CN115268472A - Intelligent ship navigation collision avoidance behavior coding method based on machine language expression - Google Patents
Intelligent ship navigation collision avoidance behavior coding method based on machine language expression Download PDFInfo
- Publication number
- CN115268472A CN115268472A CN202211185837.9A CN202211185837A CN115268472A CN 115268472 A CN115268472 A CN 115268472A CN 202211185837 A CN202211185837 A CN 202211185837A CN 115268472 A CN115268472 A CN 115268472A
- Authority
- CN
- China
- Prior art keywords
- ship
- speed
- collision avoidance
- target
- behavior
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 230000006399 behavior Effects 0.000 claims abstract description 104
- 230000009471 action Effects 0.000 claims description 49
- 238000012360 testing method Methods 0.000 claims description 19
- 238000004458 analytical method Methods 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 9
- 230000001939 inductive effect Effects 0.000 claims description 7
- 230000008859 change Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 claims description 3
- 230000007717 exclusion Effects 0.000 claims description 3
- 230000004927 fusion Effects 0.000 claims description 2
- 230000001105 regulatory effect Effects 0.000 claims description 2
- LFULEKSKNZEWOE-UHFFFAOYSA-N propanil Chemical compound CCC(=O)NC1=CC=C(Cl)C(Cl)=C1 LFULEKSKNZEWOE-UHFFFAOYSA-N 0.000 claims 1
- 238000013461 design Methods 0.000 abstract description 6
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 101150080778 INPP5D gene Proteins 0.000 description 292
- 230000001133 acceleration Effects 0.000 description 8
- 230000000875 corresponding effect Effects 0.000 description 8
- 238000011161 development Methods 0.000 description 7
- 238000013139 quantization Methods 0.000 description 5
- 230000009467 reduction Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000013459 approach Methods 0.000 description 2
- 230000003542 behavioural effect Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000000638 solvent extraction Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/0206—Control of position or course in two dimensions specially adapted to water vehicles
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a method for coding intelligent ship navigation collision avoidance behaviors based on machine language expression, belonging to the technical field of ship control; a ship intelligent navigation collision avoidance behavior coding method based on machine language expression adopts a 01 code and ILP method to design a coding form of a ship driving behavior index and a ship state index, preliminarily constructs a ship driving machine language system with universality and practical application, and is beneficial to providing a universal coding form for ship machine driving: a universal and practical ship driving machine language system is explored and constructed, instruction requirements of driving behaviors and ship states are analyzed according to 3 ship navigation collision avoidance behaviors, and coding forms of ship driving behavior indexes and ship state indexes are innovatively designed by adopting a 01 code and ILP method.
Description
Technical Field
The invention relates to the technical field of ship control, in particular to a ship intelligent navigation collision avoidance behavior coding method based on machine language expression.
Background
The intelligent navigation of the ship is an inevitable trend of future development of the shipping industry and is also a focus of international intelligent ship technical competition in recent years. The method not only can solve the challenge of further cross-over improvement of shipping efficiency, safety and economy, but also is actively integrated into the future intelligent world to realize the necessary way and trend of shipping industry transition and upgrade. The development of the intelligent navigation technology of the ship is not only a significant historical opportunity for China to lead the development of the international intelligent ship, make leading international shipping rules, promote the international shipping status and guarantee the national safety, but also an objective requirement for the national development, particularly the construction of ocean strong countries, traffic strong countries and shipping strong countries.
According to the grade division standard of the current intelligent ship navigation system, the intelligent ship navigation system can be divided into 10 grades from auxiliary driving to autonomous driving. The grade evaluation of the ship intelligent navigation system of the existing coastal intelligent container commercial ship in China is rated as coastal R11 grade (high remote control driving grade with people on the ship) and conditional A2 grade (middle autonomous driving grade). It is expected that with the increase of the number of intelligent navigation ships in China, the intelligent level of a navigation system of the intelligent navigation ship is required to be continuously improved, human decision and control are gradually changed into human decision machine control, and then autonomous decision and control of the machine are realized. In the process, how to make the machine drive transfer the excellent driving experience of the captain of the human, make the instruction of the human convert into the machine language, make the future mechanism communication between the unmanned ship and the unmanned ship of the world become possible, it is the basis of the development of intelligent navigation and intelligent navigation of the ship. At present, no systematic research can sufficiently combine with a machine language expression mode to describe collision avoidance behaviors in intelligent navigation of a ship, and an efficient coding mode is adopted to support intelligent driving of the ship.
Disclosure of Invention
The invention aims to realize efficient coding of intelligent ship navigation collision avoidance behaviors so as to realize intelligent control of a ship, and provides a coding method of the intelligent ship navigation collision avoidance behaviors based on machine language expression.
In order to achieve the purpose, the invention adopts the following technical scheme:
a method for coding intelligent ship navigation collision avoidance behaviors based on machine language expression is characterized in that aiming at ship machine driving collision avoidance behaviors, under the instruction requirements of driving behaviors and ship states, a machine language is explored to code driving behavior indexes and ship state indexes to form a set of machine language coding scheme with universality and practical application, and theoretical and technical support is provided for promoting intelligent ship navigation development, and the method specifically comprises the following steps:
s1, dividing driving behavior instructions into a vehicle using and a rudder according to requirements, and specifically, dividing driving behavior instructions into a rudder rotating, a speed regulating and a reversingC b Harmony vehicleC f The rotary rudder further comprises a left rotary rudderD l Right-handed rudderD r Non-rotating rudderD n The speed regulation further comprises accelerationV i And a decelerationV d Without speed-regulating actionV n (ii) a The vehicle-using and rudder-using commands can be independently or simultaneously carried out through a command setInstDividing the driving behavior instruction into the following specific sections:
Inst=｛｛D l ，D r ，D n ｝，｛V i ，V d ，V n ｝，｛C b ，C f ｝｝（1）
s2, in the collision avoidance process of the intelligent sailing of the ship, the speed, the position coordinates and the course of the ship are obtained by means of an Automatic Identification System (AIS) and a radar, the relative Distance, the relative course and the relative speed between the ship and a target ship are calculated according to the speed, the position coordinates and the course of the ship, the Distance of the closest point of approach (DCPA) and the minimum Time of The Closest Point of Approach (TCPA) between the ship and the target ship are calculated according to the relative Distance, the relative course and the relative speed between the ship and the target ship, and the ship state instruction is divided by combining the data indexes;
and S3, aiming at 3 collision avoidance behavior indexes, namely the relative distance, the relative course and the relative speed between the ship and the target ship during the ship navigation, analyzing the instruction requirements of the driving behavior and the ship state, and designing the coding forms of the ship driving behavior indexes and the ship state indexes by adopting 01 codes and inductive logic program design coding rules.
