CN110440804A - Intelligent navigation method suitable for polar region ice navigation - Google Patents
Intelligent navigation method suitable for polar region ice navigation Download PDFInfo
- Publication number
- CN110440804A CN110440804A CN201910710705.5A CN201910710705A CN110440804A CN 110440804 A CN110440804 A CN 110440804A CN 201910710705 A CN201910710705 A CN 201910710705A CN 110440804 A CN110440804 A CN 110440804A
- Authority
- CN
- China
- Prior art keywords
- ship
- ice
- navigation
- polar region
- polar
- 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- Withdrawn
Links
Classifications
-
- 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
- G01C21/203—Specially adapted for sailing ships
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/26—Visual data mining; Browsing structured data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Remote Sensing (AREA)
- Databases & Information Systems (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Radar, Positioning & Navigation (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- Automation & Control Theory (AREA)
- Navigation (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a kind of intelligent navigation methods suitable for polar region ice navigation, on the basis of sea ice merges display technology in polar region, carry out the flight course planning under the environment of polar region using Q-Learning algorithm.It will be more than that the high ice concentration region of ship ice-breaking capacity is considered as barrier under the environment of polar region, during simulated training, only need whether environmental feedback travels the signal in infeasible region to ship, by successive ignition training, ship can be automatically found a Ship's Optimum Route, without the variation of Environment Obstacles object during concern ship running, application risk Bayesian networks technique assesses navigation risk, provides navigation risk for ship and reminds.
Description
Technical field
The present invention relates to intelligent navigational fields, more particularly to a kind of intelligent navigation side suitable for polar region ice navigation
Method.
Background technique
In recent years, the problems such as global climate increasingly warms, and south poles sea ice is caused to melt, sea-level rise.At the same time,
In summer, the condition in polar region sea area has been able to satisfy the normal pass of ship.With the development of ice-breaking technology, marine technology, I
State is reached the voyage of North America east bank by " northwest course line ", will substantially be shortened by the voyage that " northeast course line " reaches Europe, is saved
The transportation cost of ship is saved.Polar region mining deposits abundant, the scientific investigation team that attract various countries comes to explore, in recent years, more
Start to drive towards polar region sea area come more ships.
Currently, lacking since map open platform Baidu map, Google Maps, Mapbox are defaulted as Mercator projection
The projection of polar region, developer lack polar region sea area map data mining platform, and no APP is shown and divided to polar region sea area ice condition information
Analysis prediction, crewman can not be clearly understood that ice concentration and depth information around ship.It is suitable for and polar region due to lacking
The navigation in sea area, path planning system, ship are selected polar region ice navigation Intermediate Course largely according to navigator's
Experience is completed, and the safety of navigation cannot be ensured effectively.Research is suitable for the intelligent navigation APP of polar region ice navigation
The top priority of system is collected including ice concentration, the sea areas such as sea ice thickness ice condition information, by ice condition data and electron sea
Figure additive fusion provides surrounding sea ice condition information for ship in real time.Destination path is planned on this basis, thus really
Protect the passage normally and efficiently of ship.Any support ice formation path planning, at the same it is reliable in view of being limited on physics and operation
Tool is all popular in maritime world.So a kind of intelligent navigation method suitable for polar region ice navigation is to follow-up work
Development has great significance.
Summary of the invention
The present invention is analyzed by polar region ice condition data and Multidimensional Comprehensive display research, and the ship to travel in polar region, which provides, works as
The preceding and following sea area ice condition information.The ice condition data in sea area analysis shows that on the basis of, severe sea condition and ice navigation are carried out
Safety analysis research passes through the support of data mining and extractive technique, utilizes ice formation course line Q-Learning nitrification enhancement
Method provides the optimal feasible path towards destination for user, provides convenience for arctic navigation.The analysis of mass data is ground
Study carefully, predict ice condition, the optimal path of future time section is provided, provides selection for the trip planning of ship.
Projection is defined to complete polar region map picture using ArcGIS, picture is corresponding with spatial positional information, with
EPSG:3571 carries out projection setting, to figure layer point of addition point geographic position name is corresponding with longitude and latitude, while correcting warp
The precision of latitude stops addition when longitude and latitude error is less than 0.0001 degree, the production of polar region base map is completed, with pole in server
Ground projection vector figure address, polar stereographic projection map overlay is shown in platform.
Ice concentration, the sea ice thickness data file of FTP downloading are normalized, y=(x-MinValue)/
(MaxValue-MinValue), x, y are respectively to convert the maximum value that forward and backward value MaxValue, MinValue is respectively sample
And minimum value, data are converted into the decimal between (0,1), adverse effect caused by unusual sample data is eliminated, data is limited
It is scheduled on the range being easily processed, output result is corresponding with rgb value, ice concentration and sea ice thickness are indicated with color saturation
Size.Ergodic Matrices, by the corresponding ice concentration and sea ice depth rgb value label of representing of longitude and latitude in polar stereographic projection base map
In.
On the basis of sea ice information is merged and shown, path planning is carried out in conjunction with meteorological condition, uses ice formation course line Q-
Learning nitrification enhancement carries out flight course planning to destination, is first abstracted an ambient condition and turns to following expression:
S=(Xg, Xo1, Xo2, Xo3, Xo4)
Wherein, S is an ambient condition in environment space, XgIndicate the orientation situation of the ship target point to be reached,
Xo1, Xo2, Xo3, Xo4For the obstacle information in front of ship, by the distribution of obstacles situation within the scope of 120 ° in front of ship according to angle
Degree is separated into four quantity of state X with distanceo1,Xo2,Xo3,Xo4, wherein angle is with course for 0 ° of benchmark, and barrier and ship connect
Angle of the line relative to course, Xo1,Xo2,Xo3,Xo4The angular range of representative is [- 60 °, -30 °] respectively, [- 30 °, 0 °], [0 °,
30 °], ship movement is reduced to straight line, is advanced with the course for being located at 30 ° on the left of ship's head, to be located at ship by [30 °, 60 °]
30 ° of course is advanced on the right side of first direction, static, respectively corresponds the serial number of 1-5, and the speed of navigation is according to the property of ship polar region environment
Energy model calculates, and the execution time acted each time is fixed；Excitation function is set, when the behavior of ship make itself and target point away from
From approaching, excitation function is awarded, maximum value of awarding when reaching target point.When ship close to can not ice-breaking area
When, penalty value is returned, and when ship and a certain high ice concentration region are bumped against, excitation function should return to one and greatly punish
Penalties guarantees that ship avoids colliding with the region；Movement selection is carried out using ε-greedy strategy, ship is in current ambient conditions
Under there is the probability of ε to randomly choose a movement, and the movement value function for having the probability selection of 1- ε to make current ambient conditions is maximum
Movement, wherein ε be a lesser constant.
Detailed description of the invention
Fig. 1 is a kind of intelligent navigation APP system for being suitable for the intelligent navigation method of polar region ice navigation using the present invention.
Fig. 2 is the query path functional flow diagram for the intelligent navigation method that the present invention is suitable for polar region ice navigation.
Specific embodiment
A specific embodiment of the invention is further illustrated with reference to the accompanying drawing.
As shown in Figure 1, a kind of excellent with data mining and extractive technique, multiple target suitable for the fusion display of polar region sea area ice condition
Change the technologies such as algorithm, safety of ship analytical technology based on data-driven and combines the intelligent navigation APP for forming polar region ice navigation
System.
As shown in Figure 1, ice condition display module is defined projection to complete polar region map picture using ArcGIS, make picture
With spatial positional information, projection setting is carried out with EPSG:3571, by the method for point of addition point by geographic position name
It is corresponding with longitude and latitude, stop addition when inspection data longitude and latitude error is less than 0.0001 degree, completes the calibration of longitude and latitude precision.With
Polar stereographic projection polar plot address in server, polar stereographic projection map overlay is shown in platform.Sea ice information data is carried out
Normalized, y=(x-MinValue)/(MaxValue-MinValue), x, y are respectively to convert forward and backward value
MaxValue, MinValue are respectively the maximum value and minimum value of sample, and data are converted into the decimal between (0,1), processing
Result afterwards is corresponding with rgb value, and the saturation degree of color indicates the size of ice concentration and sea ice thickness.Data are corresponding
Color mark is in polar stereographic projection base map.
As shown in Figure 1, ice prediction module uses data mining statistical learning technology, ice condition data are compiled, it will be through
Latitude matrix is corresponding with ice condition data, establishes model using causal influence network algorithm:
Wherein, Y represents certain month to be predicted sea ice density or depth,It indicated before month to be predicted K months i-th
J-th of Climatic of observation point, Climatic be sea surface temperature, sea-level pressure, ocean surface wind speed,ForRecurrence
Coefficient, Z-kIndicate k months ice concentrations or thickness, q before month to be predicted-kFor Z-kRegression coefficient, O indicate the data moon
Number amount, M are grid data number, and N indicates the Climatic quantity, and ε is constant；Prediction data is compared with truthful data
It is right, prediction accuracy is stepped up, the predictive information of ice condition is provided.
As shown in Figure 1, flight course planning module multiple-objection optimization Q-Learning nitrification enhancement navigates to destination
Line gauge is drawn, and is first abstracted an ambient condition and is turned to following expression:
S=(Xg, Xo1, Xo2, Xo3, Xo4)
Wherein, S is an ambient condition in environment space, XgIndicate the orientation situation of the ship target point to be reached,
Xo1, Xo2, Xo3, Xo4For the obstacle information in front of ship, by the distribution of obstacles situation within the scope of 120 ° in front of ship according to angle
Degree is separated into four quantity of state X with distanceo1,Xo2,Xo3,Xo4, wherein angle is with course for 0 ° of benchmark, and barrier and ship connect
Angle of the line relative to course, Xo1,Xo2,Xo3,Xo4The angular range of representative is [- 60 °, -30 °] respectively, [- 30 °, 0 °], [0 °,
30 °], ship movement is reduced to straight line, is advanced with the course for being located at 30 ° on the left of ship's head, to be located at ship by [30 °, 60 °]
30 ° of course is advanced on the right side of first direction, static, respectively corresponds the serial number of 1-5, and the speed of navigation is according to the property of ship polar region environment
Energy model calculates, and the execution time acted each time is fixed；Excitation function is set, when the behavior of ship make itself and target point away from
From approaching, excitation function is awarded, maximum value of awarding when reaching target point.When ship close to can not ice-breaking area
When, penalty value is returned, and when ship and a certain high ice concentration region are bumped against, excitation function should return to one and greatly punish
Penalties guarantees that ship avoids colliding with the region；Movement selection is carried out using ε-greedy strategy, ship is in current ambient conditions
Under there is the probability of ε to randomly choose a movement, and the movement value function for having the probability selection of 1- ε to make current ambient conditions is maximum
Movement, wherein ε be a lesser constant.
As shown in Figure 1, risk reminding module uses safety of ship analytical technology, vessel underway oceangoing ship is mentioned using Bayesian network
It is reminded for navigation risk, node is the factor for causing ship that accident occurs in arctic navigation, is passed through by expert investigation and history
Data are tested to obtain the dependence between node, correlation point is carried out to bivariate using Pearson product-moment correlation coefficient method
Analysis, production node complete correlation among nodes analysis, arctic navigation risk bayesian network structure figure are constructed, to vessel underway oceangoing ship
Navigation risk is provided to remind.
As shown in Fig. 2, user inputs destination longitude and latitude, system carries out on the basis of current context information after determination
Path planning, display can navigate by water Ship's Optimum Route, show that future can by ice prediction and flight course planning module if cannot navigate by water
Ship's Optimum Route is navigated by water, and shows and can navigate by water the date.
It should be noted that specific embodiment of the present invention illustrates inventive principle and the scope of application.But it is unlimited
In a kind of embodiment, under the exploration of experimental spirits, according to other solutions that the invention occurs, the invention is belonged to
Protection scope.
Claims (1)
1. a kind of intelligent navigation method suitable for polar region ice navigation, which comprises the following steps:
Step 1: making the map with polar region sea ice information；
Step 2: flight course planning being carried out to destination using ice formation course line Q-Learning nitrification enhancement, constructs Bayesian network
Network provides navigation risk for ship and reminds；Bayesian network is constructed the following steps are included: to lead to ship thing in arctic navigation
Therefore factor be node, construct complete Bayesian network, using Bayesian network to vessel underway oceangoing ship provide navigation risk remind,
Dependence between node is obtained by historical data or the method for expert investigation, while using Pearson product-moment correlation coefficient method
Correlation analysis is carried out to bivariate, after completing node and correlation analysis, constructs arctic navigation risk Bayesian network
Structure chart；
Step 1 the following steps are included:
Step 11: projection is defined to polar region map picture, picture is corresponding with spatial positional information, come with EPSG:3571
Projection setting is carried out, it is to figure layer point of addition point that geographic position name is corresponding with longitude and latitude, while the precision of longitude and latitude is corrected,
Stop addition when longitude and latitude error is less than 0.0001 degree；
Step 12: base map is published to server by production polar stereographic projection vector base map；
Step 13: using map overlay method, polar stereographic projection polar plot address in invoking server, by polar stereographic projection map overlay
It is shown in platform, hundred-mark system sea ice density corresponding with longitude and latitude matrix and sea ice thickness data is carried out at following normalization
Reason: y=(x-MinValue)/(MaxValue-MinValue), x, y are respectively to convert forward and backward value, MaxValue,
MinValue is respectively the maximum value and minimum value of sample, and data are converted into the decimal between (0,1), by output result conversion
For rgb value, the size of ice concentration and sea ice depth is represented using the saturation degree of color, loops through longitude, latitude, sea ice
Closeness, sea ice matrix of depths, every time traversal by longitude and latitude it is corresponding represent ice concentration and sea ice depth rgb value label exist
In polar stereographic projection base map；
Step 2 the following steps are included:
Step 21: environment state abstraction in a ship's navigation is turned into following expression:
S=(Xg, Xo1, Xo2, Xo3, Xo4) wherein, S is an ambient condition in environment space, XgIndicate the ship mesh to be reached
The orientation situation of punctuate, Xo1, Xo2, Xo3, Xo4For the obstacle information in front of ship, by the obstacle within the scope of 120 ° in front of ship
Object distribution situation is separated into four quantity of state X according to angle and distanceo1,Xo2,Xo3,Xo4, wherein angle is with course for 0 ° of base
Standard, the angle of barrier and ship line relative to course, Xo1,Xo2,Xo3,Xo4The angular range of representative be respectively [- 60 ° ,-
30 °], [- 30 °, 0 °], [0 °, 30 °], the movement of ship is reduced to straight ahead by [30 °, 60 °], and straight line retreats, to be located at
30 ° of course is advanced on the left of ship's head, is advanced with the course for being located at 30 ° on the right side of ship's head, static, respectively corresponds 1-5's
The speed of serial number, navigation is calculated according to the performance model of ship polar region environment, and the execution time acted each time is fixed；
Step 22: setting excitation function, when the behavior that ship is made makes closer at a distance from target point, excitation function should
Award, guarantee ship can be constantly close to target point, and when reaching target point in subsequent exploration, excitation function
A great reward value should be returned to, during ship's navigation, not allowing to navigate by water is being more than the high sea ice of ship ice-breaking capacity
Closeness region, when ship is close to these regions, excitation function should return to a penalty value, and when ship with it is a certain
When high ice concentration region is bumped against, excitation function should return to a very big penalty value, guarantee in heuristic process later,
Ship avoids colliding with the region as far as possible；
Step 23: movement selection being carried out using ε-greedy strategy, ship has the probability of ε to randomly choose under current ambient conditions
One movement, and the maximum movement of movement value function for thering is the probability selection of 1- ε to make current ambient conditions, wherein ε is one
Lesser constant.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910710705.5A CN110440804A (en) | 2019-08-02 | 2019-08-02 | Intelligent navigation method suitable for polar region ice navigation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910710705.5A CN110440804A (en) | 2019-08-02 | 2019-08-02 | Intelligent navigation method suitable for polar region ice navigation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110440804A true CN110440804A (en) | 2019-11-12 |
Family
ID=68432913
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910710705.5A Withdrawn CN110440804A (en) | 2019-08-02 | 2019-08-02 | Intelligent navigation method suitable for polar region ice navigation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110440804A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112799069A (en) * | 2020-12-30 | 2021-05-14 | 上海海事大学 | Ice region navigation sea ice obstacle avoidance path generation method based on navigation radar image |
| CN113157841A (en) * | 2021-04-06 | 2021-07-23 | 中国科学院西北生态环境资源研究院 | Channel detection method and device, electronic equipment and readable storage medium |
| CN114460934A (en) * | 2022-01-05 | 2022-05-10 | 武汉理工大学 | Visual guidance method, system and device for navigation of icebreaker and storage medium |
| CN116011255A (en) * | 2023-01-18 | 2023-04-25 | 上海交通大学 | Polar region navigation window period assessment system based on ship ice effect three-dimensional visual simulation |
-
2019
- 2019-08-02 CN CN201910710705.5A patent/CN110440804A/en not_active Withdrawn
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112799069A (en) * | 2020-12-30 | 2021-05-14 | 上海海事大学 | Ice region navigation sea ice obstacle avoidance path generation method based on navigation radar image |
| CN112799069B (en) * | 2020-12-30 | 2024-02-13 | 上海海事大学 | Method for generating sea ice obstacle avoidance path of ice region navigation based on navigation radar image |
| CN113157841A (en) * | 2021-04-06 | 2021-07-23 | 中国科学院西北生态环境资源研究院 | Channel detection method and device, electronic equipment and readable storage medium |
| CN114460934A (en) * | 2022-01-05 | 2022-05-10 | 武汉理工大学 | Visual guidance method, system and device for navigation of icebreaker and storage medium |
| CN114460934B (en) * | 2022-01-05 | 2023-08-15 | 武汉理工大学 | Visual guidance method, system and device for ice breaker navigation and storage medium |
| CN116011255A (en) * | 2023-01-18 | 2023-04-25 | 上海交通大学 | Polar region navigation window period assessment system based on ship ice effect three-dimensional visual simulation |
| CN116011255B (en) * | 2023-01-18 | 2023-08-18 | 上海交通大学 | Polar region navigation window period assessment system based on ship ice effect three-dimensional visual simulation |
| US11954806B1 (en) * | 2023-01-18 | 2024-04-09 | Shanghai Jiao Tong University | Polar navigation window period assessment system based on three-dimensional visualization simulation of ship-ice interaction |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110440804A (en) | Intelligent navigation method suitable for polar region ice navigation | |
| Pereira et al. | Risk‐aware path planning for autonomous underwater vehicles using predictive ocean models | |
| Sun et al. | Proximity based automatic data annotation for autonomous driving | |
| Tsou et al. | An Ant Colony Algorithm for efficient ship routing | |
| CN110442665A (en) | Polar region sea area ice condition merges display methods | |
| Christensen et al. | A risk-based approach for determining the future potential of commercial shipping in the Arctic | |
| CN103376109A (en) | Route planning apparatus and route plan verifying method | |
| CN111307160B (en) | Correction method and device for long-endurance flight path of autonomous underwater vehicle | |
| CN105539797A (en) | Navigation method and system of wind power assisting navigation ship based on ECDIS | |
| CN110889198A (en) | Multi-factor joint learning-based dead reckoning probability distribution prediction method and system | |
| CN112347218B (en) | Unmanned ship environment map generation method and unmanned ship sensing system | |
| Chen et al. | FVCOM model estimate of the location of Air France 447 | |
| CN117232520A (en) | Ship intelligent navigation system and navigation method suitable for offshore navigation | |
| CN110319839A (en) | A kind of intelligent navigation APP system suitable for polar region ice navigation | |
| US20220214689A1 (en) | Autonomous Vessel and Infrastructure for Supporting an Autonomous Vessel on Inland Waterways | |
| Zaccone et al. | An optimization tool for ship route planning in real weather scenarios | |
| KR102143322B1 (en) | Method for supplying safe-voyage in Northic sea | |
| Busznyák et al. | On-board diagnostic-based positioning as an additional information source of driver assistant systems | |
| CN113525626B (en) | Ship predicted track correction method based on sea-land constraint | |
| Pan et al. | Research on ship arrival law based on route matching and deep learning | |
| Smith et al. | Autonomous passage planning for a polar vessel | |
| Schütz | Dynamic routing through waters partially covered with sea ice | |
| Zhang et al. | Density-weighted ant colony algorithm for ship trajectory reconstruction | |
| Sasa et al. | Estimation and comparison of accuracy in various data resolutions on optimal ship routing across the North Pacific Ocean | |
| Liu | Research and design of intelligent planning system for ship route based on electronic chart display and information system |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20191112 |