CN111721321B - Intelligent collision avoidance method and system for navigation mark - Google Patents

Intelligent collision avoidance method and system for navigation mark Download PDF

Info

Publication number
CN111721321B
CN111721321B CN202010427248.1A CN202010427248A CN111721321B CN 111721321 B CN111721321 B CN 111721321B CN 202010427248 A CN202010427248 A CN 202010427248A CN 111721321 B CN111721321 B CN 111721321B
Authority
CN
China
Prior art keywords
ship
time
navigation mark
track
collision
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.)
Active
Application number
CN202010427248.1A
Other languages
Chinese (zh)
Other versions
CN111721321A (en
Inventor
李锋
甄涛
江道伟
陈国伟
谢奎
邵哲平
洪长华
薛晗
孙洪波
洪炜煜
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Aids To Navigation Department Of Donghai Navigation Safety Administration Mot
Jimei University
Original Assignee
Shanghai Aids To Navigation Department Of Donghai Navigation Safety Administration Mot
Jimei University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shanghai Aids To Navigation Department Of Donghai Navigation Safety Administration Mot, Jimei University filed Critical Shanghai Aids To Navigation Department Of Donghai Navigation Safety Administration Mot
Priority to CN202010427248.1A priority Critical patent/CN111721321B/en
Publication of CN111721321A publication Critical patent/CN111721321A/en
Application granted granted Critical
Publication of CN111721321B publication Critical patent/CN111721321B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C25/00—Manufacturing, calibrating, cleaning, or repairing instruments or devices referred to in the other groups of this subclass
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/20—Instruments for performing navigational calculations
    • G01C21/22—Plotting boards

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Manufacturing & Machinery (AREA)
  • Automation & Control Theory (AREA)
  • Traffic Control Systems (AREA)
  • Navigation (AREA)

Abstract

The invention discloses an intelligent navigation mark collision avoidance method, which realizes intelligent prediction and early warning of ship-navigation mark collision risks by predicting ship tracks, constructing a 3D scene and estimating collision risks; predicting the ship track, namely predicting the ship position, the heading and the track of the ship characteristic points according to the ship characteristic point model and the real-time and historical ship position and heading of the ship, wherein the ship track is predicted according to a Kalman filtering method; constructing a 3D scene, and performing three-dimensional visual display on the cloth of the navigation mark, wherein the three-dimensional visual display comprises a schematic diagram of the cloth of the navigation mark; and (4) estimating collision risks, namely calculating the nearest meeting distance between the navigation mark and the ship characteristic point model and the driving distance required by the ship to reach the meeting point, estimating collision risks in real time, and prejudging whether the ship and the navigation mark collide. The invention also discloses a device based on the method. The invention has the advantages that: the intelligent prediction and early warning of the collision risk of the ship and the navigation mark can be effectively realized.

Description

Intelligent collision avoidance method and system for navigation mark
Technical Field
The invention relates to the technical field of ship track prediction, in particular to an intelligent navigation mark collision avoidance method and system.
Background
In the prior art, various intelligent prediction algorithms are adopted to predict the navigation track of a ship in the field of navigation prediction and control of the ship, so that the navigation safety of the ship is ensured. In the prior art, a method for predicting a ship navigation track adopts a Kalman filter for calculation. However, the conventional kalman filter uses the same sampling interval throughout the filtering process. However, as AIS location update reporting intervals vary with ship speeds, communication between the coastal station and the buoy is blocked and network delays are incurred, and sometimes correct location data may not be received or the location reports are not updated in a timely manner. In the prior art, a Kalman filter is introduced to predict the ship track by adopting a self-adaptive mechanism, but the problems of untimely updating of position data or large estimation error still exist.
In addition, in the prior art, especially in the field of navigation prediction of navigation marks and ships, a method for visualizing the navigation marks and intelligently predicting and early warning the collision risk of the ships and the navigation marks is lacked.
Disclosure of Invention
The invention aims to solve the technical problem of how to realize intelligent prediction and early warning of the collision risk of a ship and a navigation mark, and provides an intelligent navigation mark collision avoidance method and system aiming at the technical problem to be solved.
In order to achieve the purpose, the invention provides the following technical scheme: an intelligent navigation mark collision avoidance method realizes intelligent prediction and early warning of ship-navigation mark collision risks by predicting ship tracks, constructing a 3D scene and estimating collision risks;
predicting a ship track, namely predicting the ship position, heading and track of the ship characteristic points according to the ship characteristic point model and the real-time and historical ship position and heading of the ship, wherein the ship track is predicted according to a Kalman filtering method, and different time scales are introduced into the Kalman filtering method;
constructing a 3D scene, and performing three-dimensional visual display on the cloth of the navigation mark, wherein the three-dimensional visual display comprises a schematic diagram of the cloth of the navigation mark;
and (4) estimating collision risks, namely calculating the nearest meeting distance between the navigation mark and the ship characteristic point model and the driving distance required by the ship to reach the meeting point, estimating collision risks in real time, and prejudging whether the ship and the navigation mark collide.
Further, the mathematical model of the ship motion system adopted in the process of predicting the ship track is as follows:
Figure GDA0003352485920000021
Figure GDA0003352485920000022
Figure GDA0003352485920000023
Figure GDA0003352485920000024
where x is a state variable, u (k) is an input vector at time k, wkSystem state noise at time k, k is the time sequence count, T (k) is the kth time interval of the AIS location update report, zkIs the measured output of time k,. epsilonkFor the noise of the measurement at time k,
Figure GDA0003352485920000025
denotes the error at time K +1, Kk+1Denotes the gain of the Kalman filtering at time k +1, x (k) denotes the state variable at time k,
Figure GDA0003352485920000026
the average value of the time interval is represented,
Figure GDA0003352485920000027
indicating movement at time kMechanical measurements, v (k +1) representing the velocity of the vessel at time k +1,
Figure GDA0003352485920000028
covariance, P, representing the estimation error of k +1 timekDenotes the covariance of the error at time k, E denotes the mean, QkRepresenting the covariance of the system noise at time k.
Further, the covariance of the system noise at time k is calculated according to an adaptive mechanism for selecting the state noise covariance, and the calculation model is as follows:
Figure GDA0003352485920000029
Figure GDA00033524859200000210
wherein beta is a forgetting factor, and the value range of beta is 0.95<β<0.99，λkIs the gain factor at time k.
Further, the longitude and latitude of the ship at the instant of the k +1th time interval in the process of predicting the ship track can be calculated as follows:
Figure GDA00033524859200000211
λ(k+1)＝λ(k)+v(k)T(k)sin(θ(k))
the system state equation is:
x(k+1)＝x(k)+T(k)u(k)+w(k)
the system observations are:
z(k+1)＝x(k+1)+ε(k+1)