Preferably, the division of the ship state command in S2 specifically includes the following contents:
(1) the ship has the following motion parameters: position of the vesselB x ，B y HeadingB ag And speed of flightB v ；
(2) The motion parameters of the target ship are as follows: target vessel positionT x ，T y HeadingT ag And speed of flightT v ；
(3) Ship safety state parameters: distance to meet recently (DCPA), time to meet minimum (TCPA).
Preferably, the design of the ship driving behavior index coding form by using the 01 code in S3 specifically includes the following contents:
a1, operation behavior coding: the driving behavior commands are respectively coded, wherein A represents a left-handed rudderD l B represents a right-handed rudderD r And C represents accelerationV i And D represents decelerationV d E represents backingC b F represents a main vehicleC f G represents reserved attribute behavior; a, B, C, D, E and F are all 01 codes, 0 is set to indicate that the behavior is not executed, and 1 indicates that the behavior is executed;
a2, encoding the operation behavior degree: will turn left rudderD l Right-handed rudderD r Selecting an action degree unit as an angle, and setting an interval unit as 1 degree; accelerationV i And a decelerationV d The unit of action degree is selected as a section, and the unit of interval is 0.1 section; back-upC b The unit of action degree is selected as second, and the unit of interval is 5s; vehicle for correctingC f Selecting action degree unit as minute, and interval unit as 10min; mutual exclusion exists among the operation behavior degree data, namely the single-class behavior degree data at any moment are unique;
and A3, combining the operation behavior attribute codes and the operation behavior degree codes in the A1 and the A2 to obtain a ship driving behavior index code.
Preferably, the designing of the ship state index coding form by using the inductive logic programming coding rule in S3 refers to comprehensively considering the states of the ship and the target ship and the ship attribute logic, and implementing ship state coding at each moment by using the inductive logic programming coding rule in combination with COLREGS-related regulations, and specifically includes the following contents:
b1, ship overtaking scene rules and codes: the two motor boats run in the same direction, when the speed of the rear boat is higher than that of the front boat, a overtaking situation is formed, and the overtaking situation follows the rule of COLREGS thirteenth, wherein the specific rule is as follows: in the specified range of the track to be tracked, under the 5-level sea condition, the test ship sails at the operation speed, the stern of the ship moves to the +/-67.5-degree direction within 3nm, the speed of the other ship is greater than that of the test ship to accelerate, and the tracked situation is formed; the test ship is positioned in the 3nm of the stern direction +/-67.5 degrees of the other ship, and the other ship is accelerated to be tracked;
combining the analysis, if the target ship overtakes the ship and the ship is a straight ship, keeping the direction and speed; if the ship overtakes the target ship and is a way-giving ship, steering collision avoidance control is carried out, and the specific mathematical analysis is as follows:
b1.1, two motorized ships, when the target ship tracks the ship, the ship is a straight ship, and the mathematical and logical codes are as follows:
wherein X is the ship, Y is the target ship, V t Is the target ship speed, V 0 The speed of the ship is defined, alpha is the angle of the target ship relative to the ship, and beta is the azimuth of the ship relative to the target ship; "stand-on ship (X)" indicates that the ship is a straight ship and keeps the direction and the speed; "power-drive vessels (X)" and "power-drive vessels (Y)" indicate that both vessels are motorized vessels; "fusion Risk (X, Y)" indicates that avoidance responsibility and Collision avoidance operation between the own ship and the target ship are all taken into consideration;
b1.2, when the ship tracks over the target ship, the ship is a way-giving ship and needs to carry out collision avoidance operation of left steering or right steering, and mathematical and logical codes are as follows:
wherein, the 'digit-way ship (X) represents that the ship is a way yielding ship, and the' keep court (Y) represents that the target ship is a straight ship;
b2, ship encounter scene rules and codes: when two motor-driven ships meet in opposite or nearly opposite directions to form collision danger, a meeting situation is formed, and the fourth regulation of COLREGS is followed, wherein the specific rule is as follows: under 5-level sea conditions and below, the visibility is not lower than 1km, a test ship can sail in an entrance channel correctly identified based on an electronic chart and a light buoy, and during the process, the test ship meets an oncoming ship in the opposite direction in the channel, and the channel is designed according to a double-line channel, and the meeting ship runs along the central line of the opposite channel;
the calculation rule of the width of the multi-line channel is as follows:
W=QA+b+Qc （4）
A=n（Lsin（θτ）+B）（5）
wherein, W represents the navigation width of the channel, m; q represents the number of lane lines; a represents the width of the flight path band, m; c represents the margin width between the ship and the bottom line of the channel, m; b represents the margin width between the ships, m, when the ship intersection density is large, the margin width between the ships can be increased properly; n represents the ship drift multiple; l represents the design captain, m; theta tau represents wind and flow pressure deflection angle; b represents the designed ship width, m;
combining the analysis, when a meeting local area is formed between the ship and the target ship, the ship and the target ship are both yielding ships, the two ships have the same yielding responsibility and need to carry out right-turning collision avoidance operation, and the mathematical logic codes are as follows:
b3, ship cross meeting scene rules and coding: the two ships can only see the sidelights on one side of the other side, the course lines of the two ships are crossed, the ship (the ship displaying the green sidelight) with the other ship on the starboard of the ship is the way-giving ship, the other ship (the ship displaying the red sidelight) should give way, and the other ship, namely the straight ship, should keep course and speed, and the specific rule is as follows: testing the operating speed of the ship under 5-level sea conditions and below, and sailing by other ships within 3nm of any one side of the course of the ship; the bow of the two ships is crossed, the crossing bulwark angle is larger than 5 degrees and smaller than 112.5 degrees, the ship speed of the other ship is +/-20 percent of the ship speed of the test ship, and the two ships form a crossing meeting situation;
by combining the above analysis, when the ship forms a cross meeting situation and there is a collision risk, the ship on the port side gives way to the ship on the starboard side, and assumes the avoidance responsibility, and the specific mathematical analysis is as follows:
b3.1, when the ship and the target ship form meeting collision avoidance in which port sides meet in a cross mode, the ship is a straight ship, and direction and speed are guaranteed; the target ship executes collision avoidance operation, and only when the target ship does not fulfill the responsibility and obligation of the yielding ship to carry out collision avoidance operation or form a tight office, the ship needs to carry out collision avoidance operation, and the mathematical logic codes are as follows:
b3.2, when the relative position or the bulwark angle of the coming ship is [5 degrees, 67.5 degrees ], the coming ship meets in a small-angle crossing mode, the ship is a way-giving ship, the avoiding effect of right steering is better than that of speed change, the ship needs to adopt collision avoidance operation of right steering, and mathematical and logical codes are as follows:
b3.3, starboard large-angle cross meeting means that when the bulwark angle of the target ship is [67.5, 112.5], the target ship is a straight ship and keeps the direction and the speed, the ship is a way-giving ship and needs collision avoidance operation, and the ship can carry out variable (reduced) speed avoidance, left steering avoidance and right steering avoidance at the moment, and mathematical logic codes are as follows:
compared with the prior art, the invention provides the intelligent ship navigation collision avoidance behavior coding method based on the machine language expression, and the method has the following beneficial effects:
the method avoids the limitation on the ship driving condition, adopts a coding method for the ship machine driving collision avoidance behavior, comprehensively considers the navigation rule by analyzing the ship machine driving behavior command and the state command, and formulates a corresponding comprehensive coding scheme; the method specifically comprises the following steps:
(1) The coding mode is efficient, simple and easy to realize;
(2) The intelligent ship navigation system can provide technical conditions for man-machine and machine-machine cooperative conversation during intelligent ship navigation, and provides important basic theory and technical support for intelligent development of shipping;
(3) The method adopts 01 codes and an ILP method to design the coding forms of the ship driving behavior indexes and the ship state indexes, preliminarily constructs a ship driving machine language system with universality and practical application, and is beneficial to providing the universal coding form for the ship driving machine: a universal and practical ship driving machine language system is explored and constructed, instruction requirements of driving behaviors and ship states are analyzed according to 3 ship navigation collision avoidance behaviors, and coding forms of ship driving behavior indexes and ship state indexes are innovatively designed by adopting a 01 code and ILP method.
Drawings
FIG. 1 is a schematic flow chart of a method for coding intelligent navigation collision avoidance behaviors of a ship based on machine language expression according to the present invention;
FIG. 2 is a schematic diagram of a two-ship overtaking scene of the intelligent ship navigation collision avoidance behavior coding method based on machine language expression provided by the invention;
FIG. 3 is a schematic diagram of a two-ship straight-channel encounter scene of the intelligent ship navigation collision avoidance behavior coding method based on machine language expression, provided by the invention;
fig. 4 is a schematic diagram of a scene of two-ship cross-meeting of the intelligent ship navigation collision avoidance behavior coding method based on machine language expression.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments.
Example 1:
referring to fig. 1-4, a method for encoding intelligent navigation collision avoidance behavior of a ship based on machine language expression includes the following steps (as shown in fig. 1):
s1, operation behavior coding: the data label processing is multipurpose, 1 and 0 indicate yes and no, and in the traditional label processing mode, if N identification results exist, the identification results are expressed by a1 multiplied by N01 array. However, considering the superposition of driving behaviors of vehicles and rudders, the operation types are exponentially increased, and if the operation types are still expressed by a 01 array, the operation types are excessively redundant, and the calculation efficiency is affected. For this, the present invention utilizes the idea of orthogonalization to superimpose 01 variables of each behavior together to form a1 × 7 01 array, and see table 1 for specific partitioning.
TABLE 1 partitioning of vessel driving behavior
| Left-handed rudder | To the right rudder | Running speed reduction | Increase of running speed | Back-up | Positive vehicle | |
| Left-handed rudder | ★ | × | √ | √ | √ | √ |
| To the right rudder | × | ★ | √ | √ | √ | √ |
| Running speed reduction | √ | √ | ★ | × | √ | √ |
| Increase of running speed | √ | √ | × | ★ | √ | √ |
| Back-up | √ | √ | √ | √ | ★ | × |
| Vehicle for correcting | √ | √ | √ | √ | × | ★ |
Note: : |, driving behavior;
x represents driving behavior that cannot be performed simultaneously;
and √ denotes a driving behavior that can be performed simultaneously.
S1.1, the encoding mode of the operational behavior attribute is shown in the table 2.
TABLE 2 operation behavior Attribute encoding
| Action 1 | Action 2 | Action 3 | Action 4 | Action 5 | Action 6 | Reservation action | |
| A | B | C | D | E | F | G | Action behavior attribute code (action identification code) |
Wherein, A represents a left rudder, B represents a right rudder, C represents acceleration, D represents deceleration, E represents reversing, and F represents driving. A. B, C, D, E, F, etc. are all 01 coded, with a 0 indicating that the action is not executed and a1 indicating execution. G is then the reserved attribute behavior. For which a corresponding action behavior attribute encoding table (see table 3) is available for grouping the entire operation behavior.
TABLE 3 action behavior Attribute encoding Table
| Quantization label | Behavior characteristicsSign for | Quantization label | Behavioral characteristics | Quantization label | Behavioral characteristics |
| [0000000] | Do not operate | [0100000] | Right-handed rudder | [1000000] | Left-hand rudder |
| [0000010] | Positive vehicle | [0100010] | Right-handed rudder and vehicle | [1000010] | Left-handed rudder and go-ahead vehicle |
| [0000100] | Back-up | [0100100] | Right-handed rudder and reverse | [1000100] | Left-handed rudder and reverse |
| [0001000] | Speed reduction | [0101000] | Right-handed rudder and speed reduction | [1001000] | Left-hand rudder and speed reduction |
| [0001010] | Vehicle correcting and decelerating | [0101010] | Right-handed rudder, forward driving and deceleration | [1001010] | Left-handed rudder, forward running and deceleration |
| [0001100] | Backing and decelerating | [0101100] | Right-handed rudder, backing car and reducing speed | [1001100] | Left-handed rudder, backing car and reducing speed |
| [0010000] | Acceleration | [0110000] | Right-handed rudder and acceleration | [1010000] | Left-handed rudder and acceleration |
| [0010010] | Driving and accelerating | [0110010] | Right rotary rudder, forward running and acceleration | [1010010] | Left-handed rudder, forward running and acceleration |
| [0010100] | Backing and accelerating | [0110100] | A right-handed rudder, a reversing car,Acceleration | [1010100] | Left-handed rudder, backing car and accelerating |
S1.2, in order to describe the operation degree of the whole ship so as to effectively refine the ship operation behavior, the invention provides operation behavior degree coding. Therefore, the captain with abundant driving experience is consulted, the unit of the action degree of the left-right rudder is selected as an angle, and the interval unit is 1 degree. The unit of the acceleration and deceleration action degree is selected as a section, and the unit of the interval is 0.1 section. The unit of the reversing action degree is selected to be second, and the unit of the interval is 5s. The unit of the action degree of the vehicle is selected as minutes, and the unit of the interval is 10min. The code is shown in table 4.
TABLE 4 encoding method of action behavior degree
| A | Action behavior attribute code (action identification code) |
| A1 | Action degree code (action degree 1) |
| A2 | Action degree code (action degree 2) |
| A3 | Action behavior degree code (action degree 3) |
| A4 | Action behavior degree code (action degree 4) |
| A5 | Action behavior degree code (action degree 5) |
| A6 | Action behavior degree code (action degree 6) |
| A7 | Action behavior degree code (action degree 7) |
Where A1, A2, A3, A4, A5, A6, etc. are all 01 coded, 0 means the action is not performed, and 1 means execution. It should be noted that there is mutual exclusion between these types of data, i.e. there is only one 1 code in A1, A2, A3, A4, A5, A6. For this, a corresponding table of encoding of the degree of operation behavior (see table 5) is available for grouping the entire degree of operation behavior.
TABLE 5 operation behavior level coding table
S1.3, the operation behavior attribute codes and the operation behavior degree codes are combined, and the invention provides a ship machine driving behavior coding mode suitable for a machine language expression mode, which is shown in a table 6.
TABLE 6 quantization of coding matrices for operational behaviors
| Action 1 | Action 2 | Action 3 | Action 4 | Action 5 | Action 6 | Reservation action | |
| A | B | C | D | E | F | G | Attribute coding (operation identification code) |
| A1 | B1 | C1 | D1 | E1 | F1 | G1 | Level code (operation level 1) |
| A2 | B2 | C2 | D2 | E2 | F2 | G2 | Level code (operation level 2) |
| A3 | B3 | C3 | D3 | E3 | F3 | G3 | Level code (operation level 3) |
| A4 | B4 | C4 | D4 | E4 | F4 | G4 | Level code (operation level 4) |
| A5 | B5 | C5 | D5 | E5 | F5 | G5 | Level code (operation level 5) |
| A6 | B6 | C6 | D6 | E6 | F6 | G6 | Level code (operation level 6) |
| A7 | B7 | C7 | D7 | E7 | F7 | G7 | Level code (operation level 7) |
The quantization matrix is used as an output template in the ship driving decision process, and the corresponding positions represent the action types and degrees respectively, so that strategy identification indication is provided for a decision link.
S2, ship state index coding: according to the ship state instruction requirement, the states of the ship and other ships need to be considered.
S2.1 and thus the code for the ship state may be represented as shown in table 7.
TABLE 7 Ship State information coding Format
Compared with ship driving behaviors, ship state information is more complex, so that degree coding cannot be simply coded by 01, real number coding is adopted, and ship state coding at each moment is achieved.
S2.2, in the coding process of ship attribute Logic, the ship is a straight ship or a way-giving ship, and different collision avoidance operation behaviors relate to COLREGS related background knowledge, so that the method adopts an Inductive Logic Programming (ILP) coding rule, the ILP is a cross field of machine learning and Logic program design, and can overcome the problems of limitation of background knowledge, limitation of a knowledge representation mechanism and the like in the traditional machine learning under the framework of first-order Logic by means of the existing theory and method of Logic program design, so that the machine better simulates human thinking, and can summarize a general rule capable of depicting special attributes of a case by enabling a computer to investigate specific cases. The form and structure of a rule is specified as:
here, "←" represents a logic inclusion symbol, the right part is called "rule body" and represents the premise of the rule, and the left part is "rule head" and represents the result of the rule. The rule body is composed of logic wordsf k The compound formula of the composition is shown in the specification,k=1，2，...，Lthe number of logical words in the rule body is called the length of the rule. [ ] of the rule header is also a logical word that represents the target category or concept determined by the rule. According to the rules, the attribute logic of the ship, namely the attributes of the straight ship or the yielding ship can be obtained.
S2.2.1, ship track-over scene rule and code
When the speed of the rear ship is higher than that of the front ship, a tracking situation is formed, and the third regulation of COLREGS is required to be followed in the tracking process: under the condition of good visibility, any ship can give way to the overtaken ship when overtaking any other ship. In the specified range of the overtaking path, under the 5-level sea condition, the test ship sails at the operation speed, the stern of the ship is within 3nm in the +/-67.5-degree direction, the speed of the other ship is higher than that of the test ship to accelerate, and the overtaking situation is formed; the test ship is positioned in the 3nm of the stern direction +/-67.5 degrees of the other ship, so that the other ship is tracked in an accelerated way, and a scene diagram is shown in figure 2.
According to the analysis, if the target ship overtakes the ship and the ship is a straight ship, keeping the direction and the speed; and if the ship overtakes the target ship and is a way-giving ship, steering collision avoidance control is carried out. Furthermore, the relative orientation of the two vessels is in the range of 112.5 °,247.5 °.
(1) Two power vessels, when the target vessel overtakes the ship, the ship is a straight ship, and the mathematical logic codes are as follows:
wherein X is the ship, Y is the target ship, v t Is the target ship speed, v 0 The speed of the ship, alpha is the angle of the ship relative to the ship, and beta is the azimuth of the ship relative to the ship. "Power-drive vessels (X)" and "power-drive vessels (Y)" represent that both vessels are motorized vessels. The speed of the target ship is smaller than that of the target ship because the target ship tracks over the target ship, and the speed ratio of the target ship to the local ship is smaller than 1. In addition, the angle of the target ship relative to the ship is [112.5 degrees ], 247.5 degrees °]Within the range.
(2) When the ship tracks over the target ship, the ship is a way-giving ship and needs to carry out collision avoidance control of left steering or right steering, and mathematical and logical codes are as follows:
in the rule logic code, description of collision avoidance control of the ship is added, and the 'give-way ship (X)', namely the ship is a way-giving ship, the target ship is a straight ship, and the ship is expressed by 'keep route (Y)', and direction and speed are guaranteed, so that the collision avoidance responsibility and collision avoidance control between the ships are considered at the same time.
S2.2.2, ship encounter scene rule and code
A encounter is one in which two powerships are in mutual sight, the first of the COLREGS' fourteenth encounters stipulates that when two powerships encounter a risk of collision in opposite or nearly opposite heading, each should turn right so that each drives across the port of the other. And the visibility is not lower than 1km under the 5-level sea state and below. The test ship can sail on the entrance channel correctly identified by the electronic chart and the light buoy, and the opposite coming ship meets the channel in the process. The channel is designed according to a double-line channel (relevant dimensions are calculated according to ship type standards and channel dimensions corresponding to the general design Specification for seaports (JTS 165-2013)), and the ship runs along the center line of the opposite channel in meeting.
Width of the multi-line channel:
W=QA+b+Qc
A=n（Lsin（θτ）+B）
in the formula: w is the navigation width (m) of the channel; q is the number of the channel lines; a is the width (m) of the flight path belt; c is the margin width (m) between the ship and the bottom line of the channel; b is the margin width (m) between the ships, and the designed ship width B is taken, so that the margin width between the ships can be properly increased when the intersection density of the ships is higher; n is the ship drift multiple; l is the design ship length (m); theta tau is wind and current pressure deflection angle (degree); and B is the designed beam (m). The corresponding scene description is shown in fig. 3.
From the above analysis, when two ships form an encounter local area, no matter the ship or the target ship is a way-giving ship, the two ships have the same avoidance responsibility and need to perform right-turning collision avoidance operation, and the corresponding logic codes are as follows:
s2.2.3, ship cross meeting scene rule and code
The cross meeting situation range is that two ships can only see the sidelight on one side of the other ship, the course lines of the two ships are crossed, the ship (the ship displaying the green sidelight) with other ship on the starboard of the ship is the way-giving ship, the other ship (the ship displaying the red sidelight) should give way, and the course and the navigation speed of the other ship, namely the straight ship, should be kept. A 2nm wide and 6nm channel is arranged on the sea area, and the electronic chart is marked; and under 5-level sea conditions and below, the operation speed of the ship is tested, and other ships sail within 3nm of the left side and the right side of the course of the ship. The two ships are crossed in the fore direction, the cross bulwark angle is larger than 5 degrees and smaller than 112.5 degrees, the ship speed of the other ship is +/-20 percent of the ship speed of the test ship, the two ships form a cross meeting situation, and the corresponding scene description is shown in fig. 4.
According to the COLREGS regulations, when a ship forms a cross meeting situation and has collision danger, a ship on a port side gives way to a ship on a starboard side to bear avoidance responsibility, and the collision avoidance can be divided into three situations:
(1) when the ship and the target ship form meeting and collision avoidance in which the port and the board meet in a cross way, the ship is a straight ship and is subjected to direction keeping and speed keeping; the target ship executes collision avoidance operation, and the ship needs collision avoidance operation only when the target ship does not fulfill the responsibility and obligation of the way-giving ship for collision avoidance operation or forms a tight office. The following regular logical coding model may be established:
(2) when the relative position or the bulwark angle of the coming ship is [5 degrees and 67.5 degrees ], the ship meets each other in a small-angle crossing way, the ship is a way-giving ship, the avoiding effect of right steering is better than speed change, and the ship needs to adopt collision avoidance control of right steering. The following rule logic coding model may be established:
(3) the starboard large-angle cross meeting means that when the bulwark angle of the target ship is 67.5 degrees and 112.5 degrees, the target ship is a straight ship and keeps the direction and the speed, the ship is a way-giving ship and needs to be controlled by collision avoidance, and the ship can carry out variable (reduced) speed avoidance, left steering avoidance and right steering avoidance. The following regular logical coding model may be established:
3) The comprehensive coding scheme for the ship machine driving comprises the following steps: the comprehensive driving behavior index code and the ship state index code are shown in the table 8.
TABLE 8 comprehensive code for marine machine driving
Compared with the table 7, the ship attribute code is added to the table, the code mode is 01 code, wherein 0 represents a straight ship, and 1 represents a way-giving ship, the former does not need to operate, and the latter needs to perform collision avoidance operation. Therefore, a machine language-oriented ship driving collision avoidance behavior coding scheme is obtained, and basic theoretical support is provided for ship collision avoidance decisions.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be considered as the technical solution of the present invention and the inventive concept thereof, which are equivalent to or changed by the present invention, within the technical scope of the present invention.
Claims (4)
1. A ship intelligent navigation collision avoidance behavior coding method based on machine language expression is characterized in that aiming at ship machine driving collision avoidance behaviors, under the instruction requirements of driving behaviors and ship states, driving behavior indexes and ship state indexes are coded through a machine language to form a set of machine language coding method with universality and practical application, and the method specifically comprises the following contents:
s1, dividing driving behavior instructions into a vehicle using and a rudder according to requirements, and specifically, dividing driving behavior instructions into a rudder rotating, a speed regulating and a reversingC b Peace vehicleC f The rotary rudder further comprises a left rotary rudderD l Right-handed rudderD r Non-rotating rudderD n The speed regulation further comprises accelerationV i And a decelerationV d Without speed-regulating actionV n (ii) a The vehicle-using and rudder-using commands can be independently or simultaneously carried out through a command setInstDividing the driving behavior instruction into the following specific sections:
Inst=｛｛D l ，D r ，D n ｝，｛V i ，V d ，V n ｝，｛C b ，C f ｝｝（1）
s2, in the collision avoidance process of intelligent sailing of the ship, the speed, the position coordinates and the course of the ship are obtained by means of an automatic ship identification system and a radar, the relative distance, the relative course and the relative speed between the ship and a target ship are calculated according to the speed, the position coordinates and the course, the closest meeting distance and the minimum meeting time between the ship and the target ship are calculated according to the relative distance, the relative course and the relative speed between the ship and the target ship, and a ship state instruction is divided by combining the data indexes;
and S3, aiming at 3 collision avoidance behavior indexes, namely the relative distance, the relative course and the relative speed between the ship and the target ship during the ship navigation, analyzing the command requirements of the driving behavior and the ship state, and designing the coding forms of the ship driving behavior index and the ship state index by adopting 01 codes and inducing logic program design coding rules.
2. The encoding method for intelligent vessel navigation collision avoidance behavior based on machine language expression of claim 1, wherein the vessel state command division mentioned in S2 specifically includes the following contents:
(1) the ship has the following motion parameters: the ship position { Bx, by }, the course Bag and the speed Bv;
(2) the motion parameters of the target ship are as follows: target vessel position { Tx, ty }, heading Tag, and speed Tv;
(3) ship safety state parameters: distance to last encounter (DCPA), time to minimum encounter (TCPA).
3. The intelligent ship sailing collision avoidance behavior coding method based on machine language expression of claim 1, wherein the ship driving behavior index coding form designed by using 01 codes mentioned in S3 specifically includes the following contents:
a1, operation behavior coding: the driving behavior commands are respectively coded, wherein A represents a left-handed rudderD l B represents a right-handed rudderD r And C represents accelerationV i And D represents decelerationV d E denotes reversingC b F represents a main vehicleC f G represents a reserved attribute behavior; a, B, C, D, E and F are all 01 codes, setting 0 represents that the behavior is not executed, and setting 1 represents execution;
a2, encoding the operation behavior degree: will turn left rudderD l Right-handed rudderD r The unit of action degree is selected as an angle, and the unit of interval is 1 degree; accelerationV i And a decelerationV d The unit of action degree is selected as a section, and the unit of interval is 0.1 section; back-upC b The unit of action degree is selected as second, and the unit of interval is 5s; positive vehicleC f Selecting action degree unit as minute, and interval unit as 10min; mutual exclusion exists among the operation behavior degree data, namely the single-class behavior degree data at any moment are unique;
and A3, combining the operation behavior attribute codes and the operation behavior degree codes in the A1 and the A2 to obtain a ship driving behavior index code.
4. The machine language expression-based intelligent ship navigation collision avoidance behavior coding method according to claim 1, wherein the designing of the ship state index coding form by using the inductive logic programming coding rule mentioned in S3 refers to the comprehensive consideration of the ship and target ship states and ship attribute logic, and the implementation of the ship state coding at each moment by using the inductive logic programming coding rule in combination with COLREGS-related regulations, and specifically includes the following contents:
b1, ship overtaking scene rules and codes: the two motor boats run in the same direction, when the speed of the rear boat is higher than that of the front boat, a overtaking situation is formed, and the overtaking situation follows the rule of COLREGS thirteenth, wherein the specific rule is as follows: in the specified range of the overtaking path, under the 5-level sea condition, the test ship sails at the operation speed, the stern of the ship is within 3nm in the +/-67.5-degree direction, the speed of the other ship is higher than that of the test ship to accelerate, and the overtaking situation is formed; the test ship is positioned in the 3nm of the stern direction +/-67.5 degrees of the other ship, and the other ship is accelerated to be tracked;
combining the analysis, if the target ship overtakes the ship and the ship is a straight ship, keeping the direction and speed; if the ship overtakes the target ship and is a way-giving ship, steering collision avoidance control is carried out, and the specific mathematical analysis is as follows:
b1.1, two motorized ships, when the target ship overtakes the ship, the ship is a straight ship, and mathematical and logical codes are as follows:
wherein X is the ship, Y is the target ship, V t Is the target ship speed, V 0 The speed of the ship, alpha is the angle of the target ship relative to the ship, and beta is the azimuth of the ship relative to the target ship; "stand-on ship (X)" indicates that the ship is a straight ship and keeps the direction and the speed; "Power-drive vessels (X)" and "power-drive vessels (Y)" indicate that both vessels are motorized vessels; "fusion Risk (X, Y)" indicates that avoidance responsibility and Collision avoidance operation between the own ship and the target ship are all taken into consideration;
b1.2, two motorized ships, when the ship tracks over the target ship, the ship is a way-yielding ship, collision avoidance operation of left steering or right steering is carried out, and mathematical logic codes are as follows:
wherein, the 'love-way ship (X) represents that the ship is a yielding ship, and the' keep court (Y) represents that the target ship is a straight ship;
b2, ship encounter scene rules and coding: when two motor-driven ships meet in opposite or nearly opposite directions to form collision danger, a meeting situation is formed, and the fourth regulation of COLREGS is followed, wherein the specific rule is as follows: under 5-level sea conditions and below, the visibility is not lower than 1km, a test ship can sail in an entrance channel correctly identified based on an electronic chart and a light buoy, and during the process, the test ship meets an oncoming ship in the opposite direction in the channel, and the channel is designed according to a double-line channel, and the meeting ship runs along the central line of the opposite channel;
the calculation rule of the width of the multi-line channel is as follows:
W=QA+b+Qc （4）
A=n（Lsin（θτ）+B）（5）
wherein, W represents the navigation width of the channel, m; q represents the number of lane lines; a represents the width of the track band, m; c represents the margin width between the ship and the bottom line of the channel, m; b represents the margin width between the ships, m, when the ship intersection density is large, the margin width between the ships can be increased properly; n represents the ship drift multiple; l represents the designed ship length, m; theta tau represents wind and flow pressure deflection angle; b represents the designed ship width, m;
combining the above analysis, when a meeting local area is formed between the ship and the target ship, the ship and the target ship both are yielding ships, both ships have the same yielding responsibility, and need to perform right-turn collision avoidance operation, and the mathematical logic codes are as follows:
b3, ship cross meeting scene rules and coding: the two ships can only see the sidelights on one side of the other side, the course lines of the two ships are crossed, the ship on the starboard of the ship is the way-giving ship, the way-giving ship is supposed to give way, and the course and the speed of the other ship, namely the straight ship, are supposed to be kept according to COLREGS regulations, and the specific rule is as follows: testing the operating speed of the ship under 5-level sea conditions and below, wherein other ships sail within a range of 3nm at any side of the left and right of the course; the bow directions of the two ships are crossed, the crossing angle is larger than 5 degrees and smaller than 112.5 degrees, the ship speed of the other ship is +/-20 percent of the navigation speed of the test ship, and the two ships form a crossing meeting situation;
in combination with the above analysis, when the ships form a cross meeting situation and there is a collision risk, the ship on the port side gives way to the ship on the starboard side to undertake avoidance responsibility, and the specific mathematical analysis is as follows:
b3.1, when the ship and the target ship form meeting collision avoidance in which port sides meet in a cross mode, the ship is a straight ship, and direction and speed are guaranteed; the target ship executes collision avoidance operation, and only when the target ship does not fulfill the responsibility and obligation of the yielding ship to carry out collision avoidance operation or form a tight office, the ship needs to carry out collision avoidance operation, and the mathematical logic codes are as follows:
b3.2, when the relative position or the bulwark angle of the coming ship is [5 degrees, 67.5 degrees ], the coming ship and the coming ship meet in a small-angle crossing way, at the moment, the ship is a yielding ship, the avoiding effect of right steering is better than speed change, the ship is required to adopt collision avoidance operation of right steering, and mathematical logic codes are as follows:
b3.3, starboard large-angle cross meeting means that when the bulwark angle of the target ship is [67.5, 112.5], the target ship is a straight ship and keeps the direction and the speed, the ship is a way-giving ship and needs collision avoidance operation, and the ship can carry out speed change or deceleration avoidance, left steering avoidance and right steering avoidance at the moment, and the mathematical logic code is as follows:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211185837.9A CN115268472B (en) | 2022-09-28 | 2022-09-28 | Intelligent ship navigation collision avoidance behavior coding method based on machine language expression |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211185837.9A CN115268472B (en) | 2022-09-28 | 2022-09-28 | Intelligent ship navigation collision avoidance behavior coding method based on machine language expression |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115268472A true CN115268472A (en) | 2022-11-01 |
| CN115268472B CN115268472B (en) | 2022-12-20 |
Family
ID=83756019
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211185837.9A Active CN115268472B (en) | 2022-09-28 | 2022-09-28 | Intelligent ship navigation collision avoidance behavior coding method based on machine language expression |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115268472B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB897101A (en) * | 1958-04-11 | 1962-05-23 | Avel Corp Geneva | Improvements in collision warning apparatus |
| GB201301488D0 (en) * | 2013-01-28 | 2013-03-13 | Kruschandl Nelson J | Autonomous Vessel Robot AI System |
| CN110083155A (en) * | 2019-04-19 | 2019-08-02 | 集美大学 | Machine learning method for realizing ship anthropomorphic intelligent collision avoidance decision |
| CN111186549A (en) * | 2020-01-15 | 2020-05-22 | 大连海事大学 | Course autopilot control system with ship collision avoidance function |
| CN113625728A (en) * | 2021-09-23 | 2021-11-09 | 大连海事大学 | Performance-limited ship collision avoidance path planning method |
| CN113703463A (en) * | 2021-09-23 | 2021-11-26 | 大连海事大学 | Elite multi-population evolution algorithm-based ship collision avoidance path planning method |
| WO2021249645A1 (en) * | 2020-06-11 | 2021-12-16 | Abb Oy | Apparatus, method and computer program for controlling propulsion of marine vessel |
| CN114550499A (en) * | 2022-02-14 | 2022-05-27 | 武汉理工大学 | Intelligent ship collision avoidance method and device integrating COLREGs scene knowledge |
| CN114896819A (en) * | 2022-06-13 | 2022-08-12 | 北京航空航天大学 | Planning method for multi-search and rescue equipment cooperative search and rescue task in medium and far sea area |
-
2022
- 2022-09-28 CN CN202211185837.9A patent/CN115268472B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB897101A (en) * | 1958-04-11 | 1962-05-23 | Avel Corp Geneva | Improvements in collision warning apparatus |
| GB201301488D0 (en) * | 2013-01-28 | 2013-03-13 | Kruschandl Nelson J | Autonomous Vessel Robot AI System |
| CN110083155A (en) * | 2019-04-19 | 2019-08-02 | 集美大学 | Machine learning method for realizing ship anthropomorphic intelligent collision avoidance decision |
| CN111186549A (en) * | 2020-01-15 | 2020-05-22 | 大连海事大学 | Course autopilot control system with ship collision avoidance function |
| WO2021249645A1 (en) * | 2020-06-11 | 2021-12-16 | Abb Oy | Apparatus, method and computer program for controlling propulsion of marine vessel |
| CN113625728A (en) * | 2021-09-23 | 2021-11-09 | 大连海事大学 | Performance-limited ship collision avoidance path planning method |
| CN113703463A (en) * | 2021-09-23 | 2021-11-26 | 大连海事大学 | Elite multi-population evolution algorithm-based ship collision avoidance path planning method |
| CN114550499A (en) * | 2022-02-14 | 2022-05-27 | 武汉理工大学 | Intelligent ship collision avoidance method and device integrating COLREGs scene knowledge |
| CN114896819A (en) * | 2022-06-13 | 2022-08-12 | 北京航空航天大学 | Planning method for multi-search and rescue equipment cooperative search and rescue task in medium and far sea area |
Non-Patent Citations (6)
| Title |
|---|
| JIANAN LUO 等: "《Study on the Risk Model of the Intelligent Ship Navigation》", 《WILEY》 * |
| RICARDO DOS SANTOS FERREIRA 等: "《Comparative analysis of machine learning prediction models of container ships propulsion power》", 《OCEAN ENGINEERING》 * |
| XIONGFEI GENG 等: "《Motion Plan of Maritime Autonomous Surface Ships by Dynamic Programming for Collision Avoidance and Speed Optimization》", 《SENSORS》 * |
| 倪生科: "《基于规则的船舶智能避碰决策关键技术研究》", 《中国博士学位论文全文数据库工程科技Ⅰ辑》 * |
| 周壮壮 等: "《基于四元船舶领域的船舶碰撞危险度模型》", 《武汉理工大学学报(交通科学与工程版)》 * |
| 桂士宏 等: "《国外军用领域人工智能发展规划及舰船智能化技术运用》", 《舰 船 科 学 技 术》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115268472B (en) | 2022-12-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Porathe | Maritime Autonomous Surface Ships (MASS) and the COLREGS: Do we need quantified rules or is “the ordinary practice of seamen” specific enough? | |
| CN109597417B (en) | Multi-USV group collaborative collision avoidance planning method based on collision avoidance criterion | |
| Zaccone et al. | A colreg-compliant ship collision avoidance algorithm | |
| CN108459602B (en) | Autonomous berthing method of under-actuated unmanned ship in multi-obstacle complex environment | |
| CN113096446B (en) | Multi-ship collision avoidance decision-making method under hybrid navigation scene, storage medium and processor | |
| CN110444046B (en) | Method for analyzing cluster situation of non-conflict meeting ship in limited water area | |
| Porathe | Safety of autonomous shipping: COLREGS and interaction between manned and unmanned ships | |
| CN112965475A (en) | Obstacle collision prevention method based on dynamic navigation ship domain and collision prevention rule | |
| CN114895673A (en) | Ship collision avoidance decision method based on deep reinforcement learning under rule constraint | |
| Wang et al. | A novel maritime autonomous navigation decision-making system: Modeling, integration, and real ship trial | |
| CN116360434A (en) | Ship path planning method based on improved CSAC-APF algorithm | |
| CN114550499B (en) | Ship intelligent collision prevention method and device integrating COLREGs scene knowledge | |
| CN111679674B (en) | Flexible meeting evasion method for unmanned ship | |
| CN115268472B (en) | Intelligent ship navigation collision avoidance behavior coding method based on machine language expression | |
| Jing et al. | Path planning and navigation of oceanic autonomous sailboats and vessels: A survey | |
| Zhao et al. | Decision-making for the autonomous navigation of USVs based on deep reinforcement learning under IALA maritime buoyage system | |
| CN117008606A (en) | Autonomous navigation decision-making method and device for ship in water area with line system | |
| CN110796142A (en) | Method for acquiring relative field between ships based on ship cluster situation | |
| Eriksen et al. | Short-term ASV collision avoidance with static and moving obstacles | |
| CN110928278A (en) | Autonomous safe sailing method for unmanned ship and unmanned ship | |
| CN114035574A (en) | Autonomous obstacle avoidance method for unmanned surface vehicle | |
| Zhang et al. | Collision avoidance path planning of open water ships based on velocity obstacle method | |
| CN110796899A (en) | Ship-shore relative field acquisition method based on ship cluster situation in limited water area | |
| Wang et al. | COLREGs-Compliant Collision Avoidance Method for Autonomous Ships via Deep Reinforcement Learning | |
| CN114912658A (en) | Computer integrated control method for full-seas search and rescue instruction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |