CN109299816A - Ship track real-time predicting method based on Rolling Planning strategy - Google Patents

Ship track real-time predicting method based on Rolling Planning strategy Download PDF

Info

Publication number
CN109299816A
CN109299816A CN201811015766.1A CN201811015766A CN109299816A CN 109299816 A CN109299816 A CN 109299816A CN 201811015766 A CN201811015766 A CN 201811015766A CN 109299816 A CN109299816 A CN 109299816A
Authority
CN
China
Prior art keywords
ship
sampling instant
track
time
real
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.)
Withdrawn
Application number
CN201811015766.1A
Other languages
Chinese (zh)
Inventor
韩云祥
赵景波
李广军
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.)
Jiangsu University of Technology
Original Assignee
Jiangsu University of Technology
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 Jiangsu University of Technology filed Critical Jiangsu University of Technology
Priority to CN201811015766.1A priority Critical patent/CN109299816A/en
Publication of CN109299816A publication Critical patent/CN109299816A/en
Withdrawn legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem

Abstract

The present invention relates to a kind of ship track real-time predicting method based on Rolling Planning strategy, comprises the following steps, and obtains the real-time and historical position information of ship by sea radar first and does preliminary treatment；Then ship track data is pre-processed in each sampling instant, then ship track data is clustered in each sampling instant, parameter training is carried out using Hidden Markov Model to ship track data again and in each sampling instant, then in each sampling instant according to Hidden Markov Model parameter, hidden state q corresponding to current time observation is obtained using Viterbi algorithm, finally pass through setting prediction time domain W in each sampling instant, hidden state q based on ship current time, obtain the position prediction value O of future time period ship, speculate to be rolled in each sampling instant to the track of ship in future time period.Present invention rolling predicts that accuracy is preferable to ship track in real time, to provide a strong guarantee for subsequent ship conflict Resolution.

Description

Ship track real-time predicting method based on Rolling Planning strategy
The application is application No. is 201410849496.X, and invention and created name is " ship track real-time predicting method ", The applying date are as follows: the divisional application of the application for a patent for invention on December 30th, 2014.
Technical field
The present invention relates to a kind of sea area traffic control method more particularly to a kind of ship tracks based on Rolling Planning strategy Real-time predicting method.
Background technique
With the fast development of global shipping business, the traffic in the busy sea area in part is further crowded.It is close in vessel traffic flow Collect complex sea area, still uses sail plan not combine the regulation model of artificial interval allotment not for the collision scenario between ship Adapt to the fast development of shipping business.To guarantee the personal distance between ship, implementing effectively conflict allotment just becomes sea area friendship The emphasis of siphunculus system work.Ship conflict Resolution is a key technology in navigational field, safely and efficiently frees scheme pair It is of great significance in increasing sea area ship flow and ensuring that sea-freight is safe.
In order to improve the efficiency of navigation of ship, marine radar automatic plotter has been widely applied to ship monitor at present In collision prevention, which provides reference frame by extracting the judgement of collision scenario between ship of ship relevant information.Although this Class equipment greatly reduces the load manually monitored, but it does not have the automatic conflict Resolution function of ship.And ship conflict solution De- is based on the basis of the prediction to ship track, in ship real navigation, by meteorological condition, navigation equipment and driving The influence of the various factors such as member's operation, its operating status often not exclusively belongs to a certain specific motion state, in ship rail The influence for needing to consider various enchancement factors during mark prediction, by obtaining the newest characteristic of all kinds of enchancement factors to its future Implement rolling forecast and enhance the robustness of its trajectory predictions in track.
Summary of the invention
The technical problem to be solved in the present invention is to provide a kind of robustness preferably based on the ship rail of Rolling Planning strategy The ship trajectory predictions precision of mark real-time predicting method, this method is higher.
Realize that the technical solution of the object of the invention is to provide a kind of ship track based on Rolling Planning strategy and predicts in real time Method comprises the following steps:
1. obtaining the real-time and historical position information of ship by sea radar, the location information of each ship is discrete two-dimensional Position sequence x'=[x1',x2',...,xn'] and y'=[y1',y2',...,yn'], by application wavelet transformation theory to original Discrete two-dimensional position sequence x'=[x1',x2',...,xn'] and y'=[y1',y2',...,yn'] preliminary treatment is carried out, to obtain Take the denoising discrete two-dimensional position sequence x=[x of ship1,x2,...,xn] and y=[y1,y2,...,yn]；
2. pre-processing in each sampling instant to ship track data, discrete two-dimensional position is denoised according to acquired ship Sequence x=[x1,x2,...,xn] and y=[y1,y2,...,yn], processing is carried out to it using first-order difference method and obtains new ship Oceangoing ship discrete location sequence △ x=[△ x1,△x2,...,△xn-1] and △ y=[△ y1,△y2,...,△yn-1], wherein △ xi= xi+1-xi,△yi=yi+1-yi(i=1,2 ..., n-1)；
3. being clustered in each sampling instant to ship track data, to ship discrete two-dimensional position sequence △ new after processing X and △ y is clustered number M' by setting, is clustered respectively to it using genetic algorithm for clustering；
4. parameter training is carried out using Hidden Markov Model to ship track data in each sampling instant, by that will locate Vessel motion track data △ x and △ y after reason is considered as the aobvious observation of hidden Markov models, by setting hidden state number N and parameter update period τ ', roll the newest Hidden Markov of acquisition according to T' nearest position detection value and using B-W algorithm Model parameter λ '；
5. obtaining current time sight using Viterbi algorithm according to Hidden Markov Model parameter in each sampling instant Hidden state q corresponding to measured value；
6. predicting time domain W, the hidden state q based on ship current time in each sampling instant by setting, obtaining future The position prediction value O of period ship speculates to roll in each sampling instant to the track of ship in future time period.
Further, the step 1. in, by application wavelet transformation theory to original discrete two-dimensional position sequence x'= [x1',x2',...,xn'] and y'=[y1',y2',...,yn'] preliminary treatment is carried out, to obtain the denoising discrete two-dimensional of ship Position sequence x=[x1,x2,...,xn] and y=[y1,y2,...,yn]: for given original two dimensional sequence data x'= [x1',x2',...,xn'], approximation is carried out to it respectively using the linear representation of following form:
Wherein:
F'(x') indicate that, to the function expression obtained after data smoothing processing, ψ (x') indicates female wave, and δ, J and K are small Wave conversion constant, ψJ,K(x') transition form of female wave, c are indicatedJ,KIndicate the function coefficients obtained by wavelet transform procedure, its body Wavelet ψ is showedJ,K(x') to the weight size of entire approximation to function, if this coefficient very little, then it means wavelet ψJ,K(x') Weight it is also smaller, thus can be under the premise of not influence function key property, by wavelet ψ during approximation to functionJ,K (x') it removes；In real data treatment process, implements " threshold transition " by given threshold χ, work as cJ,KWhen < χ, c is setJ,K =0；The selection of threshold function table uses the following two kinds mode:
With
For y'=[y1',y2',...,yn'], denoising is also carried out using the above method.
Further, the step 4. in determine that track Hidden Markov Model the parameter lambda '=process of (π, A, B) is as follows:
4.1) variable assigns initial value: application, which is uniformly distributed, gives variable πi, aijAnd bj(ok) assign initial valueWithAnd make It meets constraint condition:WithThus it obtains λ0=(π0,A0,B0), wherein okIndicate a certain aobvious observation, π0、A0And B0It is by element respectivelyWithThe matrix of composition, Enable parameter l=0, o=(ot-T'+1,...,ot-1,ot) be current time t before T' historical position observation；
4.2) E-M algorithm is executed:
4.2.1) E- step: by λlCalculate ξe(i, j) and γe(si)；
VariableSo
Wherein s indicates a certain hidden state；
4.2.2 it) M- step: usesEstimate respectively Count πi, aijAnd bj(ok) and thus obtain λl+1；
4.2.3) recycle: l=l+1 repeats E- step and M- step, until πi、aijAnd bj(ok) convergence, i.e.,
|P(o|λl+1)-P(o|λl) | < ε, wherein parameter ε=0.00001, return step 4.2.4)；
4.2.4): enabling λ '=λl+1, algorithm terminates.
Further, 5. the middle iterative process for determining the most preferably hidden status switch of ship track is as follows for the step:
5.1) variable assigns initial value: enabling g=2, βT'(si(the s of)=1i∈ S), δ1(si)=πibi(o1), ψ1(si)=0, wherein
, wherein variable ψg(sj) indicate to make variable δg-1(si)aijThe hidden state s of the ship track being maximizedi, parameter S expression The set of hidden state；
5.2) recursive process:
5.3) moment updates: enabling g=g+1, if g≤T', return step 5.2), otherwise iteration ends and go to step 5.4)；
5.4)Go to step 5.5)；
5.5) optimal hidden status switch obtains:
5.5.1) variable assigns initial value: enabling g=T'-1；
5.5.2) backward recursion:
5.5.3) moment updates: g=g-1 is enabled, if g >=1, return step 5.5.2), otherwise terminate.
Further, the step 3. in, cluster number M' value be 4.
Further, the step 4. in, the value of state number N is 3, and it is 30 seconds that parameter, which updates period τ ', T' 10.
Further, the step 6. in, prediction time domain W be 300 seconds.
The present invention has the effect of positive: (1) present invention during ship track is predicted in real time, incorporated it is random because The influence of element, used rolling track prediction scheme can extract the changing condition of extraneous enchancement factor in time, improve ship The accuracy of oceangoing ship trajectory predictions.
(2) the present invention is based on different performance index, the real-time prediction result in ship track can be in the presence of the multiple of conflict Trajectory planning scheme is freed in ship offer, improves the economy of vessel motion and the utilization rate of sea area resources.
Detailed description of the invention
Fig. 1 is the short-term track product process schematic diagram of vessel motion in the present invention.
Specific embodiment
(embodiment 1)
See that Fig. 1, the ship track real-time predicting method based on Rolling Planning strategy of the present embodiment include following several steps It is rapid:
1. obtaining the real-time and historical position information of ship by sea radar, the location information of each ship is discrete two-dimensional Position sequence x'=[x1',x2',...,xn'] and y'=[y1',y2',...,yn'], by application wavelet transformation theory to original Discrete two-dimensional position sequence x'=[x1',x2',...,xn'] and y'=[y1',y2',...,yn'] preliminary treatment is carried out, to obtain Take the denoising discrete two-dimensional position sequence x=[x of ship1,x2,...,xn] and y=[y1,y2,...,yn]: y=[y1,y2,..., yn]: for given original two dimensional sequence data x'=[x1',x2',...,xn'], utilize the linear representation point of following form It is other that approximation is carried out to it:
Wherein:
F'(x') indicate that, to the function expression obtained after data smoothing processing, ψ (x') indicates female wave, and δ, J and K are small Wave conversion constant, ψJ,K(x') transition form of female wave, c are indicatedJ,KIndicate the function coefficients obtained by wavelet transform procedure, its body Wavelet ψ is showedJ,K(x') to the weight size of entire approximation to function, if this coefficient very little, then it means wavelet ψJ,K(x') Weight it is also smaller, thus can be under the premise of not influence function key property, by wavelet ψ during approximation to functionJ,K (x') it removes；In real data treatment process, implements " threshold transition " by given threshold χ, work as cJ,KWhen < χ, c is setJ,K =0；The selection of threshold function table uses the following two kinds mode:
With
For y'=[y1',y2',...,yn'], denoising is also carried out using the above method；
2. pre-processing in each sampling instant to ship track data, discrete two-dimensional position is denoised according to acquired ship Sequence x=[x1,x2,...,xn] and y=[y1,y2,...,yn], processing is carried out to it using first-order difference method and obtains new ship Oceangoing ship discrete location sequence △ x=[△ x1,△x2,...,△xn-1] and △ y=[△ y1,△y2,...,△yn-1], wherein △ xi= xi+1-xi,△yi=yi+1-yi(i=1,2 ..., n-1)；
3. being clustered in each sampling instant to ship track data, to ship discrete two-dimensional position sequence △ new after processing X and △ y is clustered number M' by setting, is clustered respectively to it using genetic algorithm for clustering；
4. parameter training is carried out using Hidden Markov Model to ship track data in each sampling instant, by that will locate Vessel motion track data △ x and △ y after reason is considered as the aobvious observation of hidden Markov models, by setting hidden state number N and parameter update period τ ', roll the newest Hidden Markov of acquisition according to T' nearest position detection value and using B-W algorithm Model parameter λ '；Determine that track Hidden Markov Model the parameter lambda '=process of (π, A, B) is as follows:
4.1) variable assigns initial value: application, which is uniformly distributed, gives variable πi, aijAnd bj(ok) assign initial valueWithAnd It is set to meet constraint condition:WithThus To λ0=(π0,A0,B0), wherein okIndicate a certain aobvious observation, π0、A0And B0It is by element respectivelyWithThe square of composition Battle array, enables parameter l=0, o=(ot-T'+1,...,ot-1,ot) be current time t before T' historical position observation；
4.2) E-M algorithm is executed:
4.2.1) E- step: by λlCalculate ξe(i, j) and γe(si)；
VariableSo
Wherein s indicates a certain hidden state；
4.2.2 it) M- step: usesEstimate respectively Count πi, aijAnd bj(ok) and thus obtain λl+1；
4.2.3) recycle: l=l+1 repeats E- step and M- step, until πi、aijAnd bj(ok) convergence, i.e.,
|P(o|λl+1)-P(o|λl) | < ε, wherein parameter ε=0.00001, return step 4.2.4)；
4.2.4): enabling λ '=λl+1, algorithm terminates.
5. obtaining current time sight using Viterbi algorithm according to Hidden Markov Model parameter in each sampling instant Hidden state q corresponding to measured value:
5.1) variable assigns initial value: enabling g=2, βT'(si(the s of)=1i∈ S), δ1(si)=πibi(o1), ψ1(si)=0, wherein
, wherein variable ψg(sj) indicate to make variable δg-1(si)aijThe hidden state s of the ship track being maximizedi, parameter S expression The set of hidden state；
5.2) recursive process:
5.3) moment updates: enabling g=g+1, if g≤T', return step 5.2), otherwise iteration ends and go to step 5.4)；
5.4)Go to step 5.5)；
5.5) optimal hidden status switch obtains:
5.5.1) variable assigns initial value: enabling g=T'-1；
5.5.2) backward recursion:
5.5.3) moment updates: g=g-1 is enabled, if g >=1, return step 5.5.2), otherwise terminate.
6. predicting time domain W, the hidden state q based on ship current time in each sampling instant by setting, obtaining future The position prediction value O of period ship.
The value of above-mentioned cluster number M' is 4, the value of state number N is 3, and it is 30 seconds, T' 10 that parameter, which updates period τ ', Predict that time domain W is 300 seconds.
(application examples, navigation traffic control method)
The navigation traffic control method of the present embodiment comprises the following steps:
Step A, ship is obtained according to the ship track real-time predicting method based on Rolling Planning strategy that embodiment 1 obtains The track of ship in the future time period that each sampling instant speculates；
Step B, sea is obtained based on the current operating status of ship and historical position observation sequence in each sampling instant The numerical value of domain wind field variable, detailed process is as follows:
B.1 the stop position of ship) is set as track reference coordinate origin；
B.2) when ship is in straight running condition and at the uniform velocity turning operating status, sea area wind field linear filtering mould is constructed Type；
B.3 the numerical value of wind field variable) is obtained according to constructed Filtering Model.
Step C, it in each sampling instant, is needed when the ship of operating status and setting based on each ship is run in sea area The safety regulation collection of satisfaction, when being possible to be in the presence of violating safety regulation between ship, to its dynamic behaviour implementing monitoring And timely warning information is provided for maritime traffic control centre；
Step D, when warning information occurs, under the premise of meeting ship physical property and sea area traffic rules, pass through Set optimizing index function and incorporate wind field variable value, using Model Predictive Control Theory method to ship collision avoidance track into Row Rolling Planning, and program results are transferred to each ship and are executed, detailed process is as follows:
D.1 termination reference point locations P, the collision avoidance policy control time domain Θ, trajectory predictions of ship collision avoidance trajectory planning) are set Time domain Υ；
D.2 under the premise of) being set in given optimizing index function, it is based on cooperative collision avoidance trajectory planning thought, by giving Each ship assigns different weights and incorporates real-time wind field variable filtering numerical value, obtains the collision avoidance track of each ship and keeps away It hits control strategy and program results is transferred to each ship and execute, and each ship only implements its first in Rolling Planning interval Optimal Control Strategy；
D.3) in next sampling instant, step is repeated D.2 until each ship reaches it and frees terminal.
Above-mentioned termination reference point locations P is set as next navigation channel point of vessel position conflict point, when collision avoidance policy control Domain Θ is 300 seconds；Trajectory predictions time domain Υ is 300 seconds.
Obviously, the above embodiment is merely an example for clearly illustrating the present invention, and is not to of the invention The restriction of embodiment.For those of ordinary skill in the art, it can also be made on the basis of the above description Its various forms of variation or variation.There is no necessity and possibility to exhaust all the enbodiments.And these belong to this hair The obvious changes or variations that bright spirit is extended out are still in the protection scope of this invention.

Claims (2)

1. a kind of ship track real-time predicting method based on Rolling Planning strategy, it is characterised in that comprise the following steps:
1. obtaining the real-time and historical position information of ship by sea radar, the location information of each ship is discrete two-dimensional position Sequence x'=[x1',x2',K,xn'] and y'=[y1',y2',K,yn'], by application wavelet transformation theory to original discrete two-dimensional Position sequence x'=[x1',x2',K,xn'] and y'=[y1',y2',K,yn'] preliminary treatment is carried out, to obtain the denoising of ship Discrete two-dimensional position sequence x=[x1,x2,K,xn] and y=[y1,y2,K,yn]: for given original two dimensional sequence data x'= [x1',x2',K,xn'], approximation is carried out to it respectively using the linear representation of following form:
Wherein:
F'(x' it) indicates to the function expression obtained after data smoothing processing, ψ (x') indicates that female wave, δ, J and K are that small echo becomes Change constant, ψJ,K(x') transition form of female wave, c are indicatedJ,KIndicate the function coefficients obtained by wavelet transform procedure；In actual number According in treatment process, implements " threshold transition " by given threshold χ, work as cJ,KWhen < χ, c is setJ,K=0；The choosing of threshold function table It takes using the following two kinds mode:
With
For y'=[y1',y2',K,yn'], denoising is also carried out using the above method；
2. pre-processing in each sampling instant to ship track data, discrete two-dimensional position sequence is denoised according to acquired ship X=[x1,x2,K,xn] and y=[y1,y2,K,yn], processing is carried out to it using first-order difference method and obtains new ship discrete bits Set sequence △ x=[△ x1,△x2,K,△xn-1] and △ y=[△ y1,△y2,K,△yn-1], wherein △ xi=xi+1-xi,△yi =yi+1-yi, i=1,2, K, n-1；
3. ship track data is clustered in each sampling instant, to ship discrete two-dimensional position sequence △ x new after processing and △ y is clustered number M' by setting, is clustered respectively to it using genetic algorithm for clustering；
4. parameter training is carried out using Hidden Markov Model to ship track data in each sampling instant, after it will handle Vessel motion track data △ x and △ y be considered as the aobvious observations of hidden Markov models, by set hidden state number N and Parameter updates period τ ', rolls the newest Hidden Markov mould of acquisition according to T' nearest position detection value and using B-W algorithm Shape parameter λ '；
5. obtaining current time observation using Viterbi algorithm according to Hidden Markov Model parameter in each sampling instant Corresponding hidden state q:
5.1) variable assigns initial value: enabling g=2, βT′(si)=1, si∈ S, δ1(si)=πibi(o1), ψ1(si)=0, whereinWherein variable ψg(sj) Expression makes variable δg-1(si)aijThe hidden state s of the ship track being maximizedi, parameter S indicates the set of hidden state；
5.2) recursive process:
5.3) moment updates: enabling g=g+1, if g≤T', return step 5.2), otherwise iteration ends and go to step 5.4)；
5.4)Go to step 5.5)；
5.5) optimal hidden status switch obtains:
5.5.1) variable assigns initial value: enabling g=T'-1；
5.5.2) backward recursion:
5.5.3) moment updates: g=g-1 is enabled, if g >=1, return step 5.5.2), otherwise terminate；
6. predicting time domain W, the hidden state q based on ship current time in each sampling instant by setting, obtaining future time period The position prediction value O of ship speculates to roll in each sampling instant to the track of ship in future time period.
2. the ship track real-time predicting method according to claim 1 based on Rolling Planning strategy, it is characterised in that: institute State step 3. in, cluster number M' value be 4.
CN201811015766.1A 2014-12-30 2014-12-30 Ship track real-time predicting method based on Rolling Planning strategy Withdrawn CN109299816A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811015766.1A CN109299816A (en) 2014-12-30 2014-12-30 Ship track real-time predicting method based on Rolling Planning strategy

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811015766.1A CN109299816A (en) 2014-12-30 2014-12-30 Ship track real-time predicting method based on Rolling Planning strategy
CN201410849496.XA CN104484726B (en) 2014-12-30 2014-12-30 Ship track real-time predicting method

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
CN201410849496.XA Division CN104484726B (en) 2014-12-30 2014-12-30 Ship track real-time predicting method

Publications (1)

Publication Number Publication Date
CN109299816A true CN109299816A (en) 2019-02-01

Family

ID=52759267

Family Applications (4)

Application Number Title Priority Date Filing Date
CN201811015762.3A Withdrawn CN109299815A (en) 2014-12-30 2014-12-30 Ship trajectory predictions method
CN201811015766.1A Withdrawn CN109299816A (en) 2014-12-30 2014-12-30 Ship track real-time predicting method based on Rolling Planning strategy
CN201811015761.9A Pending CN108875954A (en) 2014-12-30 2014-12-30 Ship trajectory predictions method based on Rolling Planning strategy
CN201410849496.XA Active CN104484726B (en) 2014-12-30 2014-12-30 Ship track real-time predicting method

Family Applications Before (1)

Application Number Title Priority Date Filing Date
CN201811015762.3A Withdrawn CN109299815A (en) 2014-12-30 2014-12-30 Ship trajectory predictions method

Family Applications After (2)

Application Number Title Priority Date Filing Date
CN201811015761.9A Pending CN108875954A (en) 2014-12-30 2014-12-30 Ship trajectory predictions method based on Rolling Planning strategy
CN201410849496.XA Active CN104484726B (en) 2014-12-30 2014-12-30 Ship track real-time predicting method

Country Status (1)

Country Link
CN (4) CN109299815A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111428816A (en) * 2020-04-17 2020-07-17 贵州电网有限责任公司 Non-invasive load decomposition method

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104899263B (en) * 2015-05-22 2018-01-26 华中师范大学 A kind of ship track mining analysis and monitoring method based on specific region
CN105554101A (en) * 2015-12-15 2016-05-04 南京欣网互联网络科技有限公司 E-commerce system and method based on flow request
CN105930921A (en) * 2016-04-13 2016-09-07 石河子大学 Prediction method for two-spotted spider mites
CN107256438B (en) * 2017-05-26 2020-05-01 亿海蓝（北京）数据技术股份公司 Method and device for predicting residence time of ship port
CN110363094A (en) * 2019-06-20 2019-10-22 珠海云航智能技术有限公司 A kind of ship abnormal behaviour recognition methods, device and terminal device
CN110333726A (en) * 2019-07-29 2019-10-15 武汉理工大学 A kind of safety of ship DAS (Driver Assistant System) based on ship motion prediction
CN112182133B (en) * 2020-09-29 2022-02-15 南京北斗创新应用科技研究院有限公司 AIS data-based ship loitering detection method
CN112562372B (en) * 2020-11-30 2021-11-16 腾讯科技（深圳）有限公司 Track data processing method and related device
CN112966332B (en) * 2021-03-02 2022-06-21 武汉理工大学 Conflict detection method based on multi-ship motion uncertainty, memory and processor
CN113110468B (en) * 2021-04-22 2022-07-26 中国船舶重工集团公司第七0七研究所九江分部 Control method applied to autonomous berthing of under-actuated double-paddle double-rudder ship
CN113221450B (en) * 2021-04-27 2024-03-12 中国科学院国家空间科学中心 Space-time prediction method and system for sparse non-uniform time sequence data

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1280610C (en) * 2001-11-14 2006-10-18 财团法人资讯工业策进会 Flightpath position forcasting method in combined radar/automatic feedback reporting monitoring environment
CN101750061B (en) * 2008-12-10 2013-04-03 北京普及芯科技有限公司 Method and device for target flight path detection/course prediction
CN102819663B (en) * 2012-07-17 2015-04-08 哈尔滨工程大学 Method for forecasting ship wake based on optimized support vector regression parameter
CN102999789A (en) * 2012-11-19 2013-03-27 浙江工商大学 Digital city safety precaution method based on semi-hidden-markov model
CN102968625B (en) * 2012-12-14 2015-06-10 南京思创信息技术有限公司 Ship distinguishing and tracking method based on trail

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111428816A (en) * 2020-04-17 2020-07-17 贵州电网有限责任公司 Non-invasive load decomposition method
CN111428816B (en) * 2020-04-17 2023-01-20 贵州电网有限责任公司 Non-invasive load decomposition method

Also Published As

Publication number Publication date
CN104484726A (en) 2015-04-01
CN109299815A (en) 2019-02-01
CN108875954A (en) 2018-11-23
CN104484726B (en) 2018-10-09

Similar Documents

Publication Publication Date Title
CN104484726B (en) Ship track real-time predicting method
CN104462856B (en) Ship conflict method for early warning
CN104537891B (en) A kind of boats and ships track real-time predicting method
CN104504277B (en) A kind of ship conflict method for early warning
CN110648014B (en) Regional wind power prediction method and system based on space-time quantile regression
CN104504934B (en) A kind of navigation traffic control method
CN114022847A (en) Intelligent agent trajectory prediction method, system, equipment and storage medium
CN105654139A (en) Real-time online multi-target tracking method adopting temporal dynamic appearance model
CN102142085B (en) Robust tracking method for moving flame target in forest region monitoring video
CN107045803A (en) Navigation traffic control method based on Rolling Planning strategy
CN112907970B (en) Variable lane steering control method based on vehicle queuing length change rate
CN116383685A (en) Vehicle lane change detection method based on space-time interaction diagram attention network
CN109711431A (en) The method for tracking target of local block convolution, system and storage medium at one
CN106571067A (en) Planning method for ship conflict resolution based on rolling planning strategy
Kaur et al. Stock market forecasting using ANFIS with OWA operator
CN109002860B (en) Intelligent adaptive matching prediction method for sudden change wind speed along high-speed railway
Zhang et al. Motion Prediction of Tugboats Using Hidden Markov Model
CN104966080B (en) A kind of sea monitoring data sequent infrared image Weak target determination method and device
Huang Short-term wind speed forecasting model based on an attention-gated recurrent neural network and error correction strategy
CN112307893A (en) Vehicle tracking method in community monitoring scene
CN115239767A (en) Dynamic passenger flow behavior situation prediction method, system, storage medium and equipment
CN112966641A (en) Intelligent decision-making method for multiple sensors and multiple targets and storage medium
CN114676824A (en) Grey correlation analysis and power distribution network line loss prediction method under PSO-RBF
CN117094432A (en) Ocean thermal wave event prediction method based on interpretive deep learning
CN117875359A (en) Graph contrast learning resistance attack method based on momentum gradient candidates

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
WW01 Invention patent application withdrawn after publication

Application publication date: 20190201

WW01 Invention patent application withdrawn after publication