where x is a state variable, u (k) is an input vector at time k, wkSystem state noise at time k, k is the time sequence count, T (k) is the kth time interval of the AIS location update report, zkIs the measured output of time k,. epsilonkIs the measurement noise at time k.
Further, the collision risk estimation process includes the steps of:
s01, calculating the nearest meeting between the navigation mark and the ship characteristic point modelDistance and driving distance required by ship to reach meeting point, wherein the ship track is set as Q1Q2Observation line is P1P2；
s02, for quick repulsion test, set line segment P1P2The diagonal rectangle is R, and a line segment Q is set1Q2The diagonal rectangle is T, if R and T are not intersected, the two segments cannot be intersected, and the two tracks cannot be collided;
s03, performing a straddle test, wherein if two line segments intersect, the two line segments inevitably straddle each other, and if P is1P2Straddle Q1Q2Then vector P1And P2Two points are located at Q2And Q1The two sides of the straight line where the two tracks are located can not collide if the two tracks do not pass the straddle test, and the two tracks can not collide if the two tracks pass the straddle test.
The invention also aims to provide an intelligent collision avoidance system for a navigation mark, which comprises a ship track prediction device, a 3D scene display device and a collision risk estimation device, wherein the ship track prediction device predicts the ship track according to the method, the 3D scene display device performs three-dimensional visual display on the arrangement of the navigation mark according to the method, and the collision risk estimation device estimates the collision risk of the ship and the navigation mark according to the method.
Further, the 3D scene display device is a mobile terminal provided with a 3D scene display system based on a 3D max modeling tool and UNITY 3D publishing, and the mobile terminal includes a mobile terminal based on IOS, ANDROID, PC, WEB, PS3 and XBOX.
Compared with the prior art, the invention has the beneficial effects that:
according to the invention, different time scales are introduced into Kalman filtering, a self-adaptive variable parameter time scale Kalman filter is designed, and buoy drift of different report intervals is estimated by adopting APVTS-KF. Therefore, the Kalman filtering can more accurately solve the problem of drift estimation under the actual condition of the buoy dynamic interval position report. Before reaching steady state, a larger state noise covariance is chosen to speed up convergence speed. After reaching steady state, convergence accuracy can be improved by selecting a smaller state noise covariance. Under different working states, the noise covariance of the Kalman filter is subjected to self-adaptive updating and matching, rapid convergence and high estimation accuracy can be obtained under various working environments, and therefore trajectory prediction of AIS position sampling release updating period along with uncertain changes of ship speed dynamics can be processed. In addition, the navigation mark is displayed in a three-dimensional view mode, so that navigation mark configuration can be comprehensively informed to navigation mark users, ship navigation is greatly facilitated, collision risk intelligent prediction and early warning are carried out through the terminal, collision can be effectively avoided, and ship danger avoidance is informed in advance.
Drawings
FIG. 1 is a schematic flow chart of the method of the present invention.
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. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, the present embodiment discloses an intelligent collision avoidance method for a navigation mark, which realizes intelligent prediction and early warning of a collision risk between a ship and a navigation mark by predicting a ship track, constructing a 3D scene, and estimating the collision risk;
the method comprises the steps of predicting ship tracks, predicting positions, heading and tracks of ship feature points according to ship feature point models and ship real-time and historical positions and heading of ships, wherein the ship tracks are predicted according to a Kalman filtering method, and different time scales are introduced into the Kalman filtering method.
The filtering method of the kalman filter adopted in this embodiment may be as follows, where the mathematical model of the ship motion system adopted in the process of predicting the ship track is:
Figure GDA0003352485920000041
Figure GDA0003352485920000042
Figure GDA0003352485920000043
Figure GDA0003352485920000044
where x is a state variable, u (k) is an input vector at time k, wkSystem state noise at time k, k is the time sequence count, T (k) is the kth time interval of the AIS location update report, zkIs the measured output of time k,. epsilonkFor the noise of the measurement at time k,
Figure GDA0003352485920000045
denotes the error at time K +1, Kk+1Denotes the gain of the Kalman filtering at time k +1, x (k) denotes the state variable at time k,
Figure GDA0003352485920000046
the average value of the time interval is represented,
Figure GDA0003352485920000047
represents a measurement of the dynamics at the moment k, v (k +1) represents the speed of the vessel at the moment k +1,
Figure GDA0003352485920000051
covariance, P, representing the estimation error of k +1 timekDenotes the covariance of the error at time k, E denotes the mean, QkRepresenting the covariance of the system noise at time k.
System noise inThe covariance at the moment k is calculated according to the adaptive mechanism of the covariance of the selected state noise, and the calculation model is
Figure GDA0003352485920000052
Figure GDA0003352485920000053
Wherein beta is a forgetting factor, and the value range of beta is 0.95<β<0.99，λkIs the gain factor at time k. Before reaching steady state, lambdakThis results in a larger covariance of the state noise, which speeds up convergence. After reaching steady state, λkIs smaller, which results in a smaller state noise covariance to improve convergence accuracy. Adaptive strategy influences Q in mathematical model of ship motion system by adjustingkTo adjust the covariance of the estimation error
Figure GDA0003352485920000054
The setting and proving process of the mathematical model of the ship motion system adopted in the process of measuring the ship track is as follows:
in this embodiment, the motion model of the ship is expressed as follows:
t (k) is defined as the kth time interval of the AIS location update report.
Figure GDA0003352485920000055
Indicating the latitude of the vessel. λ (k) represents the longitude of the ship. v (k) represents the speed of the ship. θ (k) represents the heading of the vessel. The latitude and longitude of the buoy at the instant of time k +1th can be calculated as:
Figure GDA0003352485920000056
λ(k+1)＝λ(k)+v(k)T(k)sin(θ(k)) (2)
define x as a state variable:
Figure GDA0003352485920000057
define u (k) as the input vector:
Figure GDA0003352485920000058
consider whether noise is present in an actual system. Defined as the system noise. The system state equations of equations (1) and (2) can be written as:
x(k+1)＝x(k)+T(k)u(k)+w(k) (5)
define z as the measurement vector. Defined as measurement noise. The system observations can be written as:
z(k+1)＝x(k+1)+ε(k+1) (6)
the ship track prediction definition and the certification based on the adaptive variable parameter time scale Kalman filter are as follows:
definition of
Figure GDA0003352485920000061
State estimation of x (k), i.e.
Figure GDA0003352485920000062
Definition of
Figure GDA0003352485920000063
Is the average value of T (k), i.e.
Figure GDA0003352485920000064
The Kalman filtering is realized by two steps of time updating and measurement updating. The time update step predicts an estimate of the state, the time going one step forward.
Figure GDA0003352485920000065
Expressed as a posteriori estimate of the state at time k, calculated as follows
Figure GDA0003352485920000066
The measurement update step determines a posteriori estimates of time k
Figure GDA0003352485920000067
Measurement of state dynamics
Figure GDA0003352485920000068
Can be updated as follows
Figure GDA0003352485920000069
Where K is the kalman filter gain. Error of the k-th time
Figure GDA00033524859200000610
Given by:
Figure GDA00033524859200000611
it is possible to set the ratio of hypothesis 1: e [ w ]k]0. system state noise wkAnd the measurement output zkIs irrelevant. QkRepresenting the covariance of the system noise at time k, is calculated as follows:
Figure GDA00033524859200000612
assume 2: e [ epsilon ]k]0. measurement noise epsilonkAnd xkAnd not related.
Figure GDA00033524859200000613
Representing the covariance of the measurement noise at time k + 1.
Figure GDA00033524859200000614
Preferably, in the present embodiment, for the system state equation (5) containing system state noise and the system observation equation (6) containing measurement noise, the adaptive variable parameter time scale kalman filtering method is given by equations (14) to (17).
Figure GDA00033524859200000615
Figure GDA00033524859200000616
Figure GDA00033524859200000617
Figure GDA0003352485920000071
In the formula
Figure GDA0003352485920000072
Covariance of the estimation error at time k. PkIs the covariance of the error at time k.
The proof procedure in this particular example is possible as follows:
substituting (5) into (11) can obtain:
Figure GDA0003352485920000073
substituting (10) into (18) yields:
Figure GDA0003352485920000074
substituting (6) into (19) yields:
Figure GDA0003352485920000075
substituting (5) into (20) yields:
Figure GDA0003352485920000076
substituting (9) into (21) yields:
Figure GDA0003352485920000077
the formula (14) is proved.
Figure GDA0003352485920000078
The calculation is as follows:
Figure GDA0003352485920000079
substituting (5) into (23) yields:
Figure GDA00033524859200000710
substituting (9) into (24) yields:
Figure GDA00033524859200000711
from hypothesis 1 we can derive:
Figure GDA0003352485920000081
the pattern of formula (15) is shown.
Figure GDA0003352485920000082
Substituting (10) into (27) yields:
Figure GDA0003352485920000083
substituting (6) into (27) yields:
Figure GDA0003352485920000084
from hypothesis 2 we can derive:
Figure GDA0003352485920000085
substituting (23) into (30) yields:
Figure GDA0003352485920000086
equation (16) is obtained. To (31) regarding Kk+1The derivation can be:
Figure GDA0003352485920000087
namely:
Figure GDA0003352485920000088
the following therefore holds:
Figure GDA0003352485920000089
the formula (17) is shown.
Substituting (34) into (31) yields:
Figure GDA0003352485920000091
when the state noise covariance is small, the convergence speed is low and the convergence accuracy is high. If the state motion changes too fast in a short time, the slow convergence speed easily leads to tracking failure. When the state noise covariance is large, the convergence speed is high and the precision is low. Thus, the noise covariance in the kalman filter can be adaptively updated and matched under different operating conditions. The self-adaptive Kalman filter can simultaneously obtain quick convergence and high estimation precision under a large-range working environment. In this embodiment, an adaptive mechanism for selecting the covariance of the state noise is adopted, which specifically includes the following steps:
Figure GDA0003352485920000092
Figure GDA0003352485920000093
wherein 0.95<β<0.99 is a forgetting factor, λkIs the gain factor at time k. Before reaching steady state, lambdakThis results in a larger covariance of the state noise, which speeds up convergence. After reaching steady state, λkIs smaller, which results in a smaller state noise covariance to improve convergence accuracy. The adaptive strategy affects Q in equation (15) by adjustingkTo adjust the covariance of the estimation error
Figure GDA0003352485920000094
In addition, the method also comprises the steps of carrying out three-dimensional visual display on the cloth of the navigation mark by constructing a 3D scene, wherein the three-dimensional visual display comprises a schematic diagram of the cloth of the navigation mark; preferably, the three-dimensional visual display can be performed through a 3D scene display device, the 3D scene display device is a mobile terminal provided with a 3D scene display system based on a 3D max modeling tool and a UNITY 3D release, and the mobile terminal includes a mobile terminal based on IOS, ANDROID, PC, WEB, PS3 and XBOX.
The system introduces the arrangement of the navigation mark in a three-dimensional view mode, and informs the navigation mark user of the concise and comprehensive arrangement intention of the navigation mark, thereby greatly facilitating the navigation of the ship. The arrangement of the navigation mark generally has three important positions: (1) harbor doors, such as a township corner light tower, are located in the south of the mansion door island at about 9 nautical miles; (2) a turning point, which requires an open water area for turning because the turning radius of the ship is large; (3) dangerous locations such as reef reefs, shoals, work areas, obstructions, bridges, docks, etc.
The system can complete multiple navigation aid functions of marking obstructive objects or dangerous objects, indicating the boundary of a navigation channel and the like, remind the ship to drive cautiously, help the ship to distinguish the boundary of the navigation channel, inform steering or obstructive objects, guide the ship to avoid shoals and reefs, distinguish branches and confluence of navigation sections, correctly navigate, position and steer and safely pass through the navigation channel. The distance and the direction of the key navigation mark are reminded, and the distance and the direction between the position of the ship and the key navigation mark are displayed in real time in the process of driving the ship, so that a user can conveniently determine the course in time. When the ship approaches a certain navigation mark, the system will prompt the user to flash by light and inform the user of the information transmitted by the navigation mark in a form of characters and graphic representation.
The platform of the UNITY 3D game engine is applied using a 3D Max modeling tool. The Unity 3D supports rapid release of a plurality of platforms including IOS, ANDROID, PC, WEB, PS3, XBOX and the like, and provides a wide development space for popularization of a three-dimensional system to the plurality of platforms.
And (4) estimating collision risks, namely calculating the nearest meeting distance between the navigation mark and the ship characteristic point model and the driving distance required by the ship to reach the meeting point, estimating collision risks in real time, and prejudging whether the ship and the navigation mark collide.
Preferably, the collision risk estimation process includes the steps of:
s01, calculating the nearest meeting distance between the navigation mark and the ship characteristic point model and the driving distance required by the ship to reach the meeting point, wherein the ship track is set as Q1Q2Observation line is P1P2；
s02, for quick repulsion test, set line segment P1P2The diagonal rectangle is R, and a line segment Q is set1Q2Is a pair ofThe rectangle of the angular line is T, if R and T are not intersected, the two line segments are not intersected, and the two tracks are judged not to be collided;
s03, performing a straddle test, wherein if two line segments intersect, the two line segments inevitably straddle each other, and if P is1P2Straddle Q1Q2Then vector P1And P2Two points are located at Q2And Q1The two sides of the straight line where the two tracks are located can not collide if the two tracks do not pass the straddle test, and the two tracks can not collide if the two tracks pass the straddle test.
Setting X to represent cross multiplication and X to represent dot multiplication, and judging Q1Q2Straddle P1P2The basis of (A) is as follows:
(Q1P1×Q1Q2)*(Q1Q2×Q1P2)>0 (38)
(P1Q1×P1P2)*(P1P2×P1Q2)>0 (39)
according to the method, the ship motion track is predicted by adopting a Kalman filtering technology, the accuracy and the robustness of a Kalman filtering algorithm are improved, the Kalman filtering is popularized from an integer order system to a fractional order system, and a Kalman filtering algorithm based on a self-adaptive fractional order system is deduced. The adaptive mechanism of the state noise covariance selection deduces the mathematical process of the adaptive mechanism. Compared with an integer order filter, the fractional order Kalman filter has a wider parameter selection range and higher tracking precision. Based on the adaptive fractional Kalman filtering, the fractional Kalman filtering has smaller central position error and better tracking precision than the fractional Kalman filtering and the Kalman filtering. The adaptive fractional order Kalman can better avoid the drift effect of the tracker, and has good robustness and effectiveness.
The method can be arranged in a navigation mark intelligent collision avoidance system, the system can be a shipborne mobile terminal and other equipment, has a communication function, communicates information with the AIS in real time, and can comprise a ship track prediction device, a 3D scene display device and a collision risk estimation device, wherein the ship track prediction device predicts a ship track, the 3D scene display device displays the cloth of the navigation mark in a three-dimensional view, and the collision risk estimation device estimates the collision risk of the ship and the navigation mark. In addition, the results of system prediction, display and estimation can be displayed through a display device, and the results can be displayed on a shipborne display terminal, such as a terminal display screen, so that the results can be conveniently viewed by a user.
According to the invention, different time scales are introduced into Kalman filtering, a self-adaptive variable parameter time scale Kalman filter is designed, and buoy drift of different report intervals is estimated by adopting APVTS-KF. Therefore, the Kalman filtering can more accurately solve the problem of drift estimation under the actual condition of the buoy dynamic interval position report. Before reaching steady state, a larger state noise covariance is chosen to speed up convergence speed. After reaching steady state, convergence accuracy can be improved by selecting a smaller state noise covariance. Under different working states, the noise covariance of the Kalman filter is adaptively updated and matched, and rapid convergence and high estimation accuracy can be simultaneously obtained under various working environments. In addition, the navigation mark is displayed in a three-dimensional view mode, so that navigation mark configuration can be comprehensively informed to navigation mark users, ship navigation is greatly facilitated, collision risk intelligent prediction and early warning are carried out through the terminal, collision can be effectively avoided, and ship danger avoidance is informed in advance.
The embodiments of the present invention have been described in detail with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. It will be apparent to those skilled in the art that various changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, and the scope of protection is still within the scope of the invention.

Claims (6)

1. An intelligent navigation mark collision avoidance method is characterized in that intelligent prediction and early warning of ship-navigation mark collision risks are realized by predicting ship tracks, constructing a 3D scene and estimating collision risks;
predicting a ship track, namely predicting the ship position, heading and track of the ship characteristic points according to the ship characteristic point model and the real-time and historical ship position and heading of the ship, wherein the ship track is predicted according to a Kalman filtering method, and different time scales are introduced into the Kalman filtering method;
constructing a 3D scene, and performing three-dimensional visual display on the cloth of the navigation mark, wherein the three-dimensional visual display comprises a schematic diagram of the cloth of the navigation mark;
estimating collision risk, namely calculating the nearest meeting distance between the navigation mark and the ship characteristic point model and the driving distance required by a ship reaching the meeting point, estimating collision risk in real time, and prejudging whether the ship and the navigation mark collide;
the mathematical model of the ship motion system adopted in the process of predicting the ship track is as follows:
Figure FDA0003352485910000011
Figure FDA0003352485910000012
Figure FDA0003352485910000013
Figure FDA0003352485910000014
where x is a state variable, u (k) is an input vector at time k, wkSystem state noise at time k, k is the time sequence count, T (k) is the kth time interval of the AIS location update report, zkIs the measured output of time k,. epsilonkFor the noise of the measurement at time k,
Figure FDA0003352485910000015
denotes the error at time K +1, Kk+1Denotes the gain of the Kalman filtering at time k +1, x (k) denotes the state variable at time k,
Figure FDA0003352485910000016
the average value of the time interval is represented,
Figure FDA0003352485910000017
represents a measurement of the dynamics at the moment k, v (k +1) represents the speed of the vessel at the moment k +1,
Figure FDA0003352485910000018
covariance, P, representing the estimation error of k +1 timekDenotes the covariance of the error at time k, E denotes the mean, QkRepresenting the covariance of the system noise at time k.
2. The intelligent beacon collision avoidance method of claim 1, wherein the covariance of the system noise at time k is calculated according to an adaptive mechanism that selects the state noise covariance, and the calculation model is as follows:
Figure FDA0003352485910000019
wherein beta is a forgetting factor, and the value range of beta is 0.95<β<0.99，λkIs the gain factor at time k.
3. The intelligent navigation mark collision avoidance method according to claim 2, wherein the longitude and latitude of the ship at the instant of the k +1th time interval in the process of predicting the ship track can be calculated as:
Figure FDA0003352485910000021
λ(k+1)＝λ(k)+v(k)T(k)sin(θ(k))
the system state equation is:
x(k+1)＝x(k)+T(k)u(k)+w(k)
the system observations are:
z(k+1)＝x(k+1)+ε(k+1)
where x is a state variable, u (k) is an input vector at time k, wkSystem state noise at time k, k is the time sequence count, T (k) is the kth time interval of the AIS location update report, zkIs the measured output of time k,. epsilonkIs the measurement noise at time k.
4. The intelligent navigation mark collision avoidance method according to claim 1, wherein the collision risk estimation process comprises the following steps:
s01, calculating the nearest meeting distance between the navigation mark and the ship characteristic point model and the driving distance required by the ship to reach the meeting point, wherein the ship track is set as Q1Q2Observation line is P1P2；
s02, for quick repulsion test, set line segment P1P2The diagonal rectangle is R, and a line segment Q is set1Q2The diagonal rectangle is T, if R and T are not intersected, the two segments cannot be intersected, and the two tracks cannot be collided;
s03, performing a straddle test, wherein if two line segments intersect, the two line segments inevitably straddle each other, and if P is1P2Straddle Q1Q2Then vector P1And P2Two points are located at Q2And Q1The two sides of the straight line where the two tracks are located can not collide if the two tracks do not pass the straddle test, and the two tracks can not collide if the two tracks pass the straddle test.
5. An intelligent navigation mark collision avoidance system, which is characterized by comprising a ship track prediction device, a 3D scene display device and a collision risk estimation device, wherein the ship track prediction device predicts a ship track according to the method of any one of claims 1 to 3, the 3D scene display device displays the arrangement of navigation marks in a three-dimensional view according to the method of claim 1, and the collision risk estimation device estimates the collision risk of a ship and the navigation marks according to the method of claim 1 or 4.
6. The intelligent navigation mark collision avoidance system of claim 5, wherein the 3D scene display device is a mobile terminal equipped with a 3D scene display system based on a 3D Max modeling tool and UNITY 3D publishing, and the mobile terminal comprises a mobile terminal based on IOS, ANDROID, PC, WEB, PS3 and XBOX.
CN202010427248.1A 2020-05-19 2020-05-19 Intelligent collision avoidance method and system for navigation mark Active CN111721321B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010427248.1A CN111721321B (en) 2020-05-19 2020-05-19 Intelligent collision avoidance method and system for navigation mark

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010427248.1A CN111721321B (en) 2020-05-19 2020-05-19 Intelligent collision avoidance method and system for navigation mark

Publications (2)

Publication Number Publication Date
CN111721321A CN111721321A (en) 2020-09-29
CN111721321B true CN111721321B (en) 2022-02-22

Family

ID=72564719

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010427248.1A Active CN111721321B (en) 2020-05-19 2020-05-19 Intelligent collision avoidance method and system for navigation mark

Country Status (1)

Country Link
CN (1) CN111721321B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113326640B (en) * 2021-08-03 2021-11-09 北京海兰信数据科技股份有限公司 Method and system for determining collision risk degree of inland ship
CN114464014A (en) * 2022-01-05 2022-05-10 集美大学 Fuzzy logic based regional ship collision risk processing method, system and medium
CN115346398A (en) * 2022-08-19 2022-11-15 江苏恒达微波技术开发有限公司 Ship navigation track monitoring system based on KA waveband satellite data

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2789981A2 (en) * 2013-04-12 2014-10-15 Furuno Electric Co., Ltd. Method and device for displaying route
CN104778837A (en) * 2015-04-14 2015-07-15 吉林大学 Multi-time scale forecasting method for road traffic running situation
CN105590339A (en) * 2015-12-15 2016-05-18 广州南沙3D打印创新研究院 Collision detection method and system for three-dimensional model
CN108922247A (en) * 2018-07-25 2018-11-30 重庆大学 A kind of ship based on AIS-navigation mark Risk-Degree of Collision evaluation method
CN109084747A (en) * 2018-06-26 2018-12-25 武汉理工大学 Water transportation panorama three-dimension navigation system and method based on general three-dimensional engine

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2789981A2 (en) * 2013-04-12 2014-10-15 Furuno Electric Co., Ltd. Method and device for displaying route
CN104778837A (en) * 2015-04-14 2015-07-15 吉林大学 Multi-time scale forecasting method for road traffic running situation
CN105590339A (en) * 2015-12-15 2016-05-18 广州南沙3D打印创新研究院 Collision detection method and system for three-dimensional model
CN109084747A (en) * 2018-06-26 2018-12-25 武汉理工大学 Water transportation panorama three-dimension navigation system and method based on general three-dimensional engine
CN108922247A (en) * 2018-07-25 2018-11-30 重庆大学 A kind of ship based on AIS-navigation mark Risk-Degree of Collision evaluation method

Also Published As

Publication number Publication date
CN111721321A (en) 2020-09-29

Similar Documents

Publication Publication Date Title
CN111721321B (en) Intelligent collision avoidance method and system for navigation mark
CN107329477B (en) Unmanned ship navigation and automatic driving equipment and method thereof
CN108711312B (en) Ship and static object collision risk pre-judging method based on BP neural network
CN106094606A (en) A kind of unmanned surface vehicle navigation and control remote-controlled operation platform
CN108664020A (en) A kind of unmanned boat dynamic obstacle avoidance algorithm based on Speed Obstacles method and dynamic window method
CN106558219B (en) Vehicle track prediction method and device
CN105070101B (en) Cartridge type platform traction Risk-warning and visualization system
CN103353756B (en) Based on the vessel underway oceangoing ship method for real-time monitoring of AIS and VTS information fusion
CN112327845B (en) Unmanned ship formation control method, unmanned ship and computer-readable storage medium
Pietrzykowski et al. Marine integrated navigational decision support system
CN107942095A (en) Wave glider actual heading Forecasting Methodology
KR101658133B1 (en) System for calculating collision risk of vessel
KR101275277B1 (en) Route searching support system of ship for collision avoidance by using the generation of quadrilateral fairway units
CN117250961B (en) Unmanned moving target ship cooperative formation control method and system
Blaich et al. Probabilistic collision avoidance for vessels
CN115373388A (en) Path planning system and method for ship-based robot
CN114061565B (en) Unmanned ship SLAM and application method thereof
CN104180801B (en) Forecasting Methodology and system based on ADS B system track points
JP6984874B2 (en) Voyage planning method and voyage planning system
CN116564136B (en) AIS-based ship collision prediction method
Assaf et al. Ship track estimation using GPS data and Kalman Filter
CN117232520A (en) Ship intelligent navigation system and navigation method suitable for offshore navigation
Dove et al. An automatic collision avoidance and guidance system for marine vehicles in confined waters
Praczyk Correction of navigational information supplied to biomimetic autonomous underwater vehicle
CN111984006B (en) Unmanned ship multi-target meeting collision avoidance method integrating ocean current and scale difference influences

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