Intelligent ship autonomous navigation route planning method based on maritime collision avoidance rule

By improving the virtual obstacle line design and obstacle detection sector evaluation of collision risk, and optimizing the track with the dynamic window method, the problem of intelligent ships complying with maritime collision avoidance rules and inefficiency in multiple ship encounters is solved, and safe and efficient trajectory planning.

CN118938932APending Publication Date: 2024-11-12DALIAN MARITIME UNIVERSITY
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411168315.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-23
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

The existing intelligent ship's autonomous navigation track planning method is difficult to effectively comply with maritime collision avoidance rules when dealing with multiple ship encounters and complex navigation environments, and there is a risk of violations. It is impossible to obtain the optimal safe speed vector, which is inefficient.

Method used

Adopt intelligent ship's autonomous navigation track planning method based on maritime collision avoidance rules, improve virtual obstacle line design, set up obstacle detection sectors, evaluate collision risk, optimize track planning, and use Dynamic window method to search for the optimal speed solution.

Benefits of technology

It realizes safe collision avoidance in complex scenarios of multiple ships, improves computing efficiency, ensures that the track complies with maritime collision avoidance rules, and optimizes track safety and efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118938932A_ABST
    Figure CN118938932A_ABST
Patent Text Reader

Abstract

The invention discloses an intelligent ship autonomous navigation route planning method based on a maritime collision avoidance rule, and the method comprises the following steps: building a ship meeting situation quantitative identification model, and judging the meeting type of a barrier ship and an intelligent ship; a virtual obstacle line of ship navigation is improved, obstacles are detected, maritime rule constraint inspection is added, and the actual maritime collision avoidance rule is met; establishing a strategy of an obstacle detection sector to cope with a scene of multiple dynamic obstacle ships, taking a dangerous distance as an obstacle avoidance circle of the dynamic obstacle ships, and evaluating risk values of all dynamic obstacles in the obstacle avoidance circle of the dynamic obstacles; setting that the intelligent ship only performs collision avoidance operation on the most dangerous obstacle ship with the highest risk degree to realize autonomous navigation route planning of the intelligent ship; and optimizing the autonomous navigation track planning of the intelligent ship by using the evaluation function to obtain an optimal autonomous navigation track of the intelligent ship. According to the method, the safety of the intelligent ship in the sailing process can be ensured, and the practicability of the method in the actual sailing state is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of autonomous navigation planning for intelligent ships, and in particular to a method for autonomous navigation track planning for intelligent ships based on maritime collision avoidance rules. Background Art

[0002] Maritime collision avoidance rules are important regulations and guidelines to ensure the safe navigation of smart ships. These rules provide clear avoidance criteria and navigation specifications for smart ships in complex navigation environments, which help reduce the risk of ship collisions and ensure the safety of ship navigation. Complying with maritime collision avoidance rules in track planning can enable smart ships to respond to various navigation conditions more standardizedly, improve the safety and reliability of smart ships in actual shipping routes, and is a prerequisite for the orderly development of smart shipping.

[0003] In recent years, the problem of trajectory planning based on maritime collision avoidance rules has attracted much attention, and planning strategies based on artificial potential field method, speed obstacle method, virtual obstacle method and other technologies have been proposed successively. Among them, the virtual obstacle method realizes planning that meets maritime collision avoidance rules by constructing virtual obstacles. However, this method generally establishes a single virtual obstacle to constrain the navigation range of the ship, which may violate the maritime collision avoidance rules. To address this problem, the existing method proposes a safe collision avoidance strategy based on an improved virtual obstacle line, designs multiple virtual obstacle lines according to the maritime collision avoidance rules, and realizes safe planning under various encounter situations. However, when dealing with the problem of multiple ship encounters, this method solves all obstacle ships at the same time, resulting in a decrease in the efficiency of the method; on the other hand, this method cannot obtain the optimal solution of the safe speed vector in the case of ship encounter situations, so the optimization of the collision avoidance strategy is a problem worthy of study. Although scholars have proposed a variety of autonomous navigation trajectory planning methods for intelligent ships under maritime collision avoidance rules, the existing planning methods still have the following problems:

[0004] First, in the existing virtual obstacle method, although it is possible to achieve planning that meets the maritime collision avoidance rules by constructing virtual obstacles, special encounter situations that violate the maritime collision avoidance rules may exist based on the constraints on the navigation range of ships caused by a single virtual obstacle.

[0005] Second, in the existing intelligent ship collision avoidance trajectory planning methods based on maritime collision avoidance rules, the current methods are mainly applicable to simple scenarios where two ships meet. However, when dealing with complex situations such as multiple ships meeting and the presence of a large number of dynamic obstacles, the safety of the existing methods still needs to be improved.

[0006] Third, in the existing safe collision avoidance method based on the improved virtual obstacle line, the ship can solve a feasible safe speed vector when dealing with the encounter situation, but cannot obtain the optimal solution of the speed vector, and the collision avoidance track needs to be further optimized. Summary of the invention

[0007] In order to solve the above problems, the technical solution adopted by the present invention is: a method for autonomous navigation track planning of an intelligent ship based on maritime collision avoidance rules, comprising the following steps:

[0008] Establish a quantitative identification model for ship encounter situations to determine the encounter types between obstacle ships and intelligent ships;

[0009] Improve the virtual obstacle line for ship navigation, detect obstacles, add inspections of maritime rules constraints, and comply with actual maritime collision avoidance rules;

[0010] Establish an obstacle detection sector strategy to deal with the scenario of multiple dynamic obstacle ships, use the dangerous distance as the dynamic obstacle ship avoidance circle, evaluate the risk value of all dynamic obstacles in the dynamic obstacle avoidance circle, and set the intelligent ship to perform collision avoidance operations only for the most dangerous obstacle ship with the highest risk, so as to realize the autonomous navigation track planning of the intelligent ship;

[0011] The evaluation function is used to optimize the intelligent ship's autonomous navigation trajectory planning to obtain the optimal intelligent ship's autonomous navigation trajectory.

[0012] Further: the process of establishing a quantitative identification model for ship encounter situations and determining the encounter type between the obstacle ship and the intelligent ship is as follows;

[0013] Taking the location of the smart ship as the center and heading 000° as the reference, the encounter area is divided into six parts: 1 , S 2 , S 3 , S 4 , S 5 , S 6 , when the obstacle ship is located in the area S extending π / 8 to the left and right of the reference course 1 When the smart ship is in a confrontation situation,

[0014] When the obstacle ship is located in the area S behind the regional intelligent ship [5π / 8:11π / 8] 4 When the smart ship is in a state of overtaking,

[0015] The obstacle ship is located in the area S of the remaining angle range 2 , S 3 , S 5 and S 6 At the same time, there will be cross-encounter situations with smart ships.

[0016] Further: the process of improving the virtual obstacle line for ship navigation is as follows:

[0017] In the encounter situation, taking the encounter scenario as an example, the process of improving the virtual obstacle line is as follows:

[0018] First, connect to the Smart Ship Center P 0 The center of the obstacle ship O 1 , the length of the connection is recorded as H 0 When two ships meet, the maritime collision rules require the give-way ship to turn right to avoid the obstacle, so the obstacle ship passes through the center O 1 Move left and P 0 O 1 The vertical virtual obstacle line, let the other end point be P 1 , O 1 P 1 The length is denoted as H 1 , H 1 The values ​​are as follows:

[0019]

[0020] Among them, L M is the length of the smart ship; L ob is the length of the obstacle vessel;

[0021] Pass point P 1 Make O along the direction of the smart ship 1 P 1 The other end point is denoted as P 2 , line segment P 1 P 2 The length is H 2 , then H 2 The value of should meet the following conditions:

[0022] H 2 =H 0 +L ob

[0023] Pass point P 2 Make a P along the direction of the smart ship 1 P 2 The other end point is denoted as P 3 , P 2 P 3 The length is H 3 To solve the problem of invalid expansion of random tree to the rear of intelligent ship, H 3 The value should be much larger than H 1 With H 2 ;

[0024] In other situations, the principle of obstacle line design is the same as the above process.

[0025] Further: the process of establishing the obstacle detection sector strategy to cope with the scenario of multiple dynamic obstacle ships is as follows:

[0026] Consider that the intelligent ship encounters multiple dynamic obstacle ships during navigation;

[0027] First, the obstacle detection area is established, with the predicted position P of the smart ship as the fan center, and the fan-shaped area angle is defined as The smart ship divides this fan-shaped area into two equal parts along its speed direction.

[0028] Secondly, the distance between each obstacle ship and the smart ship is i Perform calculations;

[0029] Then, set the critical distance of dangerous dynamic obstacles dist min , and define the area within the critical range as the danger zone, dist min It should be adjusted according to the environment where the smart ship is located. When the smart ship is in an area with dense obstacles, the distance should be increased. min To expand the danger zone and improve the dynamic obstacle avoidance capability of the smart ship in a complex environment; then, determine whether the distance between the dynamic obstacle ship entering the danger zone and the smart ship is decreasing; when the distance between the dynamic obstacle ship entering the danger zone and the smart ship is decreasing, adjust the track of the smart ship to avoid collision;

[0030] Establish a ship collision risk model to reflect the collision risk between the ship and each obstacle ship;

[0031] Define the coordinates of the smart ship and the obstacle ship in the earth coordinate system as (x m ,y m )、(x T ,y T ); draw the true speed vector line of the intelligent ship at the center position of the obstacle ship, so as to obtain the relative vector line between the two ships;

[0032] Then draw a perpendicular line to this relative motion line at the center of the smart ship. The intersection of the perpendicular line and the relative motion line represents the moment when the two ships are closest. The distance from the foot of the perpendicular to the center of the smart ship is the closest encounter point distance DCPA. The closest encounter time TCPA is obtained according to the relative speed of the two ships.

[0033] Further: The coordinates of the smart ship and the obstacle ship in the earth coordinate system are defined as (x m ,y m )、(x T ,y T); draw the true vector line of the speed of the smart ship at the center of the obstacle ship, so as to obtain the relative vector line between the two ships; then draw a perpendicular line to this relative motion line at the center of the smart ship, and the intersection of the perpendicular line and the relative motion line represents the moment when the two ships are closest, where the distance from the foot of the perpendicular to the center of the smart ship is the closest encounter point distance DCPA. According to the relative speed of the two ships, the relevant parameters of the closest encounter time TCPA are obtained and the solution process is as follows:

[0034] (1) Intelligent ship motion vector:

[0035] The speed is:

[0036]

[0037] Among them, U x with U y are the components of the smart ship speed on the x-axis and y-axis respectively;

[0038] Sailing directions:

[0039]

[0040] Among them, the angle component α of the smart ship’s navigation direction 0 The form is as follows:

[0041]

[0042] (2) Obstacle ship motion vector:

[0043] The speed is:

[0044]

[0045] in, and are the x-axis and y-axis components of the obstacle ship's velocity respectively

[0046] Sailing directions:

[0047]

[0048] Among them, the angle component α of the obstacle ship’s navigation direction 1 The form is as follows:

[0049]

[0050] (3) Relative motion vector:

[0051] The relative speed between the smart ship and the obstacle ship is:

[0052]

[0053] in, and are the components of the relative velocity on the x-axis and y-axis respectively, and It is expressed as:

[0054]

[0055] The relative heading is:

[0056]

[0057] Among them, the relative heading angle component α between the intelligent ship and the obstacle ship 2 The form is as follows:

[0058]

[0059] The relative distance between the smart ship and the obstacle ship is;

[0060]

[0061] The true position of the smart ship relative to the obstacle ship is:

[0062]

[0063] Among them, α 3 The form is:

[0064]

[0065] The relative positions of the obstacle ship and the intelligent ship are:

[0066] θ T =α T -φ

[0067] DCPA is:

[0068] DCPA=D R sin(φ R -α T -π)

[0069] TCPA is:

[0070]

[0071] The collision risk CRI estimated based on the DCPA and TCPA of the obstacle ship i for.

[0072]

[0073] Further: the obstacle exploration sector In the following form:

[0074]

[0075] Further: also includes

[0076] When the critical danger distance between two obstacle ships is dist ob , the corresponding coordinates of the obstacle ship a are (x a ,y a ), the speed is (U xa ,U ya ), the corresponding critical safety distance is R a ; The corresponding coordinates of the obstacle ship b are (x b ,y b ), the speed is (U xb ,U yb ), the corresponding critical safety distance is R b , when the distance between the two obstacle ships is less than dist ob When , the two obstacle ships at this time are merged into a new obstacle ship. The critical safety distance R′ corresponding to the new obstacle ship is:

[0077]

[0078] The corresponding coordinates of the new obstacle ship are Speed

[0079] Further: the process of optimizing the intelligent ship autonomous navigation track planning by using the evaluation function to obtain the optimal intelligent ship autonomous navigation track is as follows:

[0080] When the dynamic window method is used for speed sampling, the speed of the smart ship is subject to the following constraints:

[0081] The maximum and minimum speed range constraints that can be provided by the intelligent ship power system are as follows:

[0082] V i ={(U i ,r Wi )∣U i ∈[U min ,U max ],r Wi ∈[r Wmin ,r Wmax ]}

[0083] Among them, U min is the minimum speed of the smart ship; U max is the maximum speed of the smart ship; r Wmin is the minimum angular velocity of the smart ship; r Wmax V is the maximum angular velocity of the smart ship;i is the maximum and minimum speed range constraints that the intelligent ship power system can provide, U i and r Wi are the velocity and angular velocity that satisfy the constraints respectively;

[0084] The speed range V of the actual achievable speed range of intelligent ships j The constraints are as follows:

[0085]

[0086] Among them, U j and r Wj They are the speed intervals V that meet the actual achievable speed range of intelligent ships. j The speed and angular velocity of the constraints; U c is the current speed of the smart ship; is the maximum acceleration of the smart ship; is the maximum deceleration of the smart ship; r Wc is the current angular velocity of the smart ship; It is the minimum angular acceleration of smart ships; is the maximum angular deceleration of the smart ship;

[0087] The constraints that the smart ship has enough time to reduce its speed and angular velocity to zero before encountering an obstacle are as follows:

[0088]

[0089] Among them, dist is the shortest distance between the corresponding simulated trajectory and the obstacle at the current speed of the intelligent ship;

[0090] U k and r Wk are the speed and angular velocity respectively satisfying the constraint that the intelligent ship has enough time to reduce the speed and angular velocity to zero before encountering an obstacle; is the maximum deceleration of the smart ship; It is the minimum angular acceleration of smart ships;

[0091] Based on different speed and angular velocity sampling combinations and combined with the forward simulation time, the predicted navigation trajectory in the next cycle is derived respectively, and then each trajectory is comprehensively evaluated using the evaluation function.

[0092] The basic form of the evaluation function is as follows:

[0093] G(U,r W )=α·heading(U,r W )+β·driftangle(U,r W)+γ·distance(U,r W )+κ·velocity(U,r W )+λ·palstance(U,r W )

[0094] Where α, β, γ, κ and λ are the weight coefficients of each evaluation function, palstance(U,r W ) is the evaluation function of the yaw angular velocity; driftangle(U,r W ) is the evaluation function of the sideslip angle σ, heading(U,r W ) is the deflection angle evaluation function; distance(U,r W ) is the distance evaluation function, which indicates the shortest distance between the trajectory corresponding to the current sampling speed and the dynamic obstacle; velocity(U,r W ) is the speed evaluation function,

[0095] Deflection angle evaluation function heading(U,r W ) is designed as follows:

[0096] heading(U,r W )=π-Δζ

[0097] Where Δζ is the angular error between the position direction of the trajectory end point generated at the current sampling speed within the sampling period and the line connecting the target point;

[0098] Among them, the bow angular velocity evaluation function palstance(U,r W ) is designed as follows:

[0099] palstance(U,r W )=(r Wmax -|r W |) / r Wmax

[0100] In the formula, r W is the current bow angular velocity of the smart ship; r Wmax is the maximum bow angular velocity of the smart ship;

[0101] Side slip angle evaluation function driftangle(U,r W ) is designed as follows:

[0102] driftangle(U,r W )=π-|σ|

[0103] The present invention provides a method for autonomous navigation track planning of intelligent ships based on maritime collision avoidance rules, which has the following advantages:

[0104] First, compared with the existing planning method that considers all encountering ships at the same time, a trajectory planning strategy based on the collision risk model is established to give priority to the most dangerous obstacles. By evaluating the collision risk faced by ships during navigation, the most dangerous obstacle ship is quickly identified and the corresponding collision avoidance trajectory is planned, which enables safe collision avoidance of intelligent ships in complex situations where multiple ships encounter each other, and improves the computing efficiency in complex scenarios.

[0105] Second, compared with the existing planning method based on a single virtual obstacle, a trajectory planning strategy based on multiple virtual obstacle lines was established. The virtual obstacles were simplified into virtual obstacle lines without width, which reduced the computational complexity of collision detection using the fast random expansion tree and improved the efficiency of collision detection. Corresponding virtual obstacle lines were designed for encounter, overtaking, and crossing situations, solving the problem that the generated tree in the existing method did not strictly comply with maritime collision avoidance rules.

[0106] Third, compared with the existing planning method based on maritime collision avoidance rules, the present invention introduces the dynamic window method, takes the ship status as an indicator to construct an evaluation function, searches for the optimal solution in the feasible safe speed set, and realizes the optimization of the safe trajectory. BRIEF DESCRIPTION OF THE DRAWINGS

[0107] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0108] Figure 1 is a flow chart of the present invention;

[0109] Figure 2 Schematic diagram I of the virtual obstacle line proposed by the present invention;

[0110] Figure 3 Schematic diagram II of the virtual obstacle line proposed by the present invention;

[0111] Figure 4 Schematic diagram III of the virtual obstacle line proposed by the present invention;

[0112] Figure 5 Schematic diagram IV of the virtual obstacle line proposed by the present invention;

[0113] Figure 6 This is a schematic diagram of the obstacle avoidance process in a multi-ship encounter situation where an intelligent ship encounters obstacle ship 2, obstacle ship 4, and obstacle ship 5 at the same time, and the intelligent ship turns right to avoid the obstacle;

[0114] Figure 7Schematic diagram of the obstacle avoidance process in a multi-ship encounter situation where an intelligent ship encounters obstacle ship 6 and obstacle ship 8 at the same time and the intelligent ship turns right to avoid the obstacle;

[0115] Figure 8 Schematic diagram of the obstacle avoidance process in a multi-ship encounter situation where an intelligent ship encounters obstacle ship 6 and obstacle ship 7 at the same time and the intelligent ship turns right to avoid the obstacle;

[0116] Fig. 9 Schematic diagram of the intelligent ship arriving at the destination during the obstacle avoidance process in a multi-ship encounter situation. DETAILED DESCRIPTION

[0117] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0118] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present invention and its application or use. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0119] Figure 1 is a flow chart of the present invention;

[0120] A method for autonomous navigation track planning of an intelligent ship based on maritime collision avoidance rules comprises the following steps:

[0121] S1: Establish a quantitative identification model for ship encounter situations to determine the encounter type between the obstacle ship and the intelligent ship;

[0122] S2: Improve the virtual obstacle line for ship navigation, detect obstacles, add inspections of maritime rules constraints, and comply with actual maritime collision avoidance rules;

[0123] S3: Establish an obstacle detection sector strategy to deal with the scenario of multiple dynamic obstacle ships, use the dangerous distance as the dynamic obstacle ship avoidance circle, evaluate the risk value of all dynamic obstacles in the dynamic obstacle avoidance circle, and set the intelligent ship to perform collision avoidance operations only for the most dangerous obstacle ship with the highest risk to realize autonomous navigation track planning of the intelligent ship;

[0124] S4: The evaluation function is used to optimize the intelligent ship autonomous navigation trajectory planning to obtain the optimal intelligent ship autonomous navigation trajectory.

[0125] The steps S1 / S2 / S3 / S4 are performed sequentially;

[0126] Further: the process of establishing a quantitative identification model for ship encounter situations and determining the encounter type between the obstacle ship and the intelligent ship is as follows;

[0127] Taking the location of the smart ship as the center and heading 000° as the reference, the encounter area is divided into six parts: 1 , S 2 , S 3 , S 4 , S 5 , S 6 , when the obstacle ship is located in the area S extending π / 8 to the left and right of the reference course 1 When the smart ship is in a confrontation situation,

[0128] When the obstacle ship is located in the area S behind the regional intelligent ship [5π / 8:11π / 8] 4 When the smart ship is in a state of overtaking,

[0129] The obstacle ship is located in the area S of the remaining angle range 2 , S 3 , S 5 and S 6 At the same time, there will be cross-encounter situations with smart ships.

[0130] Further: the process of improving the virtual obstacle line for ship navigation is as follows:

[0131] In the encounter situation, taking the encounter scenario as an example, the process of improving the virtual obstacle line is as follows:

[0132] First, connect to the Smart Ship Center P 0 The center of the obstacle ship O 1 , the length of the connection is recorded as H 0 When two ships meet, the maritime collision rules require the give-way ship to turn right to avoid the obstacle, so the obstacle ship passes through the center O 1 Move left and P 0 O 1 The vertical virtual obstacle line, let the other end point be P 1 , O 1 P 1 The length is denoted as H 1 , H 1 The values ​​are as follows:

[0133]

[0134] Among them, L M is the length of the smart ship; L ob is the length of the obstacle vessel;

[0135] Pass point P 1 Make O along the direction of the smart ship 1 P 1 The other end point is denoted as P 2 , line segment P 1 P 2 The length is H 2 , then H 2 The value of should meet the following conditions:

[0136] H 2 =H 0 +L ob (2)

[0137] Pass point P 2 Make a P along the direction of the smart ship 1 P 2 The other end point is denoted as P 3 , P 2 P 3 The length is H 3 To solve the problem of invalid expansion of random tree to the rear of intelligent ship, H 3 The value should be much larger than H 1 With H 2 ;

[0138] In other situations, the principle of obstacle line design is the same as the above process.

[0139] Further: the process of establishing the obstacle detection sector strategy to cope with the scenario of multiple dynamic obstacle ships is as follows:

[0140] Consider that the intelligent ship encounters multiple dynamic obstacle ships during navigation;

[0141] First, the obstacle detection area is established, with the predicted position P of the smart ship as the fan center, and the fan-shaped area angle is defined as The smart ship divides this fan-shaped area into two equal parts along its speed direction.

[0142] Secondly, the distance between each obstacle ship and the smart ship is i Perform calculations;

[0143] Then, set the critical distance of dangerous dynamic obstacles dist min , and define the area within the critical range as the danger zone, dist min It should be adjusted according to the environment where the smart ship is located. When the smart ship is in an area with dense obstacles, the distance should be increased. minTo expand the danger zone and improve the dynamic obstacle avoidance capability of the smart ship in a complex environment; then, determine whether the distance between the dynamic obstacle ship entering the danger zone and the smart ship is decreasing; when the distance between the dynamic obstacle ship entering the danger zone and the smart ship is decreasing, adjust the track of the smart ship to avoid collision;

[0144] Establish a ship collision risk model to reflect the collision risk between the ship and each obstacle ship;

[0145] Define the coordinates of the smart ship and the obstacle ship in the earth coordinate system as (x m ,y m )、(x T ,y T ); draw the true speed vector line of the intelligent ship at the center position of the obstacle ship, so as to obtain the relative vector line between the two ships;

[0146] Then draw a perpendicular line to this relative motion line at the center of the smart ship. The intersection of the perpendicular line and the relative motion line represents the moment when the two ships are closest. The distance from the foot of the perpendicular to the center of the smart ship is the closest encounter point distance DCPA. The closest encounter time TCPA is obtained according to the relative speed of the two ships.

[0147] Further: The coordinates of the smart ship and the obstacle ship in the earth coordinate system are defined as (x m ,y m )、(x T ,y T ); draw the true vector line of the speed of the smart ship at the center of the obstacle ship, so as to obtain the relative vector line between the two ships; then draw a perpendicular line to this relative motion line at the center of the smart ship, and the intersection of the perpendicular line and the relative motion line represents the moment when the two ships are closest, where the distance from the foot of the perpendicular to the center of the smart ship is the closest encounter point distance DCPA. According to the relative speed of the two ships, the relevant parameters of the closest encounter time TCPA are obtained and the solution process is as follows:

[0148] (1) Intelligent ship motion vector:

[0149] The speed is:

[0150]

[0151] Among them, U x with U y are the components of the smart ship speed on the x-axis and y-axis respectively;

[0152] Sailing directions:

[0153]

[0154] Among them, the angle component α of the smart ship’s navigation direction0 The form is as follows:

[0155]

[0156] (2) Obstacle ship motion vector:

[0157] The speed is:

[0158]

[0159] in, and are the x-axis and y-axis components of the obstacle ship's velocity respectively

[0160] Sailing directions:

[0161]

[0162] Among them, the angle component α of the obstacle ship’s navigation direction 1 The form is as follows:

[0163]

[0164] (3) Relative motion vector:

[0165] The relative speed between the smart ship and the obstacle ship is:

[0166]

[0167] in, and are the components of the relative velocity on the x-axis and y-axis respectively, and It is expressed as:

[0168]

[0169] The relative heading is:

[0170]

[0171] Among them, the relative heading angle component α between the intelligent ship and the obstacle ship 2 The form is as follows:

[0172]

[0173] The relative distance between the smart ship and the obstacle ship is;

[0174]

[0175] The true position of the smart ship relative to the obstacle ship is:

[0176]

[0177] Among them, α 3 The form is:

[0178]

[0179] The relative positions of the obstacle ship and the intelligent ship are:

[0180] θ T =α T -φ (16)

[0181] DCPA is:

[0182] DCPA=D R sin(φ R -α T -π) (17)

[0183] TCPA is:

[0184]

[0185] The collision risk CRI estimated based on the DCPA and TCPA of the obstacle ship i for.

[0186]

[0187] Further: the obstacle exploration sector In the following form:

[0188]

[0189] The method further includes:

[0190] When the critical danger distance between two obstacle ships is dist ob , the corresponding coordinates of the obstacle ship a are (x a ,y a ), the speed is (U xa ,U ya ), the corresponding critical safety distance is R a ; The corresponding coordinates of the obstacle ship b are (x b ,y b ), the speed is (U xb ,U yb ), the corresponding critical safety distance is R b , when the distance between the two obstacle ships is less than dist ob When , the two obstacle ships at this time are merged into a new obstacle ship. At this time, the critical safety distance R′ corresponding to the new obstacle ship is:

[0191]

[0192] The corresponding coordinates of the new obstacle ship are Speed

[0193] Further: the process of optimizing the intelligent ship autonomous navigation track planning by using the evaluation function to obtain the optimal intelligent ship autonomous navigation track is as follows:

[0194] When the dynamic window method is used for speed sampling, the speed of the smart ship is subject to the following constraints:

[0195] The maximum and minimum speed range constraints that can be provided by the intelligent ship power system are as follows:

[0196] V i ={(U i ,r Wi )∣U i ∈[U min ,U max ],r Wi ∈[r Wmin ,r Wmax ]}

[0197] Among them, U min is the minimum speed of the smart ship; U max is the maximum speed of the smart ship; r Wmin is the minimum angular velocity of the smart ship; r Wmax V is the maximum angular velocity of the smart ship; i is the maximum and minimum speed range constraints that the intelligent ship power system can provide, U i and r Wi are the velocity and angular velocity that satisfy the constraints respectively;

[0198] The speed range V of the actual achievable speed range of intelligent ships j The constraints are as follows:

[0199]

[0200] Among them, U j and r Wj They are the speed intervals V that meet the actual achievable speed range of intelligent ships. j The speed and angular velocity of the constraints; U c is the current speed of the smart ship; is the maximum acceleration of the smart ship; is the maximum deceleration of the smart ship; r Wc is the current angular velocity of the smart ship; It is the minimum angular acceleration of smart ships; is the maximum angular deceleration of the smart ship;

[0201] The constraints that the smart ship has enough time to reduce its speed and angular velocity to zero before encountering an obstacle are as follows:

[0202]

[0203] Among them, dist is the shortest distance between the corresponding simulated trajectory and the obstacle at the current speed of the intelligent ship;

[0204] U k and r Wk are the speed and angular velocity respectively satisfying the constraint that the intelligent ship has enough time to reduce the speed and angular velocity to zero before encountering an obstacle; is the maximum deceleration of the smart ship; It is the minimum angular acceleration of smart ships;

[0205] Based on different speed and angular velocity sampling combinations and combined with the forward simulation time, the predicted navigation trajectory in the next cycle is derived respectively, and then each trajectory is comprehensively evaluated using the evaluation function.

[0206] Further: The basic form of the evaluation function is as follows:

[0207] G(U,r W )=α·heading(U,r W )+β·driftangle(U,r W )+γ·distance(U,r W )+κ·velocity(U,r W )+λ·palstance(U,r W )

[0208] Where α, β, γ, κ and λ are the weight coefficients of each evaluation function, palstance(U,r W ) is the evaluation function of the yaw angular velocity; driftangle(U,r W ) is the evaluation function of the sideslip angle σ, heading(U,r W ) is the deflection angle evaluation function; distance(U,r W ) is the distance evaluation function, which indicates the shortest distance between the trajectory corresponding to the current sampling speed and the dynamic obstacle; velocity(U,r W ) is the speed evaluation function,

[0209] Deflection angle evaluation function heading(U,r W ) is designed as follows:

[0210] heading(U,r W )=π-Δζ

[0211] Where Δζ is the angular error between the position direction of the trajectory end point generated at the current sampling speed within the sampling period and the line connecting the target point;

[0212] Among them, the bow angular velocity evaluation function palstance(U,r W ) is designed as follows:

[0213] palstance(U,r W )=(r Wmax -|r W |) / r Wmax

[0214] In the formula, r W is the current bow angular velocity of the smart ship; r Wmax is the maximum bow angular velocity of the smart ship;

[0215] Side slip angle evaluation function driftangle(U,r W ) is designed as follows:

[0216] driftangle(U,r W )=π-|σ|

[0217] The present invention is further described below in conjunction with the accompanying drawings. Figure 2-5 It is a schematic diagram of the virtual obstacle line proposed by the present invention. Considering that the ship encounter occurs in open waters, there are few static obstacle ships in open waters, and the number of static obstacles in the test environment is reduced. Set 8 obstacle ships, assuming that the obstacle ships do not have the ability to avoid obstacles and all move in a uniform straight line.

[0218] exist Figure 6 In the figure, the smart ship has sailed a short distance towards the destination. If it continues to sail in this direction, the smart ship will form a left crossing situation with obstacle ship 2, a face-to-face situation with obstacle ship 4, and a right crossing situation with obstacle ship 5. According to the proposed obstacle avoidance strategy, obstacle ship 5 is the most dangerous obstacle ship. Therefore, a virtual obstacle line is established for obstacle ship 5 in the face-to-face situation to guide the smart ship to turn right and avoid it according to the maritime collision avoidance rules.

[0219] exist Figure 7 In the figure, the smart ship and obstacle ship 6 are in an overtaking situation, and the smart ship and obstacle ship 8 are in a right crossing situation. According to the proposed obstacle avoidance strategy, obstacle ship 6 is the most dangerous obstacle ship. Therefore, a virtual obstacle line corresponding to the overtaking situation is established for obstacle ship 6, and the smart ship is guided to turn right to avoid it according to the maritime collision avoidance rules.

[0220] exist Figure 8In the figure, the smart ship will overtake both obstacle ships 6 and 7. According to the guidance of maritime collision avoidance rules, when the smart ship avoids collision with obstacle ship 6, the smart ship is not a give-way ship, and when the smart ship avoids collision with obstacle ship 7, the smart ship is a give-way ship. Therefore, in this situation, it is only necessary to establish a virtual obstacle line for the overtaking situation for obstacle ship 7 to guide the smart ship to turn right and avoid it.

[0221] Fig. 9 It is a schematic diagram of the intelligent ship arriving at the destination. The results show that in a multi-ship encounter situation, the present invention can correctly analyze the current situation of the intelligent ship and the obstacle ship, and can take avoidance actions that comply with the rules to avoid collision.

[0222] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for autonomous navigation trajectory planning of intelligent ships based on maritime collision avoidance rules. Features: The following steps are involved: Establish a quantitative identification model for ship encounter situations to determine the encounter types between obstacle ships and intelligent ships; Improve the virtual obstacle line for ship navigation, detect obstacles, add inspections of maritime rules constraints, and comply with actual maritime collision avoidance rules; Establish an obstacle detection sector strategy to deal with the scenario of multiple dynamic obstacle ships, use the dangerous distance as the dynamic obstacle ship avoidance circle, evaluate the risk value of all dynamic obstacles in the dynamic obstacle avoidance circle, and set the intelligent ship to perform collision avoidance operations only for the most dangerous obstacle ship with the highest risk, so as to realize the autonomous navigation track planning of the intelligent ship; The evaluation function is used to optimize the intelligent ship's autonomous navigation trajectory planning to obtain the optimal intelligent ship's autonomous navigation trajectory.

2. According to claim 1, a method for autonomous navigation track planning of an intelligent ship based on maritime collision avoidance rules, Features: The process of establishing a quantitative identification model for ship encounter situations and determining the encounter type between an obstacle ship and an intelligent ship is as follows: Taking the location of the smart ship as the center and heading 000° as the reference, the encounter area is divided into six parts: 1 , S 2 , S 3 , S 4 , S 5 , S 6 , when the obstacle ship is located in the area S extending π / 8 to the left and right of the reference course 1 When the smart ship is in a confrontation situation, When the obstacle ship is located in the area S behind the regional intelligent ship [5π / 8:11π / 8] 4 When the smart ship is in a state of overtaking, The obstacle ship is located in the area S of the remaining angle range 2 , S 3 , S 5 and S 6 At the same time, there will be cross-encounter situations with smart ships.

3. According to the method for autonomous navigation track planning of intelligent ships based on maritime collision avoidance rules as described in claim 1, Features: The process of improving the virtual obstacle line for ship navigation is as follows: In the encounter situation, taking the encounter scenario as an example, the process of improving the virtual obstacle line is as follows: First, connect to the Smart Ship Center P 0 The center of the obstacle ship O 1 , the length of the connection is recorded as H 0 When two ships meet, the maritime collision rules require the give-way ship to turn right to avoid the obstacle, so the obstacle ship passes through the center O 1 Move left and P 0 O 1 The vertical virtual obstacle line, let the other end point be P 1 , O 1 P 1 The length is denoted as H 1 , H 1 The values ​​are as follows: Among them, l M is the length of the smart ship; L ob is the length of the obstacle vessel; Pass point P 1 Make O along the direction of the smart ship 1 P 1 The other end point is denoted as P 2 , line segment P 1 P 2 The length is H 2 , then H 2 The value of should meet the following conditions: H 2 =H 0 +L ob (2) Pass point P 2 Make a P along the direction of the smart ship 1 P 2 The other end point is denoted as P 3 , P 2 P 3 The length is H 3 To solve the problem of invalid expansion of random tree to the rear of intelligent ship, H 3 The value should be much larger than H 1 With H 2 ; In other situations, the principle of obstacle line design is the same as the above process.

4. According to the method for autonomous navigation track planning of intelligent ships based on maritime collision avoidance rules as described in claim 1, Features: The process of establishing the obstacle detection sector strategy to cope with the scenario of multiple dynamic obstacle ships is as follows: Consider that the intelligent ship encounters multiple dynamic obstacle ships during navigation; First, the obstacle detection area is established, with the predicted position P of the smart ship as the fan center, and the fan-shaped area angle is defined as The smart ship divides this fan-shaped area into two equal parts along its speed direction. Secondly, the distance between each obstacle ship and the smart ship is i Perform calculations; Then, set the critical distance of dangerous dynamic obstacles dist min , and define the area within the critical range as the danger zone, dist min It should be adjusted according to the environment where the smart ship is located. When the smart ship is in an area with dense obstacles, the distance should be increased. min To expand the danger zone and improve the dynamic obstacle avoidance capability of the smart ship in a complex environment; then, determine whether the distance between the dynamic obstacle ship entering the danger zone and the smart ship is decreasing; when the distance between the dynamic obstacle ship entering the danger zone and the smart ship is decreasing, adjust the track of the smart ship to avoid collision; Establish a ship collision risk model to reflect the collision risk between the ship and each obstacle ship; Define the coordinates of the smart ship and the obstacle ship in the earth coordinate system as (x m ,y m )、(x T ,y T ); draw the true speed vector line of the intelligent ship at the center position of the obstacle ship, so as to obtain the relative vector line between the two ships; Then draw a perpendicular line to this relative motion line at the center of the smart ship. The intersection of the perpendicular line and the relative motion line represents the moment when the two ships are closest. The distance from the foot of the perpendicular to the center of the smart ship is the closest encounter point distance DCPA. The closest encounter time TCPA is obtained according to the relative speed of the two ships.

5. According to the method for autonomous navigation track planning of intelligent ships based on maritime collision avoidance rules as described in claim 1, Features: The coordinates of the intelligent ship and the obstacle ship in the earth coordinate system are defined as (x m ,y m )、(x T ,y T ); draw the true vector line of the speed of the smart ship at the center of the obstacle ship, so as to obtain the relative vector line between the two ships; then draw a perpendicular line to this relative motion line at the center of the smart ship, and the intersection of the perpendicular line and the relative motion line represents the moment when the two ships are closest, where the distance from the foot of the perpendicular to the center of the smart ship is the closest encounter point distance DCPA. According to the relative speed of the two ships, the relevant parameters of the closest encounter time TCPA are obtained and the solution process is as follows: (1) Intelligent ship motion vector: The speed is: Among them, U x with U y are the components of the smart ship speed on the x-axis and y-axis respectively; Sailing directions: Among them, the angle component α of the smart ship’s navigation direction 0 The form is as follows: (2) Obstacle ship motion vector: The speed is: in, and are the x-axis and y-axis components of the obstacle ship's velocity respectively Sailing directions: Among them, the angle component α of the obstacle ship’s navigation direction 1 The form is as follows: (3) Relative motion vector: The relative speed between the smart ship and the obstacle ship is: in, and are the components of the relative velocity on the x-axis and y-axis respectively, and It is expressed as: The relative heading is: Among them, the relative heading angle component α between the intelligent ship and the obstacle ship 2 The form is as follows: The relative distance between the smart ship and the obstacle ship is; The true position of the smart ship relative to the obstacle ship is: Among them, α 3 The form is: The relative positions of the obstacle ship and the intelligent ship are: i T =a T -f (16) DCPA is: DCPA=D R sin(φ R -a T -p) (17) TCPA is: The collision risk CRI estimated based on the DCPA and TCPA of the obstacle ship i for.

6. According to the method for autonomous navigation track planning of intelligent ships based on maritime collision avoidance rules as described in claim 1, Features: The obstacle exploration sector In the following form:

7. According to claim 1, a method for autonomous navigation track planning of intelligent ships based on maritime collision avoidance rules, Features: Also includes When the critical danger distance between two obstacle ships is dist ob , the corresponding coordinates of the obstacle ship a are (x a ,y a ), the speed is (U xa , U ya ), the corresponding critical safety distance is R a ; The corresponding coordinates of the obstacle ship b are (x b ,y b ), the speed is (U xb , U yb ), the corresponding critical safety distance is R b , when the distance between the two obstacle ships is less than dist ob When , the two obstacle ships at this time are merged into a new obstacle ship. At this time, the critical safety distance R′ corresponding to the new obstacle ship is: The corresponding coordinates of the new obstacle ship are Speed 8. According to the method for autonomous navigation track planning of intelligent ships based on maritime collision avoidance rules as described in claim 1, Features: The process of optimizing the intelligent ship autonomous navigation track planning by using the evaluation function to obtain the optimal intelligent ship autonomous navigation track is as follows: When the dynamic window method is used for speed sampling, the speed of the smart ship is subject to the following constraints: The maximum and minimum speed range constraints that can be provided by the intelligent ship power system are as follows: V i ={(U i ,r Wi )∣U i ∈[U min ,U max ],r Wi ∈[r Wmin ,r Wmax ]} Among them, U min is the minimum speed of the smart ship; U max is the maximum speed of the smart ship; r Wmin is the minimum angular velocity of the smart ship; r Wmax V is the maximum angular velocity of the smart ship; i is the maximum and minimum speed range constraints that the intelligent ship power system can provide, U i and r Wi are the velocity and angular velocity that satisfy the constraints respectively; The speed range V of the actual achievable speed range of intelligent ships j The constraints are as follows: V j =(U j ,r Wj )∣U j ∈[U c +a Umax Δt,U c +a Umin Δt], Among them, U j and r Wj They are the speed intervals V that meet the actual achievable speed range of intelligent ships. j The speed and angular velocity of the constraints; U c is the current speed of the smart ship; is the maximum acceleration of the smart ship; is the maximum deceleration of the smart ship; r Wc is the current angular velocity of the smart ship; It is the minimum angular acceleration of smart ships; is the maximum angular deceleration of the smart ship; The constraints that the smart ship has enough time to reduce its speed and angular velocity to zero before encountering an obstacle are as follows: Among them, dist is the shortest distance between the corresponding simulated trajectory and the obstacle at the current speed of the intelligent ship; U k and r Wk are the speed and angular velocity respectively satisfying the constraint that the intelligent ship has enough time to reduce the speed and angular velocity to zero before encountering an obstacle; is the maximum deceleration of the smart ship; It is the minimum angular acceleration of smart ships; Based on different speed and angular velocity sampling combinations and combined with the forward simulation time, the predicted navigation trajectory in the next cycle is derived respectively, and then each trajectory is comprehensively evaluated using the evaluation function.

9. According to claim 1, a method for autonomous navigation track planning of intelligent ships based on maritime collision avoidance rules, Features: The basic form of the evaluation function is as follows: G(U,r W )=α·heading(U,r W )+β·driftangle(U,r W )+γ·distance(U,r W )+k·velocity(U,r W )+λ·palstance(U,r W ) Where α, β, γ, k and λ are the weight coefficients of each evaluation function, palstance(U,r W ) is the evaluation function of the yaw angular velocity; driftangle(U,r W ) is the evaluation function of the sideslip angle σ, heading(U,r W ) is the deflection angle evaluation function; distance(U,r W ) is the distance evaluation function, which indicates the shortest distance between the trajectory corresponding to the current sampling speed and the dynamic obstacle; velocity(U,r W ) is the speed evaluation function, Deflection angle evaluation function heading(U,r W ) is designed as follows: heading(U,r W )=π-Δζ Where Δζ is the angular error between the position direction of the trajectory end point generated at the current sampling speed within the sampling period and the line connecting the target point; Among them, the bow angular velocity evaluation function palstance(U,r W ) is designed as follows: palstance(U,r W )=(r Wmax -|r W |) / r Wmax In the formula, r W is the current bow angular velocity of the smart ship; r Wmax is the maximum bow angular velocity of the smart ship; Side slip angle evaluation function driftangle(U,r W ) is designed as follows: driftangle(U,r W )=π-|σ|.

Citation Information

Cited By

  • Intelligent decision-making method based on maritime collision avoidance rule

    CN120823729A

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

    CN121209280A