CN103968833B - A kind of method choosing observation triangle before star pattern matching - Google Patents
A kind of method choosing observation triangle before star pattern matching Download PDFInfo
- Publication number
- CN103968833B CN103968833B CN201410182261.XA CN201410182261A CN103968833B CN 103968833 B CN103968833 B CN 103968833B CN 201410182261 A CN201410182261 A CN 201410182261A CN 103968833 B CN103968833 B CN 103968833B
- Authority
- CN
- China
- Prior art keywords
- star
- triangle
- observation
- bright
- value
- 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.)
- Expired - Fee Related
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
- G01C21/02—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by astronomical means
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- General Physics & Mathematics (AREA)
- Astronomy & Astrophysics (AREA)
- Navigation (AREA)
Abstract
A kind of method choosing observation triangle before the present invention relates to star pattern matching, comprises the following steps: primary election observation triangle；Observation triangle is carried out double optimization screening, and screening conditions include: in image planes, the minimum edge of composition observation triangle should be greater than dminIndividual pixel；Observe minimum angle in triangle in image planes and should be greater than θmin.The most existing method of method of observation triangle is chosen before the star pattern matching of the present invention, once the match is successful and the probability of attitude output can to improve 60.8%, in Practical Project, can preferentially choose three groups be identified compared with the observation triangle of bright star composition and verify, the probability i.e. having 77.5% can improve determine the successful speed of appearance to three couplings, and every frame whole day ball is determined the time consistency that appearance exports and is also greatly improved.
Description
Technical field
The present invention relates to autonomous navigation of satellite technical field, choose observation three particularly to before a kind of star pattern matching
The method of dihedral.
Background technology
In Practical Project, star sensor is mainly based on triangle star map recognizer, so that it is determined that attitude.
Triangle star map recognizer utilizes the observation star composition observation triangle occurred in star sensor visual field, with entirely
Equilateral triangle principle is match cognization standard.Choose suitably observation triangle and can improve star Pattern Recognition Algorithm
Success rate, the amount of calculation reduced in matching process, and then improve star sensor and export the performance of attitude in real time.
Zhang Guangjun et al. proposes one and chooses observation deltic method, and it chooses observation triangle principle: choose star quick
In sensor visual field, quantity is NBOne group of bright star, arbitrarily composition observation triangle, preferentially chooses the brightest star structure
The observation triangle become is identified.If a certain observation triangle combination recognition failures, choose the most in order
Remaining observation triangle sets is incompatible to be identified, until obtaining correct recognition result.Due in reality
In the situation of border, selected NBObservation star in there may be double star or wherein three bright stars in visual field
Position is approximated to the situation of straight line, and therefore this will appear from bulk redundancy coupling, extends recognition time, enters
And cause importance in star map recognition failure or the bigger attitude of output error.Take NB=6, whole day ball is traveled through imaging, system
Meter visual field occurs None-identified or exports the observation triangle number purpose probability distribution of bigger error attitude, as
Shown in Fig. 1, in whole day ball field range, in 20 the observation trianglees preferentially chosen, averagely have 7.74
Individual observation triangle None-identified or the bigger error attitude of output, i.e. when each importance in star map recognition, have 38.7%
Probability cannot the most directly export final carriage, during actual importance in star map recognition, due to star pattern matching one
Secondary amount of calculation significantly larger than visual field selects the amount of calculation required for optimum observation triangle, and sometimes
Even if importance in star map recognition success, but because of observation star position approximation straight line in visual field, cause exporting
Normal attitude, needs again to choose observation triangle and carries out match cognization, and therefore, this will have a strong impact on star
The real-time of sensor output attitude.
Summary of the invention
The invention solves the problems that technical problem of the prior art, it is provided that before a kind of star pattern matching, choose observation triangle
The method of shape.
In order to solve above-mentioned technical problem, technical scheme is specific as follows:
A kind of method choosing observation triangle before star pattern matching, comprises the following steps:
Primary election observation triangle；
Observation triangle is carried out double optimization screening, and screening conditions include:
In image planes, the minimum edge of composition observation triangle should be greater than dminIndividual pixel；
Observe minimum angle in triangle in image planes and should be greater than θmin。
In technique scheme, the method specifically includes:
Step 1: the observation star occurred in star sensor visual field is carried out ascending order arrangement by its magnitude information；
Step 2: on the basis of the N bright star, sequentially chooses N+1, N+2 bright star composition triangle, will
The value of N+3 is assigned to n；The initial value of N is set to 1；
Step 3: calculate the Atria limit length of side of composition, it is judged that whether the minimum length of side is more than dmin, if it is,
Perform step 4 to determine whether；If it is not, output " observation triangle exists double star ", perform step 7；
Step 4: judge that in triangle, whether minimum angle is more than θmin, if it is, by these three bright star records
In alternative group of triangle of observation, reject and selected a star the darkest in three stars, introduce n-th star composition new
Triangle, performs step 5；If it is not, output " three observation star approximation straight lines ", perform step 10；
Step 5: judge whether n-th bright star is last bright star in visual field, if it is, by N+1's
Value is assigned to N, performs step 6；If it is not, the value of n+1 is assigned to n, perform step 3；
Step 6: judge whether the N bright star is second from the bottom bright star in visual field, if it is, observation three
Alternative group of dihedral is chosen complete；If it is not, perform step 2；
Step 7: judge whether n-th bright star is last bright star in visual field, if it is, output is " the brightest
Star is a star in certain nebula ", the value of N+1 is assigned to N, performs step 8；If it is not, perform step
Rapid 9；
Step 8: judge whether the N bright star is second from the bottom bright star in visual field, if it is, select star to lose
Lose；If it is not, perform step 2；
Step 9: reject the secondary bright star of the minimum length of side of composition, chooses n-th bright star composition triangle, then will
The value of n+1 is assigned to n, performs step 3；
Step 10: judge whether n-th bright star is last bright star in visual field, if it is, by N+1's
Value is assigned to N, performs step 8；If it is not, perform step 11；
Step 11: reject and selected a star the darkest in three stars, introduces n-th star and forms new triangle,
Then the value of n+1 is assigned to n, performs step 3.
The present invention has a following beneficial effect:
Choose the method for observation triangle before the star pattern matching of the present invention to accelerate star sensor determine appearance output fast
Degree, improving selected observation triangle can the probability of a successful match output final carriage.
The most existing Zhang Guangjun of method choosing observation triangle before the star pattern matching of the present invention et al. proposes
Method, once the match is successful and the probability of attitude output can to improve 60.8%, in Practical Project, can be preferential
Choose three groups be identified compared with the observation triangle of bright star composition and verify, i.e. have the probability of 77.5% to improve right
Three times coupling determine the successful speed of appearance, and every frame whole day ball is determined the time consistency that appearance exports and also significantly carried
High.
Accompanying drawing explanation
With detailed description of the invention, the present invention is described in further detail below in conjunction with the accompanying drawings.
Fig. 1 is to occur None-identified in traversal whole day ball visual field or export the observation triangle of bigger error attitude
The probability distribution of number.
Fig. 2 is star sensor image space coordinate system schematic diagram.
Fig. 3 is the observation triangle number purpose probability distribution meeting constraints 1,2 in traversal whole day ball visual field
Figure.
Fig. 4 is to optimize observation triangle choosing method flow chart.
Detailed description of the invention
The invention thought of the present invention is:
In order to accelerate, star sensor determines appearance output speed, improving selected observation triangle can a successful match output
The probability of final carriage, before the present invention provides a kind of star pattern matching, " double optimization " chooses observation deltic method.
Observation triangle only meets some requirements, and could realize once mating, determine appearance and effectively calculate.For
Improving computational efficiency, the present invention, on the basis of primary election observation triangle, carries out two suboptimums to observation triangle
Change screening, the star Pattern Recognition Algorithm mated the most again.Observation Triangles Optimization is chosen, be according to meeting simultaneously
The double constraints that correct coupling calculates with high-precision fixed appearance, for improving computation rate in-orbit, needs by deeply
Analyze, be converted into the image planes immediate constraint condition of calculating concrete, easy.
During star pattern matching, for improving punctate opacity of the cornea Centroid accuracy, general employing out of focus technology, simultaneously double star
Impact on match cognization be can not ignore, and for making two observation star barycenter be kept completely separate, its length of side distance should be extremely
Less more than a certain threshold value.Factors above is as double optimization screening observation triangle restriction condition 1, this constraint bar
Part determines observation triangle minimum length of side dminChoosing of threshold value.
Three stars of composition observation triangle position relationship output final to star sensor attitude in visual field
Result impact is very big, such as, during certain three observation star approximation straight lines, it is fixed that this will result directly in star sensor
Appearance failure, this is primarily due to reference vector battle array V3×k=(V1V2…Vk) during the calculating of matrix inversion,
When observation Atria star position correlation is stronger when, (VTV) highly near-singular.Three will be observed
The limit of dihedral and the restriction relation of angle are transformed in image plane analysis, as in figure 2 it is shown, star sensor picture is empty
Between coordinate system schematic diagram, O is the initial point of star sensor image space coordinate system, and Star1, Star2, Star3 are permanent
Star projection in image plane, f is the focal length of star sensor, and h is the Star2 vertical line to base c, and θ is triangle
Minimum angle, i.e. a limit and the angle on c limit in shape, β is the angle that h limit is corresponding.Obtained by Fig. 2 triangular relationship:
H=a sin θ (1)
For avoiding the double star interference to importance in star map recognition, reduce three observation star dependencys of position in image plane simultaneously,
I.e. three observation star not conllinear in image planes, determine appearance principle analysis from QUEST, and h should be at least above double star door
Limit threshold value, it may be assumed that
H=f tan β > double star threshold value (2)
(1), (2) formula as double optimization screening observation triangle restriction condition 2, this constraints determines observation three
Dihedral minimum angle thetaminChoosing of threshold value.
The present invention, on the basis of primary election observation triangle, carries out double optimization screening to observation triangle.Just
Choosing observation triangle refers to that preferentially choosing N (N=5～6) star the brightest in visual field arbitrarily forms observation triangle,
These observation trianglees are formed observation triangle sets to be selected.Double optimization screening observation triangle is with as follows
Carry out based on constraints:
Constraints 1: in image planes, the minimum edge of composition observation triangle should be greater than dminIndividual pixel.
Constraints 2: in image planes, in observation triangle, minimum angle should be greater than θmin。
In terms of technique effect, the present invention has following performance.QUEST is used to determine appearance algorithm, if dmin=10,
θmin=15 °, whole day ball is traveled through, statistics visual field occurs directly exporting the observation triangle number of attitude
Purpose probability distribution, i.e. meets the observation triangle number purpose probability distribution of constraints 1,2, such as Fig. 3 institute
Show, in whole day ball field range, average each visual field has 12.2 observation trianglees can directly export appearance
State, the most existing method, once the match is successful and the probability of attitude output, in reality can to improve 60.8%
In engineering, can preferentially choose three groups and be identified compared with the observation triangle of bright star composition and verify, i.e. have
The probability of 77.5% can improve determine the successful speed of appearance to three couplings, and every frame whole day ball determine that appearance exports time
Between concordance be also greatly improved.
Below in conjunction with the accompanying drawings the present invention is described in detail.
Fig. 4 is to optimize observation triangle choosing method flow chart.M is the number occurring observation star in observation visual field
Mesh, dminIt is observation triangle minimum length of side threshold value, θminBeing observation triangle minimum angle threshold value, N is observation
Star sequence number, initial value is set to 1, and n is the 4th bright star sequence number beyond selected three bright stars.
Step 1: the observation star occurred in star sensor visual field is carried out ascending order arrangement (magnitude by its magnitude information
The least brightness is the highest).
Step 2: on the basis of N (initial value of N is set to a 1) bright star, sequentially choose N+1, N+2
Bright star composition triangle, is assigned to n by the value of N+3.
Step 3: calculate the Atria limit length of side of composition, it is judged that whether the minimum length of side is more than dmin, if it is,
Perform step 4 to determine whether；If it is not, output " observation triangle exists double star ", perform step 7.
Step 4: judge that in triangle, whether minimum angle is more than θmin, if it is, by these three bright star records
In alternative group of triangle of observation, reject and selected a star the darkest in three stars, introduce n-th star composition new
Triangle, performs step 5；If it is not, output " three observation star approximation straight lines ", perform step 10.
Step 5: judge whether n-th bright star is last bright star in visual field, if it is, by N+1's
Value is assigned to N, performs step 6；If it is not, the value of n+1 is assigned to n, perform step 3.
Step 6: judge whether the N bright star is second from the bottom bright star in visual field, if it is, observation three
Alternative group of dihedral is chosen complete；If it is not, perform step 2.
Step 7: judge whether n-th bright star is last bright star in visual field, if it is, output is " the brightest
Star is a star in certain nebula ", the value of N+1 is assigned to N, performs step 8；If it is not, perform step
Rapid 9.
Step 8: judge whether the N bright star is second from the bottom bright star in visual field, if it is, select star to lose
Lose；If it is not, perform step 2.
Step 9: reject the secondary bright star of the minimum length of side of composition, chooses n-th bright star composition triangle, then will
The value of n+1 is assigned to n, performs step 3.
Step 10: judge whether n-th bright star is last bright star in visual field, if it is, by N+1's
Value is assigned to N, performs step 8；If it is not, perform step 11.
Step 11: reject and selected a star the darkest in three stars, introduces n-th star and forms new triangle,
Then the value of n+1 is assigned to n, performs step 3.
Obviously, above-described embodiment is only for clearly demonstrating example, and not to embodiment
Limit.For those of ordinary skill in the field, can also be made it on the basis of the above description
The change of its multi-form or variation.Here without also cannot all of embodiment be given exhaustive.And by
What this was extended out obviously changes or changes among still in the protection domain of the invention.
Claims (1)
1. the method choosing observation triangle before a star pattern matching, it is characterised in that comprise the following steps:
Step 1: the observation star occurred in star sensor visual field is carried out ascending order arrangement by its magnitude information；
Step 2: on the basis of the N bright star, sequentially chooses N+1, N+2 bright star composition triangle, will
The value of N+3 is assigned to n；The initial value of N is set to 1；
Step 3: calculate the Atria limit length of side of composition, it is judged that whether the minimum length of side is more than dmin, if it is,
Perform step 4 to determine whether；If it is not, output " observation triangle exists double star ", perform step 7；
Step 4: judge that in triangle, whether minimum angle is more than θmin, if it is, by these three bright star records
In alternative group of triangle of observation, reject and selected a star the darkest in three stars, introduce n-th star composition
New triangle, performs step 5；If it is not, output " three observation star approximation straight lines ", perform step
10；
Step 5: judge whether n-th bright star is last bright star in visual field, if it is, by N+1's
Value is assigned to N, performs step 6；If it is not, the value of n+1 is assigned to n, perform step 3；
Step 6: judge whether the N bright star is second from the bottom bright star in visual field, if it is, observation three
Alternative group of dihedral is chosen complete；If it is not, perform step 2；
Step 7: judge whether n-th bright star is last bright star in visual field, if it is, output is " the brightest
Star is a star in certain nebula ", the value of N+1 is assigned to N, performs step 8；If it is not, perform step
Rapid 9；
Step 8: judge whether the N bright star is second from the bottom bright star in visual field, if it is, select star to lose
Lose；If it is not, perform step 2；
Step 9: reject the secondary bright star of the minimum length of side of composition, chooses n-th bright star composition triangle, then will
The value of n+1 is assigned to n, performs step 3；
Step 10: judge whether n-th bright star is last bright star in visual field, if it is, by N+1's
Value is assigned to N, performs step 8；If it is not, perform step 11；
Step 11: reject and selected a star the darkest in three stars, introduces n-th star and forms new triangle,
Then the value of n+1 is assigned to n, performs step 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410182261.XA CN103968833B (en) | 2014-04-30 | 2014-04-30 | A kind of method choosing observation triangle before star pattern matching |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410182261.XA CN103968833B (en) | 2014-04-30 | 2014-04-30 | A kind of method choosing observation triangle before star pattern matching |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103968833A CN103968833A (en) | 2014-08-06 |
| CN103968833B true CN103968833B (en) | 2017-01-04 |
Family
ID=51238626
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410182261.XA Expired - Fee Related CN103968833B (en) | 2014-04-30 | 2014-04-30 | A kind of method choosing observation triangle before star pattern matching |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103968833B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105928509B (en) * | 2016-06-22 | 2018-10-30 | 中国人民解放军63680部队 | Triangle recognition methods based on maximum interior angle hash function |
| CN110501016B (en) * | 2019-08-21 | 2021-04-23 | 中国科学院软件研究所 | Method and device for measuring satellite attitude |
| CN111024063B (en) * | 2019-12-23 | 2023-07-18 | 上海交通大学 | Star map recognition algorithm based on star point re-extraction under large maneuvering condition |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6324475B1 (en) * | 1997-02-13 | 2001-11-27 | Centre National D'etudes Spatiales | Method for identifying in a pre-established catalogue stars detected by a star sensor |
| CN101441082A (en) * | 2008-12-22 | 2009-05-27 | 中国科学院长春光学精密机械与物理研究所 | Rapid triangle star map recognition method |
| CN102313549A (en) * | 2011-07-28 | 2012-01-11 | 北京航空航天大学 | Identification method for triangular star atlas based on characteristic of inertia ratio |
-
2014
- 2014-04-30 CN CN201410182261.XA patent/CN103968833B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6324475B1 (en) * | 1997-02-13 | 2001-11-27 | Centre National D'etudes Spatiales | Method for identifying in a pre-established catalogue stars detected by a star sensor |
| CN101441082A (en) * | 2008-12-22 | 2009-05-27 | 中国科学院长春光学精密机械与物理研究所 | Rapid triangle star map recognition method |
| CN102313549A (en) * | 2011-07-28 | 2012-01-11 | 北京航空航天大学 | Identification method for triangular star atlas based on characteristic of inertia ratio |
Non-Patent Citations (2)
| Title |
|---|
| 一种改进的全天自主三角形星图识别算法;李立宏等;《光学技术》;20000731;第26卷(第4期);第372-374页 * |
| 星图识别;张广军;《北京:国防工业出版社》;20110831;第70-80页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103968833A (en) | 2014-08-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103913131B (en) | Free curve method vector measurement method based on binocular vision | |
| CN103295239B (en) | A kind of autoegistration method of the laser point cloud data based on datum plane image | |
| CN107564069A (en) | The determination method, apparatus and computer-readable recording medium of calibrating parameters | |
| CN106553195A (en) | Object 6DOF localization method and system during industrial robot crawl | |
| CN103868510A (en) | Rapid autonomous all-sky map fixed star identification method | |
| CN105303615A (en) | Combination method of two-dimensional stitching and three-dimensional surface reconstruction of image | |
| CN104537709A (en) | Real-time three-dimensional reconstruction key frame determination method based on position and orientation changes | |
| CN103968833B (en) | A kind of method choosing observation triangle before star pattern matching | |
| CN103198523A (en) | Three-dimensional non-rigid body reconstruction method and system based on multiple depth maps | |
| CN103954280B (en) | A kind of quickly and high robust autonomous fixed star recognition methods | |
| CN103453905B (en) | A kind of method for recognising star map based on plane triangle principal component analysis (PCA) | |
| CN106251326B (en) | A kind of building occlusion detection and occlusion area compensation method using ghost picture | |
| CN105913435A (en) | Multidimensional remote sensing image matching method and multidirectional remote sensing image matching system suitable for large area | |
| CN106295512A (en) | Many correction line indoor vision data base construction method based on mark and indoor orientation method | |
| CN107102332A (en) | The three-dimensional earth's surface deformation monitoring methods of InSAR based on variance components estimate and strees strain model | |
| CN104949677A (en) | Real-time orbit determination method for drift scanning geosynchronous satellite | |
| CN107341824A (en) | A kind of comprehensive evaluation index generation method of image registration | |
| CN102840860A (en) | Hybrid particle swarm algorithm-based star graph recognition method | |
| CN109443200A (en) | A kind of mapping method and device of overall Vision coordinate system and mechanical arm coordinate system | |
| CN104200512A (en) | Multiple-light source rendering method based on virtual spherical light sources | |
| IL299069B1 (en) | 3d structure inspection or metrology using deep learning | |
| CN105046691A (en) | Method for camera self-calibration based on orthogonal vanishing points | |
| CN105488535B (en) | A kind of three-dimensional point cloud matching process | |
| Li et al. | Scale-aware monocular SLAM based on convolutional neural network | |
| CN103940432A (en) | Posture determination method of star sensor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170104 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |