DCS ship control system based on artificial intelligence

Abstract

The invention discloses a DCS ship control system based on artificial intelligence, and relates to the technical field of artificial intelligence, and the system comprises a target record extraction module which is used for obtaining a collision record, and extracting a target record in the collision record based on an automatic recognition system; the early warning area division module is used for obtaining an early warning area according to the target record and a target area on the bridge; the scatter diagram establishment module is used for acquiring a planned trajectory of the to-be-detected ship passing through the early warning area and a historical actual route trajectory, and extracting a feature route; establishing a scatter diagram; and the early warning prompt module is used for obtaining an early warning value according to the position of the to-be-detected ship and the established scatter diagram, and prompting the to-be-detected ship in time. According to the method, the historical collision record is analyzed to obtain the early warning value corresponding to the deviation of the ship to be detected from the ship route, and related personnel are prompted in time, so that the collision detection reliability is improved, and the collision probability between the ship and the bridge is reduced.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, in particular to a DCS ship control system based on artificial intelligence. Background Art

[0002] DCS stands for Distributed Control System, a comprehensive control, monitoring, and management system. Applying DCS to ship control can help improve ship performance in multiple dimensions, including safety, economy, operational efficiency, and convenience. With the booming shipping industry, ship size and deadweight are increasing. Furthermore, large ships have poor maneuverability and high inertia, increasing the risk of collision between ships and bridges when navigating in bridge areas. Currently, passive anti-collision methods such as installing buoys and guardrails around bridge piers are used to warn ships. However, these methods not only occupy waterways but also have a low level of intelligence, requiring real-time monitoring by ship controllers. These methods have limited protection capabilities and are difficult to address with modern large ships. Summary of the Invention

[0003] The purpose of the present invention is to provide a DCS ship control system based on artificial intelligence to solve the problems raised in the prior art.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] An artificial intelligence-based DCS ship control system includes a target record extraction module, a warning area division module, a scatter plot establishment module, and a warning prompt module;

[0006] Target record extraction module: This module is used to obtain collision records of historical collisions between ships and bridges. An automatic identification system is deployed on the ship to obtain the actual trajectory of the ship during transportation corresponding to the collision record, as well as the planned trajectory preset based on the ship's starting point and destination. The actual trajectory and the planned trajectory are analyzed to extract the target record from the collision record.

[0007] Warning area division module: used to obtain the warning area in the planned trajectory corresponding to the ship to be detected based on the collision position corresponding to the target record and the target area pre-divided on the bridge to restrict the entry and exit of ships;

[0008] Scatter plot creation module: This module is used to obtain the planned trajectory of the ship to be detected passing through the warning area, as well as several actual routes of ships passing through the warning area in history, and extract characteristic routes from the actual routes. Based on the ship's heading and the weight of the cargo loaded corresponding to the characteristic routes, a scatter plot is created to show the relationship between the yaw angle and the ship's weight.

[0009] Early warning prompt module: It is used to monitor the early warning value of the ship to be detected in real time when it is sailing according to the planned trajectory based on the location of the ship to be detected and the established scatter plot, and prompt the relevant control personnel of the ship to take relevant measures in time according to the early warning value.

[0010] Furthermore, the target record extraction module includes a collision record acquisition unit and a target record extraction unit;

[0011] Collision record acquisition unit: used to acquire collision records of several ships of the same type as the ship to be detected and the bridge; take a panoramic image of the bridge, extract the collision time, collision location and collision ship number corresponding to each collision record, and obtain the collision ship S corresponding to a certain collision record R based on the automatic identification system deployed on the ship R Actual trajectory during transportation T R A , and the actual trajectory T R A The collision position C corresponding to R is recorded in R As the starting point, follow the ship S R The opposite direction of the actual track T R A The distance L1 forward is used as the detection distance, and the time period D corresponding to the detection distance and the ship S are obtained. R The actual trajectory T in time period D D A ;

[0012] The Automatic Identification System (AIS) can obtain the heading, position, speed, etc. of a ship while it is sailing.

[0013] Get according to Ship S R The starting point and destination of the ship are determined by the relevant personnel through the navigation tool before the transportation. The maximum width B of the ship is obtained. The range of length k on both sides of the planned trajectory is used as the ship S R The safe trajectory range, where 2k>B.

[0014] Furthermore, the target record extraction unit is used to obtain the ship S according to the automatic identification system. R The actual position C at a certain time t within the time period D t and the actual heading H1; obtain the actual position C in the planned trajectory t The most recent trajectory position P t , and the trajectory position P t As the starting point, according to the ship S RThe positive and reverse headings of the planned trajectory are calculated, and the position points corresponding to the distance L2 forward along the planned trajectory are taken as the first position P1 and the second position P2. The direction from the second position P2 to the first position P1 is taken as the predicted heading H2, and the angle between the actual heading H1 and the predicted heading H2 is taken as the characteristic angle at time t.

[0015] If the actual position C t Within the safe trajectory range, the deviation characteristic value at time t is set to 0. If the actual position C t Not within the safe trajectory, position C t The length closest to the safe trajectory range is used as the deviation characteristic value at time t: the period D is evenly divided into several sub-periods, and the average value of the deviation characteristic value at each moment in a sub-period is added to obtain the average characteristic value of a sub-period. Then, the variance S is calculated based on the average characteristic value of all sub-periods. D , if the variance S D If the angle is greater than the preset variance threshold and the characteristic angle at each moment in the time period D is less than the preset angle threshold, the collision record R is taken as the target record, and then all target records in the collision record are extracted.

[0016] When a ship gradually deviates from its course before a collision but still collides with it when its direction is similar to the planned direction, this phenomenon reveals systemic risks in ship safety management, and the area corresponding to the record indicates an area prone to accidents. If the judgment basis in this scheme is based on variance, since the average eigenvalue of the sub-period corresponding to the collision is larger (the ship deviates from its course at the time of collision), then the average eigenvalue of the sub-period corresponding to the non-collision is correspondingly smaller, which indicates that this situation is met, so variance can be used for judgment.

[0017] Furthermore, the early warning area division module includes an early warning area division unit;

[0018] Warning area division unit: used to obtain the collision position in the target record and use the area with a radius r centered at the collision position as the first warning area; obtain the target area on each bridge that is historically divided based on floating fences and guardrails to restrict ship entry and exit, and use both the first warning area and the target area as the warning areas in the planned trajectory corresponding to the ship to be detected.

[0019] Furthermore, the scatter plot establishment module includes a characteristic route extraction unit and a scatter plot establishment unit;

[0020] The characteristic route extraction unit is used to obtain a warning area A in the planned trajectory of the ship to be detected. The trajectory corresponding to the distance L3 along the planned trajectory, starting from the warning area A and following the opposite course of the ship to be detected, is used as the trajectory to be detected.

[0021] Obtain the navigation records of several ships of the same type as the ship to be detected that have not collided when passing through the warning area A; extract the ship S corresponding to a navigation record M M and the actual route trajectory T M , randomly obtain several target points from the trajectory to be detected, and then obtain the actual route trajectory T M The point closest to a target point in the trajectory is used as the characteristic distance of the target point, and the variance and average characteristic distance are calculated based on each characteristic distance. If the variance is less than the preset variance threshold and the average characteristic distance is less than the preset distance value, the actual route trajectory T is used as the target point. M As a characteristic route, several characteristic routes are obtained.

[0022] Furthermore, the scatter plot establishment unit is used to obtain the position point P of the ship to be detected at a certain time n. n , get the distance from position point P in a certain characteristic route T0 n The nearest position point P0 is taken, and the ship corresponding to a certain characteristic route T0 is taken as S0, and the time when the ship S0 is at the position point P0 is taken as t0; the predicted time is set as DP, and the position point P1 of the ship S0 at the time t0 and the time after DP is obtained, and the angle between the heading V0 of the ship S0 at the position point P0 and the heading V1 at the position point P1 is taken as the yaw angle, and the heading V0 is rotated 180 degrees in the clockwise and counterclockwise directions respectively to obtain the first range and the second range. If the heading V1 is within the first range, the yaw angle is set to be greater than 0, and if the heading V1 is within the second range, the yaw angle is set to be less than 0; and according to the actual ship weight of the ship S0 at the time t0, the coordinate point corresponding to the ship S0 is marked in the pre-established scatter plot of the yaw angle corresponding to a certain time n as a inverse of the ship weight, and then several coordinate points in the scatter plot are marked.

[0023] Furthermore, the early warning prompt module includes a location analysis early warning unit and an early warning prompt unit;

[0024] Position analysis and early warning unit: used to obtain the actual ship weight wgt and position point P of the ship to be detected at a certain time q q , actual heading H q and speed Y q , if the position point P q If the position point P is not within the safe trajectory, an early warning will be issued; q Within the safe trajectory range, the predicted heading H corresponding to the ship to be detected at a certain time q is obtained according to the target record extraction module. f , and the characteristic angle V q , if the characteristic angle V q If the angle is greater than the preset threshold, an early warning will be issued.

[0025] Furthermore, the early warning unit is used to determine if the position point P q Within the safe trajectory and with characteristic angle V q The angle is not greater than the preset threshold value, and the scatter plot corresponding to the time q is obtained. The coordinate points corresponding to the X ship weights with the smallest difference from the weight wgt in the scatter plot are obtained, and the average yaw angle corresponding to the X coordinate points is calculated to obtain the average yaw angle G; if G>0, the ship is moved along the actual heading H. q Rotate clockwise in the direction of G, with position point P q As the starting point, go forward DP×Y q The end point is taken as the target position point; if G<0, follow the actual heading H q Rotate counterclockwise in the direction of G, with position point P q As the starting point, go forward DP×Y q The end point is taken as the target position point; then the position point closest to the target position point in the planned trajectory is obtained, and the corresponding distance value is recorded as L;

[0026] Then the warning value of the ship to be detected at time q is Z q =W1×(1-e -L )+W2×(1-e -|G| ), where W1 and W2 are the first weight and the second weight respectively, and according to the warning value Z q , to give early warning to the ship to be inspected, and prompt the relevant control personnel of the ship to take relevant measures in time.

[0027] The warning value in this scheme represents the degree of deviation of the ship's position in the future. According to common sense, when a ship does not follow a pre-set appropriate trajectory, it is very easy to deviate from the planned trajectory in the target area when it is heading towards the target area, which may lead to the occurrence of bridge collision problems. Therefore, in this scheme, the ship's driving conditions are analyzed to obtain the warning value compared with the preset planned trajectory, and early warning prompts are issued to help relevant personnel make judgments and make corrections to the ship in advance, which will help reduce the risk of collision when passing through the warning area.

[0028] Compared with the prior art, the present invention has the following advantages: the present invention provides an artificial intelligence-based DCS ship control system, comprising a target record extraction module for acquiring collision records and, based on an automatic recognition system, extracting target records from the collision records; a warning area division module for obtaining a warning area based on the target records and the target area on the bridge; a scatter plot creation module for obtaining the planned trajectory of the ship to be detected passing through the warning area and the historical actual route trajectory, extracting the characteristic route, and creating a scatter plot; and a warning prompt module for obtaining a warning value based on the location of the ship to be detected and the created scatter plot, and providing timely prompts to the ship to be detected. By analyzing historical collision records, the present invention obtains a warning value corresponding to the deviation of the ship to be detected from the route, and promptly prompts relevant personnel, thereby helping to improve the reliability of collision detection and reduce the probability of collision between ships and bridges. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 This is a structural diagram of a DCS ship control system based on artificial intelligence in the present invention. DETAILED DESCRIPTION

[0030] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0031] Example: Figure 1 As shown, the present invention provides a technical solution for a DCS ship control system based on artificial intelligence, including a target record extraction module, a warning area division module, a scatter plot establishment module and a warning prompt module.

[0032] Target record extraction module: This module is used to obtain collision records of historical collisions between ships and bridges. An automatic identification system is deployed on the ship to obtain the actual trajectory of the ship during transportation corresponding to the collision record, as well as the planned trajectory preset based on the ship's starting point and destination. The actual trajectory and the planned trajectory are analyzed to extract the target record from the collision record.

[0033] The target record extraction module includes a collision record acquisition unit and a target record extraction unit;

[0034] Collision record acquisition unit: used to acquire collision records of several ships of the same type as the ship to be detected and the bridge; take a panoramic image of the bridge, extract the collision time, collision location and collision ship number corresponding to each collision record, and obtain the collision ship S corresponding to a certain collision record R based on the automatic identification system deployed on the ship R Actual trajectory during transportation T R A , and the actual trajectory T R A The collision position C corresponding to R is recorded in R As the starting point, follow the ship S R The opposite direction of the actual track T R A The distance L1 forward is used as the detection distance, and the time period D corresponding to the detection distance and the ship S are obtained. R The actual trajectory T in time period D D A ;

[0035] Get according to Ship S R The starting point and destination of the ship are determined by the relevant personnel through the navigation tool before the transportation. The maximum width B of the ship is obtained. The range of length k on both sides of the planned trajectory is used as the ship S R The safe trajectory range, where 2k>B.

[0036] Target record extraction unit: used to obtain ship S according to the automatic identification system R The actual position C at a certain time t within the time period D t and the actual heading H1; obtain the actual position C in the planned trajectory t The most recent trajectory position P t , and the trajectory position P t As the starting point, according to the ship S R The positive and reverse headings of the planned trajectory are calculated, and the position points corresponding to the distance L2 forward along the planned trajectory are taken as the first position P1 and the second position P2. The direction from the second position P2 to the first position P1 is taken as the predicted heading H2, and the angle between the actual heading H1 and the predicted heading H2 is taken as the characteristic angle at time t.

[0037] If the actual position C t Within the safe trajectory range, the deviation characteristic value at time t is set to 0. If the actual position C t Not within the safe trajectory, position C tThe length closest to the safe trajectory range is used as the deviation characteristic value at time t: the period D is evenly divided into several sub-periods, and the average value of the deviation characteristic value at each moment in a sub-period is added to obtain the average characteristic value of a sub-period. Then, the variance S is calculated based on the average characteristic value of all sub-periods. D , if the variance S D If the angle is greater than the preset variance threshold and the characteristic angle at each moment in the time period D is less than the preset angle threshold, the collision record R is taken as the target record, and then all target records in the collision record are extracted.

[0038] When a ship gradually deviates from its course before a collision but still collides with it when its direction is similar to the planned direction, this phenomenon reveals systemic risks in ship safety management, and the area corresponding to the record indicates an area prone to accidents. If the judgment basis in this scheme is based on variance, since the average eigenvalue of the sub-period corresponding to the collision is larger (the ship deviates from its course at the time of collision), then the average eigenvalue of the sub-period corresponding to the non-collision is correspondingly smaller, which indicates that this situation is met, so variance can be used for judgment.

[0039] Warning area division module: used to obtain the warning area in the planned trajectory corresponding to the ship to be detected based on the collision position corresponding to the target record and the target area pre-divided on the bridge to restrict the entry and exit of ships;

[0040] The early warning area division module includes an early warning area division unit;

[0041] Warning area division unit: used to obtain the collision position in the target record and use the area with a radius r centered at the collision position as the first warning area; obtain the target area on each bridge that is historically divided based on floating fences and guardrails to restrict ship entry and exit, and use both the first warning area and the target area as the warning areas in the planned trajectory corresponding to the ship to be detected.

[0042] Scatter plot creation module: This module is used to obtain the planned trajectory of the ship to be detected passing through the warning area, as well as several actual routes of ships passing through the warning area in history, and extract characteristic routes from the actual routes. Based on the ship's heading and the weight of the cargo loaded corresponding to the characteristic routes, a scatter plot is created to show the relationship between the yaw angle and the ship's weight.

[0043] The scatter plot establishment module includes a characteristic route extraction unit and a scatter plot establishment unit;

[0044] The characteristic route extraction unit is used to obtain a warning area A in the planned trajectory of the ship to be detected. The trajectory corresponding to the distance L3 along the planned trajectory, starting from the warning area A and following the opposite course of the ship to be detected, is used as the trajectory to be detected.

[0045] Obtain the navigation records of several ships of the same type as the ship to be detected that have not collided when passing through the warning area A; extract the ship S corresponding to a navigation record M M and the actual route trajectory T M , randomly obtain several target points from the trajectory to be detected, and then obtain the actual route trajectory T M The point closest to a target point in the trajectory is used as the characteristic distance of the target point, and the variance and average characteristic distance are calculated based on each characteristic distance. If the variance is less than the preset variance threshold and the average characteristic distance is less than the preset distance value, the actual route trajectory T is used as the target point. M As a characteristic route, several characteristic routes are obtained.

[0046] Scatter plot establishment unit: used to obtain the position point P of the ship to be detected at a certain time n n , get the distance from position point P in a certain characteristic route T0 n The nearest position point P0 is taken, and the ship corresponding to a certain characteristic route T0 is taken as S0, and the time when the ship S0 is at the position point P0 is taken as t0; the predicted time is set as DP, and the position point P1 of the ship S0 at the time t0 and the time after DP is obtained, and the angle between the heading V0 of the ship S0 at the position point P0 and the heading V1 at the position point P1 is taken as the yaw angle, and the heading V0 is rotated 180 degrees in the clockwise and counterclockwise directions respectively to obtain the first range and the second range. If the heading V1 is within the first range, the yaw angle is set to be greater than 0, and if the heading V1 is within the second range, the yaw angle is set to be less than 0; and according to the actual ship weight of the ship S0 at the time t0, the coordinate point corresponding to the ship S0 is marked in the pre-established scatter plot of the yaw angle corresponding to a certain time n as a inverse of the ship weight, and then several coordinate points in the scatter plot are marked.

[0047] Early warning prompt module: It is used to monitor the early warning value of the ship to be detected in real time when it is sailing according to the planned trajectory based on the location of the ship to be detected and the established scatter plot, and prompt the relevant control personnel of the ship to take relevant measures in time according to the early warning value.

[0048] The early warning prompt module includes a location analysis and early warning unit and an early warning prompt unit;

[0049] Position analysis and early warning unit: used to obtain the actual ship weight wgt and position point P of the ship to be detected at a certain time q q , actual heading H q and speed Y q , if the position point P q If the position point P is not within the safe trajectory, an early warning will be issued; qWithin the safe trajectory range, the predicted heading H corresponding to the ship to be detected at a certain time q is obtained according to the target record extraction module. f , and the characteristic angle V q , if the characteristic angle V q If the angle is greater than the preset threshold, an early warning will be issued.

[0050] According to the target record extraction unit in the target record extraction module, the predicted heading H corresponding to the time q is obtained. f and characteristic angle V q The process is as follows: Get the distance position point P in the planned trajectory of the ship to be detected q The most recent trajectory position P a , and the trajectory position P a As the starting point, according to the positive and reverse headings of the ship to be detected, the position points corresponding to the distance L2 forward along the planned trajectory are taken as the third position P3 and the fourth position P4, and the direction from the fourth position P4 to the third position P3 is taken as the predicted heading H f , and the actual heading H q With the predicted heading H f The angle between them is the characteristic angle V q .

[0051] Warning prompt unit: used to determine if the position point P q Within the safe trajectory and with characteristic angle V q The angle is not greater than the preset threshold value, and the scatter plot corresponding to the time q is obtained. The coordinate points corresponding to the X ship weights with the smallest difference from the weight wgt in the scatter plot are obtained, and the average yaw angle corresponding to the X coordinate points is calculated to obtain the average yaw angle G; if G>0, the ship is moved along the actual heading H. q Rotate clockwise in the direction of G, with position point P q As the starting point, go forward DP×Y q The end point is taken as the target position point; if G<0, follow the actual heading H q Rotate counterclockwise in the direction of G, with position point P q As the starting point, go forward DP×Y q The end point is taken as the target position point; then the position point closest to the target position point in the planned trajectory is obtained, and the corresponding distance value is recorded as L;

[0052] Then the warning value of the ship to be detected at time q is Z q =W1×(1-e -L )+W2×(1-e -|G| ), where W1 and W2 are the first weight and the second weight respectively, and according to the warning value Z q, to give early warning to the ship to be inspected, and prompt the relevant control personnel of the ship to take relevant measures in time.

[0053] The warning value in this scheme represents the degree of deviation of the ship's position in the future. According to common sense, when a ship does not follow a pre-set appropriate trajectory, it is very easy to deviate from the planned trajectory in the target area when it is heading towards the target area, which may lead to the occurrence of bridge collision problems. Therefore, in this scheme, the ship's driving conditions are analyzed to obtain the warning value compared with the preset planned trajectory, and early warning prompts are issued to help relevant personnel make judgments and make corrections to the ship in advance, which will help reduce the risk of collision when passing through the warning area.

[0054] Since (1-e -L ) and (1-e -|G| ) are both in the range of 0 to 1, and represent the position deviation and heading deviation of the ship to be detected, respectively. In this embodiment, the sum of W1 and W2 is set to 1, W1 is 0.6, and W2 is 0.4. Then, in this embodiment, Z q When the value is between 0 and 0.5, the ship is in a relatively safe state, but it is still necessary to remain vigilant and remind relevant personnel to continue to pay attention to the various operating parameters of the ship; Z q When the value is between 0.5 and 0.8, the ship is in a generally dangerous state and needs to remain vigilant, and the relevant personnel are reminded to take protective measures in advance; Z q When it is between 0 and 0.5, the ship faces a higher risk and must take emergency measures immediately to adjust the course as soon as possible to return the ship to a safe route as soon as possible. The specific value should be determined according to actual conditions.

[0055] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. A DCS ship control system based on artificial intelligence, characterized in that: The system includes a target record extraction module, an early warning area division module, a scatter plot establishment module and an early warning prompt module; Target record extraction module: This module is used to obtain collision records of historical collisions between ships and bridges. An automatic identification system is deployed on the ship to obtain the actual trajectory of the ship during transportation corresponding to the collision record, as well as the planned trajectory preset based on the ship's starting point and destination. The actual trajectory and the planned trajectory are analyzed to extract the target record from the collision record. Warning area division module: used to obtain the warning area in the planned trajectory corresponding to the ship to be detected based on the collision position corresponding to the target record and the target area pre-divided on the bridge to restrict the entry and exit of ships; Scatter plot creation module: used to obtain the planned trajectory of the ship to be detected passing through the warning area, as well as several actual route trajectories of ships passing through the warning area in history, and extract characteristic routes from the actual route trajectories; According to the course of the ship and the weight of the loaded cargo corresponding to the characteristic route, a scatter plot of the relationship between the yaw angle and the ship weight is established; Early warning prompt module: It is used to monitor the early warning value of the ship to be detected in real time when it is sailing according to the planned trajectory based on the location of the ship to be detected and the established scatter plot, and prompt the relevant control personnel of the ship to take relevant measures in time according to the early warning value.

2. The artificial intelligence-based DCS ship control system according to claim 1, characterized in that: The target record extraction module includes a collision record acquisition unit and a target record extraction unit; The collision record acquisition unit is used to acquire collision records of several ships of the same type as the ship to be detected and the bridge; take a panoramic image of the bridge, extract the collision time, collision location and collision ship number corresponding to each collision record, and obtain the collision ship S corresponding to a certain collision record R according to the automatic identification system deployed on the ship. R Actual trajectory during transportation T R A , and the actual trajectory T R A The collision position C corresponding to R is recorded in R As the starting point, follow the ship S R The opposite heading, along the actual trajectory T R A The distance L1 forward is used as the detection distance, and the time period D corresponding to the detection distance and the ship S are obtained. R The actual trajectory T in time period D D A ; Get the ship S R The starting point and destination of the ship are determined by the relevant personnel through the navigation tool before the transportation. The maximum width B of the ship is obtained, and the range of length k on both sides of the planned trajectory is used as the ship S R The safe trajectory range, where 2k>B.

3. The DCS ship control system based on artificial intelligence according to claim 2, characterized in that: The target record extraction unit is used to obtain the ship S according to the automatic identification system. R The actual position C at a certain time t within the time period D t and the actual heading H1; obtain the actual position C in the planned trajectory t The most recent trajectory position P t , and the trajectory position P t As the starting point, according to the ship S R The positive and reverse headings of the planned trajectory are L2 forward along the planned trajectory, and the corresponding positions are taken as the first position P1 and the second position P2. The direction from the second position P2 to the first position P1 is taken as the predicted heading H2, and the angle between the actual heading H1 and the predicted heading H2 is taken as the characteristic angle at time t. If the actual position C t Within the safe trajectory range, the deviation characteristic value at time t is set to 0. If the actual position C t Not within the safe trajectory, position C t The length closest to the safe trajectory range is used as the deviation characteristic value at time t: the period D is evenly divided into several sub-periods, and the average value of the deviation characteristic value at each moment in a sub-period is added to obtain the average characteristic value of a sub-period. Then, the variance S is calculated based on the average characteristic value of all sub-periods. D , if the variance S D If the angle is greater than the preset variance threshold and the characteristic angle at each moment in the time period D is less than the preset angle threshold, the collision record R is taken as the target record, and then all target records in the collision record are extracted.

4. The artificial intelligence-based DCS ship control system according to claim 1, characterized in that: The warning area division module includes a warning area division unit; the warning area division unit is used to obtain the collision position in the target record and use the area with the collision position as the center and a radius of r as the first warning area; Obtain the target area on each bridge that is historically divided based on floating fences and guard piles to restrict ship entry and exit, and use both the first warning area and the target area as the warning area in the planned trajectory corresponding to the ship to be detected.

5. The DCS ship control system based on artificial intelligence according to claim 1, characterized in that: The scatter plot establishment module includes a characteristic route extraction unit and a scatter plot establishment unit; the characteristic route extraction unit is used to obtain a warning area A in the planned trajectory of the ship to be detected, and use the warning area A as the starting point, and the trajectory corresponding to the distance L3 along the planned trajectory in the opposite direction of the ship to be detected as the trajectory to be detected; Obtain the navigation records of several ships of the same type as the ship to be detected that have not collided when passing through the warning area A; extract the ship S corresponding to a navigation record M M and the actual route trajectory T M , randomly obtain several target points from the trajectory to be detected, and then obtain the actual route trajectory T M The point closest to a target point is selected and the corresponding distance value is used as the characteristic distance of the target point. The variance and average characteristic distance are calculated based on each characteristic distance. If the variance is less than the preset variance threshold and the average characteristic distance is less than the preset distance value, the actual route trajectory T is converted to M As a characteristic route, several characteristic routes are obtained.

6. The artificial intelligence-based DCS ship control system according to claim 5, characterized in that: The scatter plot establishment unit is used to obtain the position point P of the ship to be detected at a certain time n. n , get the distance from the position point P in a certain characteristic route T0 n The nearest position point P0 is taken, and the ship corresponding to the characteristic route T0 is taken as S0, and the time when the ship S0 is at the position point P0 is taken as t0; the predicted time is set as DP, and the position point P1 where the ship S0 is located at the time t0 and the time after DP is obtained, and the angle between the heading V0 of the ship S0 at the position point P0 and the heading V1 at the position point P1 is taken as the yaw angle, and the heading V0 is rotated 180 degrees in the clockwise and counterclockwise directions respectively to obtain the first range and the second range, if the heading V1 is within the first range, the yaw angle is set to be greater than 0, and if the heading V1 is within the second range, the yaw angle is set to be less than 0; and according to the actual ship weight of the ship S0 at the time t0, the coordinate point corresponding to the ship S0 is marked in the pre-established scatter plot of the yaw angle corresponding to a certain time n as a inverse of the ship weight, and then several coordinate points in the scatter plot are marked.

7. The artificial intelligence-based DCS ship control system according to claim 6, characterized in that: The warning prompt module includes a position analysis warning unit and a warning prompt unit; the position analysis warning unit is used to obtain the actual ship weight wgt, position point P of the ship to be detected at a certain time q q , actual heading H q and speed Y q , if the position point P q If the vehicle is not within the safe trajectory, an early warning will be issued; If the position point P q Within the safe trajectory range, the predicted heading H corresponding to the ship to be detected at a certain time q is obtained according to the target record extraction module. f , and the characteristic angle V q , if the characteristic angle V q If the angle is greater than the preset threshold, an early warning will be issued.

8. The artificial intelligence-based DCS ship control system according to claim 7, characterized in that: The warning prompt unit is used to determine if the position point P q Within the safe trajectory and with characteristic angle V q The angle is not greater than the preset threshold value, and the scatter plot corresponding to the time q is obtained. The coordinate points corresponding to the X ship weights with the smallest difference from the weight wgt in the scatter plot are obtained, and the average yaw angle corresponding to the X coordinate points is calculated to obtain the average yaw angle G; if G>0, the ship is moved along the actual heading H. q Rotate clockwise in the direction of G, with position point P q As the starting point, go forward DP×Y q The end point is taken as the target position point; If G<0, follow H q Rotate counterclockwise in the direction of G, with position point P q As the starting point, go forward DP×Y q The end point is taken as the target position point; then the position point closest to the target position point in the planned trajectory is obtained, and the corresponding distance value is recorded as L; Then the warning value of the ship to be detected at time q is Z q =W1×(1-e -L )+W2×(1-e -|G| ), where W1 and W2 are the first weight and the second weight respectively, and according to the warning value Z q , to give early warning to the ship to be inspected, and prompt the relevant control personnel of the ship to take relevant measures in time.

Images