the like.

[0101] According to the ship data of the target ship and the obstacle parameter data of the obstacle, the initial obstacle avoidance area of ​​the obstacle is obtained, specifically:

[0102] The obstacle parameter data includes obstacle shape data, obstacle position data and obstacle movement data;

[0103] Obstacles are divided into static obstacles and dynamic obstacles according to obstacle movement data;

[0104] According to the obstacle shape data and the obstacle position data of the static obstacle, an initial obstacle avoidance area of ​​the static obstacle is obtained;

[0105] An initial obstacle avoidance area for the dynamic obstacle is obtained according to the obstacle shape data, obstacle position data and obstacle movement data of the dynamic obstacle.

[0106] In some embodiments, the initial obstacle avoidance area includes an above-water area and an underwater area.

[0107] According to the obstacle type data and the initial obstacle avoidance area of ​​the obstacle, the corresponding obstacle avoidance area is obtained, specifically:

[0108] Setting the obstacle avoidance weight corresponding to the obstacle according to the obstacle type data of the obstacle;

[0109] Obtain the impact area of ​​the obstacle according to the obstacle avoidance weight;

[0110] And according to the initial obstacle avoidance area and the impact area range of the obstacle, the corresponding benchmark obstacle avoidance area of ​​the obstacle is obtained.

[0111] In some embodiments, different obstacle types have different corresponding obstacle avoidance weights, wherein the specific value of the obstacle avoidance weight can be obtained according to the obstacle type data-obstacle avoidance weight comparison table, and the obstacle type data-obstacle avoidance weight comparison table can be preset according to the historical navigation data of the ship; in addition, each obstacle avoidance weight corresponds to an impact area range, and the specific value of the impact area range can be obtained according to the obstacle avoidance weight-impact area range comparison table, and the obstacle avoidance weight-impact area range comparison table is preset according to the historical navigation data of the ship;

[0112] The benchmark obstacle avoidance area corresponding to the obstacle is obtained by adding the initial obstacle avoidance area to the influence area.

[0113] And according to the benchmark obstacle avoidance area corresponding to the obstacles in the monitoring area, the total obstacle avoidance area of ​​the monitoring area and the obstacle avoidance coefficient of the total obstacle avoidance area are obtained, specifically:

[0114] Mark all the obstacle avoidance areas corresponding to all obstacles in the monitoring area, and record the marked areas in the monitoring area as the total obstacle avoidance area of ​​the monitoring area;

[0115] Obtaining a regional distribution map of each area in the total obstacle avoidance area, and dividing the total obstacle avoidance area into at least one sub-obstacle avoidance area according to the regional distribution map;

[0116] Get the number of area markings for each sub-obstacle avoidance area in the total obstacle avoidance area;

[0117] The obstacle avoidance coefficient corresponding to the sub-obstacle avoidance area in the total obstacle avoidance area is set according to the number of area markings.

[0118] In some embodiments, the regional distribution map of each area in the total obstacle avoidance area includes the specific distribution of each area in the total obstacle avoidance area; in addition, according to the regional concentration of each area in the total obstacle avoidance area, the concentrated areas in the total obstacle avoidance area are divided into a sub-obstacle avoidance area;

[0119] In addition, the obstacle avoidance coefficient corresponding to the sub-obstacle avoidance area is positively correlated with the area marking times of the sub-obstacle avoidance area, that is, the greater the area marking times, the greater the obstacle avoidance coefficient. The specific value of the obstacle avoidance coefficient can be obtained based on the area marking times-obstacle avoidance coefficient comparison table, and the area marking times-obstacle avoidance coefficient comparison table can be preset based on the ship's historical navigation data.

[0120] The ship's initial location and ship's destination are obtained, and a navigation route set of the target ship is generated according to the ship's initial location and the ship's destination; the navigation route set includes at least one navigation route, specifically:

[0121] Setting a path generation model, the path generation model includes a model input end, a model analysis end and a model output end; the model analysis end is provided with a path generation algorithm;

[0122] The model input terminal is used to input the ship's initial location and ship's destination of the target ship;

[0123] The model analysis end is used to train the path generation algorithm preset by the model analysis end according to the ship's initial location, the ship's destination, and the historical navigation route, and obtain the target path generation algorithm of the path generation model;

[0124] The target ship's initial location and target location are identified according to the target path generation algorithm, and a navigation path between the initial location and the target location is generated;

[0125] The model output end is used to output a navigation path set consisting of the navigation path between the ship's initial location and the ship's destination.

[0126] Get the distance between the navigation path and the total obstacle avoidance area in the monitoring area, and get the path safety factor corresponding to the navigation path based on the distance and the obstacle avoidance coefficient of the total obstacle avoidance area, specifically:

[0127] Get the interval distance between the navigation path and the intermediate interval area in the total obstacle avoidance area;

[0128] The navigation path and the interval distance between the sub-interval area in the total obstacle avoidance area are taken and marked to obtain the interval distance JL; the obstacle avoidance coefficient of the sub-interval area in the total obstacle avoidance area is taken and marked to obtain the obstacle avoidance coefficient BX;

[0129] By first calculating the function The path safety factor AX corresponding to the navigation path is calculated, where i is the number of the sub-obstacle avoidance area in the total obstacle avoidance area, m is the number of the sub-obstacle avoidance areas in the total obstacle avoidance area, and a1 and a2 are preset proportional factors.

[0130] In some embodiments, the interval distance between the navigation path and the sub-interval area in the total obstacle avoidance area refers to the vertical distance between the navigation path and the sub-interval area in the total obstacle avoidance area;

[0131] If the navigation path is separated from the neutron separation area in the total obstacle avoidance area by 10 nautical miles, the separation distance JL is taken as 2; the obstacle avoidance coefficient of the neutron obstacle avoidance area in the total obstacle avoidance area is 10, then the obstacle avoidance coefficient BX is taken as 10;

[0132] Among them, the proportional factors a1 and a2 are preset based on the influence weights of the interval distance and the obstacle avoidance coefficient on the path safety factor. For example, if the influence weight of the interval distance on the path safety factor is greater than the influence weight of the obstacle avoidance coefficient on the path safety factor, then the proportional factor a1 is greater than the proportional factor a2. The specific values ​​of the proportional factors a1 and a2 can be preset based on the actual influence weights of the path safety factor.

[0133] Get the path distance of the navigation path, and analyze the path distance and path safety factor to get the path comprehensive coefficient of the navigation path, specifically:

[0134] The path distance of the navigation path is obtained by taking values ​​and marking the path distance HJ;

[0135] Through the second calculation function The comprehensive coefficient ZX of the navigation route is calculated, where b1 and b2 are preset proportional factors;

[0136] The target navigation path of the target ship is obtained according to the path comprehensive coefficient, and the navigation of the target ship is controlled according to the target navigation path, specifically:

[0137] The comprehensive coefficients of the target ship's navigation routes are compared, and the navigation route with the largest comprehensive coefficient is recorded as the target navigation route.

[0138] In some embodiments, if the route distance of the navigation path is 1000 nautical miles, the route distance HJ is 1000; if the route safety factor AX corresponding to the navigation path is 100, and the proportional factors b1 and b2 are 0.1 and 1, then the second calculation function The calculated route comprehensive coefficient ZX of the navigation route is 200; wherein the proportional factors b1 and b2 are preset according to the influence weights of the path distance and the route safety factor on the route comprehensive coefficient. The specific values ​​of the proportional factors b1 and b2 can be preset according to the actual influence weights of the route comprehensive coefficient. In this embodiment, the values ​​of b1 and b2 are only examples.

[0139] If, when the route safety factor is the maximum, there are navigation routes with the same and maximum route safety factors, then the navigation routes with the same and maximum route safety factors are randomly selected, and the randomly selected navigation routes are recorded as the target navigation routes.

[0140] Reference Figure 2 FIG. 1 is a schematic diagram of a ship navigation obstacle avoidance control system based on artificial intelligence provided by an embodiment of the present application, the system comprising:

[0141] The first acquisition module: acquires the ship data of the target ship and sets the monitoring area of ​​the target ship according to the ship data;

[0142] The second acquisition module: acquires the regional data of the monitoring area;

[0143] Analyze the regional data to obtain obstacle parameter data and obstacle type data of obstacles in the monitoring area;

[0144] The first analysis module: obtains the initial obstacle avoidance area of ​​the obstacle according to the ship data of the target ship and the obstacle parameter data of the obstacle;

[0145] According to the obstacle type data and the initial obstacle avoidance area of ​​the obstacle, the corresponding obstacle avoidance area is obtained;

[0146] And according to the benchmark obstacle avoidance area corresponding to the obstacles in the monitoring area, the total obstacle avoidance area of ​​the monitoring area and the obstacle avoidance coefficient of the total obstacle avoidance area are obtained;

[0147] The second analysis module: obtaining the ship's initial location and ship's destination of the target ship, and generating a navigation route set of the target ship according to the ship's initial location and ship's destination; the navigation route set includes at least one navigation route;

[0148] The third analysis module: obtains the interval distance between the navigation route and the total obstacle avoidance area in the monitoring area, and obtains the path safety factor corresponding to the navigation route according to the interval distance and the obstacle avoidance coefficient of the total obstacle avoidance area;

[0149] Obtain the path distance of the navigation path, and obtain the path comprehensive coefficient of the navigation path based on the path distance and path safety factor analysis;

[0150] Obstacle avoidance control module: obtains the target navigation path of the target ship based on the path comprehensive coefficient, and controls the navigation of the target ship based on the target navigation path.

[0151] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A ship navigation obstacle avoidance control method based on artificial intelligence, characterized in that: The following steps are involved: Acquire ship data of a target ship, and set a monitoring area of ​​the target ship according to the ship data; Acquiring regional data of the monitoring area; Analyzing the area data to obtain obstacle parameter data and obstacle type data of obstacles in the monitoring area; According to the ship data of the target ship and the obstacle parameter data of the obstacle, an initial obstacle avoidance area of ​​the obstacle is obtained; According to the obstacle type data and the initial obstacle avoidance area of ​​the obstacle, the corresponding obstacle avoidance area is obtained; And according to the benchmark obstacle avoidance area corresponding to the obstacles in the monitoring area, the total obstacle avoidance area of ​​the monitoring area and the obstacle avoidance coefficient of the total obstacle avoidance area are obtained; Acquire the ship's initial location and ship's destination of the target ship, and generate a navigation route set of the target ship according to the ship's initial location and the ship's destination; the navigation route set includes at least one navigation route; Obtain the interval distance between the navigation route and the total obstacle avoidance area in the monitoring area, and obtain the path safety factor corresponding to the navigation route based on the interval distance and the obstacle avoidance coefficient of the total obstacle avoidance area; Obtaining the path distance of the navigation path, and obtaining the path comprehensive coefficient of the navigation path according to the path distance and the path safety factor analysis; A target navigation path of the target ship is obtained according to the path comprehensive coefficient, and navigation control of the target ship is performed according to the target navigation path.

2. The method for ship navigation obstacle avoidance control based on artificial intelligence according to claim 1 is characterized in that: Obtain the ship data of the target ship, and set the monitoring area of ​​the target ship according to the ship data, specifically: The ship data includes ship hull parameters, ship heading and ship speed of the target ship; Setting the monitoring direction of the target ship according to the ship's heading; Setting the monitoring distance of the target ship according to the ship hull parameters and the ship speed; The monitoring area of ​​the target ship is obtained according to the monitoring direction and the monitoring distance.

3. The method for ship navigation obstacle avoidance control based on artificial intelligence according to claim 2 is characterized in that: Obtaining regional data of the monitoring area; analyzing the regional data to obtain obstacle parameter data and obstacle type data of obstacles in the monitoring area, specifically: Setting a sensor device, and acquiring regional data of a monitoring area according to the sensor device, wherein the regional data includes sensor data of obstacles in the monitoring area; Set up obstacle recognition model; The obstacle recognition model is provided with a model input terminal, a model analysis terminal and a model output terminal; the model analysis terminal is preset with an obstacle recognition algorithm; The model input terminal is used to input sensor data of obstacles in the monitoring area; The model analysis end is used to perform algorithm training on the obstacle recognition algorithm preset by the model analysis end according to the historical sensing data and the historical obstacle recognition results, so as to obtain the target obstacle recognition algorithm of the obstacle recognition model; and performing obstacle recognition on the sensor data of obstacles in the monitoring area according to the target obstacle recognition algorithm, and obtaining an obstacle recognition result, wherein the obstacle recognition result includes obstacle parameter data and obstacle type data; The model output end is used to output obstacle parameter data and obstacle type data of obstacles in the monitoring area.

4. The method for ship navigation obstacle avoidance control based on artificial intelligence according to claim 3 is characterized in that: According to the ship data of the target ship and the obstacle parameter data of the obstacle, the initial obstacle avoidance area of ​​the obstacle is obtained, specifically: The obstacle parameter data includes obstacle shape data, obstacle position data and obstacle movement data; Classifying obstacles into static obstacles and dynamic obstacles according to the obstacle movement data; Obtaining an initial obstacle avoidance area for the static obstacle based on obstacle shape data and obstacle position data of the static obstacle; An initial obstacle avoidance area of ​​the dynamic obstacle is obtained according to the obstacle shape data, obstacle position data and obstacle movement data of the dynamic obstacle.

5. The method for ship navigation obstacle avoidance control based on artificial intelligence according to claim 4 is characterized in that: According to the obstacle type data and the initial obstacle avoidance area of ​​the obstacle, the corresponding obstacle avoidance area is obtained, specifically: Setting the obstacle avoidance weight corresponding to the obstacle according to the obstacle type data of the obstacle; Obtaining the impact area of ​​the obstacle according to the obstacle avoidance weight; And according to the initial obstacle avoidance area and the influence area range of the obstacle, the corresponding benchmark obstacle avoidance area of ​​the obstacle is obtained.

6. The method for ship navigation obstacle avoidance control based on artificial intelligence according to claim 5 is characterized in that: And according to the benchmark obstacle avoidance area corresponding to the obstacles in the monitoring area, the total obstacle avoidance area of ​​the monitoring area and the obstacle avoidance coefficient of the total obstacle avoidance area are obtained, which are specifically: Mark all the obstacle avoidance areas corresponding to all obstacles in the monitoring area, and record the marked areas in the monitoring area as the total obstacle avoidance area of ​​the monitoring area; Obtaining a regional distribution map of each area in the total obstacle avoidance area, and dividing the total obstacle avoidance area into at least one sub-obstacle avoidance area according to the regional distribution map; Obtaining the number of area markings of each sub-obstacle avoidance area in the total obstacle avoidance area; The obstacle avoidance coefficient corresponding to the sub-obstacle avoidance area in the total obstacle avoidance area is set according to the number of area markings.

7. The method for ship navigation obstacle avoidance control based on artificial intelligence according to claim 6 is characterized in that: The ship initial location and the ship destination of the target ship are obtained, and a navigation route set of the target ship is generated according to the ship initial location and the ship destination; the navigation route set includes at least one navigation route, specifically: Setting a path generation model, the path generation model comprising a model input end, a model analysis end and a model output end; the model analysis end is provided with a path generation algorithm; The model input end is used to input the ship's initial location and ship's destination of the target ship; The model analysis end is used to train the path generation algorithm preset by the model analysis end according to the ship's initial location, the ship's destination and the historical navigation route, so as to obtain the target path generation algorithm of the path generation model; and identifying the ship's initial location and the ship's target location of the target ship according to the target path generation algorithm, and generating a navigation path between the ship's initial location and the ship's target location; The model output end is used to output a navigation path set consisting of the navigation path between the ship's initial location and the ship's destination.

8. The method for ship navigation obstacle avoidance control based on artificial intelligence according to claim 7 is characterized in that: Get the distance between the navigation path and the total obstacle avoidance area in the monitoring area, and get the path safety factor corresponding to the navigation path based on the distance and the obstacle avoidance coefficient of the total obstacle avoidance area, specifically: Get the distance between the navigation path and the intermediate interval area in the total obstacle avoidance area; The navigation path and the interval distance between the sub-interval area in the total obstacle avoidance area are taken and marked to obtain the interval distance JL; the obstacle avoidance coefficient of the sub-interval area in the total obstacle avoidance area is taken and marked to obtain the obstacle avoidance coefficient BX; By first calculating the function The path safety factor AX corresponding to the navigation path is calculated, where i is the number of the sub-obstacle avoidance area in the total obstacle avoidance area, m is the number of the sub-obstacle avoidance areas in the total obstacle avoidance area, and a1 and a2 are preset proportional factors.

9. The method for ship navigation obstacle avoidance control based on artificial intelligence according to claim 8, characterized in that: The path distance of the navigation path is obtained, and the path comprehensive coefficient of the navigation path is obtained according to the path distance and the path safety factor analysis, which is specifically: The path distance of the navigation path is obtained by taking values ​​and marking the path distance HJ; Through the second calculation function The comprehensive coefficient ZX of the navigation route is calculated, where b1 and b2 are preset proportional factors; The target navigation path of the target ship is obtained according to the path comprehensive coefficient, and the navigation of the target ship is controlled according to the target navigation path, specifically: The comprehensive coefficients of the target ship's navigation routes are compared, and the navigation route with the largest comprehensive coefficient is recorded as the target navigation route.

10. A ship navigation obstacle avoidance control system based on artificial intelligence, the system is used to implement the ship navigation obstacle avoidance control method based on artificial intelligence according to any one of claims 1 to 9, characterized in that: The system includes: The first acquisition module is used to acquire the ship data of the target ship and set the monitoring area of ​​the target ship according to the ship data; The second acquisition module is used to acquire the area data of the monitoring area; Analyzing the area data to obtain obstacle parameter data and obstacle type data of obstacles in the monitoring area; The first analysis module: obtains the initial obstacle avoidance area of ​​the obstacle according to the ship data of the target ship and the obstacle parameter data of the obstacle; According to the obstacle type data and the initial obstacle avoidance area of ​​the obstacle, the corresponding obstacle avoidance area is obtained; And according to the benchmark obstacle avoidance area corresponding to the obstacles in the monitoring area, the total obstacle avoidance area of ​​the monitoring area and the obstacle avoidance coefficient of the total obstacle avoidance area are obtained; The second analysis module: obtaining the ship's initial location and ship's destination of the target ship, and generating a navigation route set of the target ship according to the ship's initial location and ship's destination; the navigation route set includes at least one navigation route; The third analysis module: obtains the interval distance between the navigation route and the total obstacle avoidance area in the monitoring area, and obtains the path safety factor corresponding to the navigation route according to the interval distance and the obstacle avoidance coefficient of the total obstacle avoidance area; Obtaining the path distance of the navigation path, and obtaining the path comprehensive coefficient of the navigation path according to the path distance and the path safety factor analysis; Obstacle avoidance control module: obtains the target navigation path of the target ship according to the path comprehensive coefficient, and performs navigation control on the target ship according to the target navigation path.

Images