CN112729277B - Star sensor star map identification method based on dynamic included angle matching - Google Patents
Star sensor star map identification method based on dynamic included angle matching Download PDFInfo
- Publication number
- CN112729277B CN112729277B CN202011590104.4A CN202011590104A CN112729277B CN 112729277 B CN112729277 B CN 112729277B CN 202011590104 A CN202011590104 A CN 202011590104A CN 112729277 B CN112729277 B CN 112729277B
- Authority
- CN
- China
- Prior art keywords
- star
- matching
- angle
- navigation
- feature
- 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
Links
Images
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
Abstract
The invention discloses a star map identification method of a star sensor based on dynamic angle matching, which belongs to the field of navigation, guidance and control of spacecrafts and comprises the following steps: step 1: constructing a navigation star database, which comprises selecting navigation star, extracting star pattern characteristics, recording and storing; step 2: the dynamic included angle initial matching comprises constructing dynamic included angle characteristics of a reference star to be identified, searching candidate navigation stars in a navigation star library, and recording a matching group for completing initial matching; step 3: the matching degree calculation comprises the steps of defining the included angle and distance characteristics of a reference star and a navigation star, calculating the characteristic matching number, defining the matching degree score, selecting the maximum score of each candidate star, and obtaining Hou Xuanxing with the maximum score as the identification result. The method uses a plurality of groups of dynamic included angle characteristics as the initial edges of the subsequent matching degree calculation, overcomes the problem of relying on the nearest neighbor star as the initial star, changes the unique matching into a plurality of groups of matching, and improves the accuracy of identification and the robustness to noise and false star.
Description
Technical Field
The present invention relates generally to the field of spacecraft navigation, guidance and control, and more particularly to a star map identification method for a star sensor.
Background
The star sensor is a celestial body sensor in an astronomical navigation system and has the characteristics of small volume, high precision, strong autonomy and the like. The method is widely applied to the field of aerospace by observing stars in the space to calculate the gesture of the spacecraft. The star map recognition algorithm is used as a key technology of the star sensor, and the realization of a reliable and rapid star map recognition algorithm is always an important research subject.
Several star map recognition algorithms have been proposed and are briefly described below.
The triangle algorithm proposed by Liebe et al uses the angular distance between every two stars as a characteristic for identification, and is simple to realize but has certain redundancy.
Mortari et al propose pyramid algorithm, based on triangle, extend to four star points, further reduce redundancy in the identification process.
Zhang Anjun et al propose an improved triangle recognition algorithm employing angular distance matching to directly store the star diagonal, speeding up recognition.
Wei et al propose to convert the star map from a Cartesian coordinate system to a polar coordinate system for identification to achieve scale and rotational invariance in the identification algorithm.
Padgett et al propose a star pattern-like trellis algorithm that matches the star pattern trellis encoded as pattern features, but the algorithm performs poorly under interference such as star and position noise.
Na et al change the hard template matching in the original grid matching process into a cost function for measuring the difference between the observation feature and the star pattern feature, and use the relative star and the like of the star as weights, thereby improving the robustness of the position noise and the star and the like noise.
The Polestar algorithm is proposed by Silani et al, which uses one star as a reference star, calculates the angular distance between the star and the adjacent star as a characteristic to form a specific binary vector, and uses the ideas of a lookup table structure and a voting method for identification.
Zhang et al used a step-wise recognition algorithm with initial matching using radial features and subsequent matching using circumferential features, but the choice of the starting edge in the circumferential features was susceptible to noise as with the grid algorithm.
Wei et al propose to eliminate the effects of various noise by using the angle between adjacent star vectors as a dynamic circumferential feature.
Samibrhai et al use additive vectors with rotational invariance as features of the identification and select the star with the highest matching degree as the identification result by voting, but this method has the problem of being too dependent on the nearest neighbor as the starting star.
However, the accuracy and robustness of current star map recognition algorithms still have room and need for further improvement.
Disclosure of Invention
In order to solve the problem that the existing identification algorithm uses the nearest neighbor star as the initial star, and under the condition of large star point extraction error, the initial star is erroneously selected to easily cause matching failure, the invention provides a star sensor star map identification method based on dynamic angle matching, which changes the prior unique matching into multiple groups of matching, and improves the identification accuracy and the robustness of position noise and false star.
A star sensor star map identification method based on dynamic included angle matching comprises the following steps:
step 1: constructing navigation star database
(1) And selecting the star meeting the identification requirement from the SAO J2000 star list as a navigation star. Unstable varistors and indistinguishable double stars in the star table are removed, sensitive stars of a star sensor are combined, and 4956 stars are finally selected as navigation stars.
(2) And respectively taking the selected navigation satellites as the centers of the fields of view, and selecting the satellites meeting the angular distance limitation as adjacent satellites according to the size of the field of view of the satellite sensor to obtain the mode characteristics of N adjacent satellites for generating the navigation satellites.
(3) Using the distance r between the navigation star and the adjacent star i Included angle theta between adjacent satellites i As a star pattern feature of the navigation star. As shown in FIG. 1, a circle is made by taking a navigation star X as a circle center and taking half of FOV/2 of the size of a field of view as a radius, N star points in the circle are taken as adjacent stars, and the N star points are respectively marked as { S ] in a counterclockwise order 1 ，S 2 ，...，S N }. Connect each adjacent star { S } 1 ，S 2 ，...，S N Respectively obtaining the included angle { theta } between each adjacent star by using the formula (1) and the formula (2) 1 ，θ 2 ，...，θ N Distance { r between reference and adjacent satellites } 1 ，r 2 ，...，r N }
Wherein x is the sum of i And y i Respectively the coordinates of the ith adjacent star in the image coordinate system, N is the total number of all adjacent stars in the field of view, and x c And y c Is the coordinates of the navigation star in the image coordinate system.
The features are uniformly arranged in a counterclockwise manner, and the starting point S i The differences selected correspond to the feature vector v θ And v r Only the cyclic shift is performed. For example, for navigation star X, S 1 Two feature vectors v obtained as starting points θ ＝{θ 1 ，θ 2 ，...，θ N }，v r ＝{r 1 ，r 2 ，...，r N And use S 6 Two feature vectors v obtained as starting points θ ＝{θ 6 ，θ 7 ，...，θ N ，...，θ 5 }，v r ＝{r 6 ，r 7 ，...，r N ，...，r 5 Only one cyclic shift between. Thus, both the angle and distance features have rotational invariance as star pattern features.
Finally, the characteristic vector v of each navigation star is obtained θ ＝{θ 1 ，θ 2 ，...，θ N }，v r ＝{r 1 ，r 2 ，...，r N }。
(4) The navigation star pattern features are stored in tables 1 and 2, respectively, the first column of the table is the unique number of the navigation star and is used as the index of each navigation star, and the following N columns are the distance features and the included angle features of N adjacent stars corresponding to the current navigation star.
TABLE 1 navigation star repository angle characterization
Table 2 navigation star-bank distance features
Step 2: initial matching of dynamic included angles
For a given star map, a reference star to be identified is selected, and its surrounding neighbors construct star pattern features. The included angle characteristic { theta } between each adjacent star can be obtained 1 ，θ 2 ，...，θ N Distance feature between reference and neighbor { r } 1 ，r 2 ，...，r N }。
The present inventors are based on the following insight of the inventors. The inventor finds that the main reason why the star map recognition method in the prior art has lower precision is that the partial pattern recognition algorithm uses the nearest neighbor as the starting star, but this is not necessarily the best choice, but instead causes recognition precision problems, and the inventor finds that the N groups of corresponding included angles and distances obtained by combining { theta } with the reference star to be recognized k ，r k As dynamic included angle feature, sequentially searching navigation star library for navigation star containing the feature as Hou Xuanxing, which can significantly identify accuracy, and simultaneously, in order to improve the robustness of the identification algorithm to noise, setting threshold value in the searching process, namely, satisfying the navigation star of formula (3) id As Hou Xuanxing.
Wherein θ k And r k As a kth angle and distance feature of the reference star,
and d l For navigating star id Is the first included angle and distance feature. Epsilon angle And epsilon distance The angle match threshold and the distance match threshold, respectively.
In short, the invention solves the problem of relying on the nearest neighbor star as the starting star by using a plurality of groups of dynamic included angle features. Through the initial identification process, hou Xuanxing star meeting the initial matching requirement is screened from the navigation star table id And record the k-th set of features of the reference star and the navigation star id The first group of features of (a) is a pair of matching groups, providing a dynamic starting edge for the next step of computing the degree of matching.
Step 3: matching degree calculation
The included angles and distances between adjacent satellites defining the reference star R are characterized as ref= { (θ) 1 ，r 1 )，(θ 2 ，r 2 )，...，(θ m ，r m ) Hou Xuanxing star id Is characterized by the included angles and the distance between adjacent stars
Wherein m and n are respectively the generated reference star R and Hou Xuanxing star id And generating the number of adjacent stars of the features.
Through the initial identification process, the kth group dynamic included angle characteristic and Hou Xuanxing star of the reference star can be known id The first group of dynamic included angle features in the model is matched. Taking the dynamic included angle characteristic as a starting point of subsequent matching, calculating a reference star R to be identified and a candidate star selected id Number of feature matches num between match 。
First, the features ref and pat are combined using matching dynamic angle features id The cyclic shifts k and l times, respectively, result in shift_ref= { (α) 1 ，β 1 )，(α 2 ，β 2 )，...，(α m ，β m )}，shift_pat id ＝{(a 1 ，b 1 )，(a 2 ，b 2 )，...，(a n ，b n ) It is easy to know at this time shift_ref, shift_pat id In (a) and (b)The starting adjacent star features are already respectively { θ } for the above-mentioned complete matching k ，r k Sum of
Namely, satisfies the formula (4): />
With the characteristic shift_ref, shift_pat after cyclic shift id Based on the information of the included angles, the cumulative angle characteristics angle_ref= { ω are respectively constructed 1 ，ω 2 ，...，ω m Sum of angle_pat id ＝{μ 1 ，μ 2 ，...，μ n -wherein each element is defined as formula (5) and formula (6), respectively:
using angle_ref and angle_pat id And (5) performing matching degree calculation. Traversing angle_ref and angle_pat id Finding the close cumulative angle omega i Sum mu j As a matching candidate, i.e., the cumulative angle constraint should satisfy equation (7):
|ω i -μ j |≤ε angle (7)
wherein i and j are traversal angle_ref and angle_pat, respectively id Index of time, ε angle A threshold range for cumulative angle matching.
When omega i Sum mu j When the above equation is satisfied, only the angle information is used at this time, and further verification using distance features is required to reduce redundancy in matching. From omega i Sum mu j The matching result indicates that the distance corresponding to the current matching candidate is beta i And b j . And if the distance constraint condition is met, the current i and j can be considered to be matched. Namely, the following conditions are satisfied:
|β i -b j |≤ε distance (8)
in the feature matching number calculation process, i and j start from 1. If the characteristics corresponding to the current i and j are matched through the accumulated angle and the distance constraint, the reference star R and Hou Xuanxing star to be identified id Number of feature matches num between match Adding one, moving to the next position and continuing to compare; otherwise, judging the current omega i Sum mu i The size of (2) is determined next, if ω i ＜μ j I=i+1; if omega i ＞μ j J=j+1 and then the process of accumulating angle matching and distance verification continues to repeat.
Finally, when i is more than or equal to m or j is more than or equal to n, ending the characteristic matching number calculation process, and obtaining the characteristic matching number num at the moment match Representing the currently identified reference star R and Hou Xuanxing star id Element pairs successfully matched with adjacent star features.
Since the number of adjacent satellites around each navigation satellite has a certain difference, only the feature matching number num is used match It is not so reasonable to measure the degree of matching. Thus, the feature matching number num is used in combination with the number of neighbors around the navigation satellite match Number num of adjacent satellites to the navigation satellite neighbor As a criterion describing the degree of matching, i.e. the degree of matching score is defined as formula (9):
obviously, the matching score similarity score has a value ranging from 0,1, the larger this value being indicative of a higher matching between the reference star and the current candidate star.
Since multiple dynamic angles will correspond to the same Hou Xuanxing, multiple matching degrees are obtained, and this Hou Xuanxing selects the largest matching degree score as its own final score. Finally, hou Xuanxing with the largest matching degree score is selected as the identification result of the reference star to be identified.
The overall recognition flow is as in fig. 1.
The invention has at least the following beneficial effects: compared with the prior art, the star map recognition method of the star sensor based on dynamic angle matching provided by the invention uses the extracted multiple groups of dynamic angle characteristics to be respectively and sequentially used as the starting edges of subsequent matching degree calculation, overcomes the problem that the nearest neighbor star is too dependent as the starting star in the past, changes the past unique matching into multiple groups of matching, and improves the recognition accuracy and the robustness to noise and false star.
Drawings
The invention is further elucidated below in connection with the embodiments with reference to the drawings.
FIG. 1 shows a flow chart of an overall star map recognition method
FIG. 2 shows a feature extraction schematic diagram of star pattern feature extraction
FIG. 3 illustrates recognition rate performance under location noise for various recognition methods in one embodiment of the invention
FIG. 4 illustrates the recognition rate performance of the recognition methods under pseudo-star interference in one embodiment of the invention
FIG. 5 shows the recognition rate performance of the recognition methods under the noise of the star
FIG. 6 shows the actual star map matching results in one embodiment of the invention
Detailed Description
It should be noted that the components in the figures may be shown exaggerated for illustrative purposes and are not necessarily to scale. In the drawings, identical or functionally identical components are provided with the same reference numerals.
In the present invention, unless specifically indicated otherwise, "disposed on …", "disposed over …" and "disposed over …" do not preclude the presence of an intermediate therebetween. Furthermore, "disposed on or above" … merely indicates the relative positional relationship between the two components, but may also be converted to "disposed under or below" …, and vice versa, under certain circumstances, such as after reversing the product direction.
In the present invention, the embodiments are merely intended to illustrate the scheme of the present invention, and should not be construed as limiting.
In the present invention, the adjectives "a" and "an" do not exclude a scenario of a plurality of elements, unless specifically indicated.
It should also be noted herein that in embodiments of the present invention, only a portion of the components or assemblies may be shown for clarity and simplicity, but those of ordinary skill in the art will appreciate that the components or assemblies may be added as needed for a particular scenario under the teachings of the present invention. In addition, features of different embodiments of the invention may be combined with each other, unless otherwise specified. For example, a feature of the second embodiment may be substituted for a corresponding feature of the first embodiment, or may have the same or similar function, and the resulting embodiment would fall within the disclosure or scope of the disclosure.
It should also be noted herein that, within the scope of the present invention, the terms "identical", "equal" and the like do not mean that the two values are absolutely equal, but rather allow for some reasonable error, that is, the terms also encompass "substantially identical", "substantially equal". By analogy, in the present invention, the term "perpendicular", "parallel" and the like in the table direction also covers the meaning of "substantially perpendicular", "substantially parallel".
The numbers of the steps of the respective methods of the present invention are not limited to the order of execution of the steps of the methods. The method steps may be performed in a different order unless otherwise indicated.
The invention is further elucidated below in connection with the embodiments with reference to the drawings.
Example 1
Stars are selected from the SAO J2000 star list, and a simulated star map generated by MATLAB 2020 is used for testing the method provided by the invention. The field angle of a star sensor for generating the simulated star map is 12 degrees multiplied by 12 degrees, the image resolution is 1024 x 1024, the pixel size is 12 mu m, the focal length is 58.4563mm, and the sensitive star is 6.0Mv. In the generation process, uniformly traversing 0-360 degrees of red warp and-90 degrees of red weft at intervals of 2 degrees to finally obtain 16200 simulated star images uniformly covering the whole celestial sphere.
The modified triangle algorithm, radial-axial algorithm, grid algorithm, and pyramid algorithm are compared with the algorithm of the present invention. In the case of an ideal simulated star map without added noise, the recognition rates of the respective recognition algorithms are obtained as shown in table 3. As can be seen from the table, the recognition algorithm provided by the invention is superior to other algorithms in recognition rate, and the recognition accuracy rate is 99.80%. Of the 16200 simulated star maps, only 32 star maps were not correctly identified. Statistics show that the number of star points in the star maps is less than 3, so that the algorithm cannot work normally.
Table 3 recognition accuracy performance of each recognition algorithm
The star sensor is disturbed by the satellite motion, vibration and other external factors, so that the imaged star point can shift, and Gaussian noise is added to the star point position in the simulated star map to simulate the situation. FIG. 3 shows the recognition accuracy of each recognition algorithm after adding position noise with a mean of 0 and a variance of 0-2 pixels. As can be seen from fig. 3, the improved triangular algorithm, radial-axial algorithm, and pyramid algorithm recognition rate rapidly decreased with increasing position noise, and the grid algorithm recognition rate decreased to 91.89%. Compared with the algorithm, when the position noise variance is 2 pixels, the recognition rate is maintained at 98.30%, so that the algorithm has stronger robustness to the position noise.
The star sensor is interfered by dust, space debris and the like to generate pseudo star points in the shooting process, and the pseudo star points are randomly placed in the simulated star map, so that the identification performance of each star map identification algorithm under the interference of the pseudo star is verified. FIG. 4 shows the recognition accuracy of each recognition algorithm after adding 0-5 pseudo-star points to the simulated star map. As can be seen from fig. 4, in the case that 5 pseudolites exist, the method provided by the invention still has the recognition accuracy of 97.83%. The method provided by the invention is insensitive to the pseudo star points, and has stronger robustness compared with other algorithms.
In the imaging process of the star sensor, a plurality of low-brightness stars can not be captured due to various interferences, so that the star points are lost, and Gaussian noise is added to the stars and the like of the stars in the simulated star map, so that the situation that some stars disappear due to the change of the stars and the like is simulated. FIG. 5 shows the recognition accuracy of each star map recognition method after Gaussian noise with the mean value of 0 and the variance of 0-1 Mv is added to the star of each star, so as to verify the robustness of each star map recognition method to the noise interference of the star. As can be seen from fig. 5, each identification method is insensitive to changes in star etc. The method is mainly used for identifying the adjacent satellites, the probability of too few adjacent satellites in a scene is increased along with the gradual increase of noise such as the satellites, and finally, the identification accuracy of the method is slightly lower than that of a pyramid algorithm. However, when the star noise deviation is 1Mv, the recognition accuracy of 97.45% is still maintained. Therefore, the method of the invention has certain robustness to star noise.
Example two
Fig. 6 is a true star map of the current model satellite download, which is used to further verify the method of the present invention. The star sensor has image resolution of 1536 x 1536, focal length of 23.905mm, field size of 20 x 20, principal point coordinates of [759, 768]. Through a star point extraction algorithm, 24 star points are extracted from the star map. The identification algorithm is executed by utilizing the star points, and the actual result is used for verification, so that the fact that 21 star points are accurately identified is finally determined, the requirement of star sensor pose determination is met, and the usability of the method is further verified.
While certain embodiments of the present invention have been described herein, those skilled in the art will appreciate that these embodiments are shown by way of example only. Numerous variations, substitutions and modifications will occur to those skilled in the art in light of the present teachings without departing from the scope of the invention. The appended claims are intended to define the scope of the invention and to cover such methods and structures within the scope of these claims themselves and their equivalents.
Claims (7)
1. A star sensor star map identification method based on dynamic angle matching is characterized by comprising the following steps:
the navigation star bank is constructed, which comprises the following steps:
selecting a star meeting the identification requirement from the star table as a navigation star; extracting features of the navigation star to obtain a feature vector of the navigation star, wherein the feature vector is used as a star mode feature of the navigation star and comprises a distance r between the navigation star and an adjacent star i Included angle theta between adjacent satellites i The method comprises the steps of carrying out a first treatment on the surface of the And recording and storing the mode characteristics of the navigation star, namely the distance characteristics and the included angle characteristics of N adjacent stars corresponding to the navigation star, and the characteristic vector v θ ＝{θ 1 ，θ 2 ，...，θ N }，v r ＝{r 1 ，r 2 ，...，r N }；
Performing initial matching of dynamic included angles, including the following steps:
selecting a reference star to be identified and adjacent stars around the reference star to construct a star pattern feature to obtain an included angle feature { theta } between each adjacent star 1 ，θ 2 ，...，θ N Distance feature between reference and neighbor { r } 1 ，r 2 ，...，r N To be combined { θ } k ，r k -as a dynamic angle feature; searching navigation star containing the dynamic included angle characteristic in navigation star library id As Hou Xuanxing; recording the k-th group characteristic of the reference star and the navigation star meeting the initial matching requirement id Is characterized by a pair of matched sets;
performing a matching degree calculation, comprising the steps of:
the included angles and distances between adjacent satellites defining the reference star R are characterized as ref= { (θ) 1 ，r 1 )，(θ 2 ，r 2 )，...，(θ m ，r m ) Hou Xuanxing star id Is characterized by the included angles and the distance between adjacent stars
Wherein m and n are respectively the generated reference star R and Hou Xuanxing star id Generating the number of adjacent stars of the features; taking a matching group meeting the initial matching requirement as a starting point of subsequent matching, and calculating a reference star R to be identified and a candidate star selected id Number of feature matches num between match The method comprises the steps of carrying out a first treatment on the surface of the Using a feature matching number num match Number num of adjacent stars to candidate star neighbor As a criterion describing the degree of matching, to obtain a degree of matching score according to the following formula: />
Wherein the matching score similarity score has a value in the range of 0,1, wherein a greater value indicates a higher matching between the reference star and the current candidate star; selecting the largest matching degree score from the plurality of matching degrees obtained by Hou Xuanxing as the final score of the user; and Hou Xuanxing with the largest final matching degree score is selected as the identification result of the reference star to be identified.
2. The method of claim 1, wherein selecting the star from the star list as the navigation star that meets the identification requirement comprises the steps of:
unstable varistors and indistinguishable double stars are removed from the SAO J2000 star list, and 4956 stars finally selected by combining sensitive stars of a star sensor and the like are combined.
3. The method according to claim 1 or claim 2, wherein the feature extraction of the navigation star comprises the steps of:
taking the navigation star X as the center of a circle, taking half of the FOV/2 of the field of view as the radius to make a circle, taking N star points in the circle as adjacent stars, and respectively marking the N star points as { S } according to the anticlockwise sequence 1 ，S 2 ，...，S N -a }; connect each adjacent star { S } 1 ，S 2 ，...，S N And navigation star X, and respectively obtaining each adjacent by using the following formulaIncluded angle { θ } between adjacent stars 1 ，θ 2 ，...，θ N Distance { r between reference and adjacent satellites } 1 ，r 2 ，...，r N }：
Wherein x is the sum of i And y i Respectively the coordinates of the ith adjacent star in the image coordinate system, N is the total number of all adjacent stars in the field of view, and x c And y c Is the coordinates of the navigation star in the image coordinate system.
4. The method of claim 1, wherein said performing a dynamic angle initial match further comprises the steps of:
setting a threshold value, the navigation star will satisfy the following formula id As Hou Xuanxing:
wherein θ k And r k As a kth angle and distance feature of the reference star,
and d l For navigating star id Is the first included angle, distance characteristic, epsilon angle And epsilon distance The angle match threshold and the distance match threshold, respectively.
5. The method according to claim 1, wherein the calculation of the reference star R to be identified and the candidate star selected id Number of feature matches num between match Comprises the following steps:
features (e.g. a character)ref and pat id The cyclic shifts k and l times, respectively, result in shift_ref= { (α) 1 ，β 1 )，(α 2 ，β 2 )，...，(α m ，β m )}，shift_pat id ＝{(a 1 ，b 1 )，(a 2 ，b 2 )，...，(a n ，b n ) -a }; construction of the cumulative angle feature angle_ref= { ω 1 ，ω 2 ，...，ω m Sum of angle_pat id ＝{μ 1 ，μ 2 ，...，μ n -wherein the elements are defined according to the following formula:
using cumulative angle features angle ref and angle pat id Calculating the matching degree to obtain the matching candidate calculated angle omega i Sum mu j Where i and j are traversal angle_ref and angle_pat, respectively id Index of time; using the above-mentioned matching candidate ω i Sum mu j Corresponding distance beta i And b j Performing distance feature verification, wherein i and j start from 1, and if the features corresponding to the current i and j are matched through the accumulated angle and the distance constraint, referencing the star R to be identified with Hou Xuanxing star id Number of feature matches num between match Adding one, moving to the next position and continuing to compare; otherwise, judging the current omega i Sum mu j The size of (2) is determined next, if ω i ＜μ j I=i+1; if omega i ＞μ j J=j+1, and then continuing to repeat the process of cumulative angle matching and distance verification; when i is more than or equal to m or j is more than or equal to n, finishing the characteristic matching number calculation process to obtain a reference star R and Hou Xuanxing star representing the current to be identified id Feature matching number num of element logarithm with successful adjacent star feature matching match 。
6. The method of claim 5, wherein the build accumulation angle feature angle_ref= { ω 1 ，ω 2 ，...，ω m Sum of angle_pat id ＝{μ 1 ，μ 2 ，...，μ n The method comprises the following steps:
the constraint accumulation angle satisfies the following formula: omega i -μ j |≤ε angle
Wherein ε is angle A threshold range for cumulative angle matching.
7. The method according to one of claims 5 and 6, wherein said matching candidate ω is used i Sum mu j Corresponding distance beta i And b j The distance characteristic verification comprises the following steps:
the constraint corresponding distance feature satisfies the following formula: beta (beta) i -b j |≤ε distance
Wherein ε is distance A threshold value for verification of the corresponding distance feature.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011590104.4A CN112729277B (en) | 2020-12-29 | 2020-12-29 | Star sensor star map identification method based on dynamic included angle matching |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011590104.4A CN112729277B (en) | 2020-12-29 | 2020-12-29 | Star sensor star map identification method based on dynamic included angle matching |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112729277A CN112729277A (en) | 2021-04-30 |
| CN112729277B true CN112729277B (en) | 2023-05-26 |
Family
ID=75607409
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011590104.4A Active CN112729277B (en) | 2020-12-29 | 2020-12-29 | Star sensor star map identification method based on dynamic included angle matching |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112729277B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113483751B (en) * | 2021-06-29 | 2022-07-05 | 北京航空航天大学 | Radial triangle mapping matrix-based star map identification method |
| CN114543789A (en) * | 2021-12-27 | 2022-05-27 | 浙江众星志连科技有限责任公司 | Star map identification method and system based on one-dimensional convolutional neural network |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5935195A (en) * | 1997-03-27 | 1999-08-10 | Matra Marconi Space Uk | Autonomous star identification |
| CN101004349A (en) * | 2007-01-22 | 2007-07-25 | 北京航空航天大学 | Method for recognising not gauged star map |
| CN103148852A (en) * | 2013-03-01 | 2013-06-12 | 国家测绘地理信息局卫星测绘应用中心 | Star pattern recognition method based on directed loop |
| CN109813303A (en) * | 2019-03-08 | 2019-05-28 | 北京航空航天大学 | A kind of method for recognising star map for not depending on calibrating parameters based on the ballot of angle mould formula cluster |
| CN110715656A (en) * | 2019-11-01 | 2020-01-21 | 北京航天自动控制研究所 | Autonomous star map identification method based on dimensionality reduction star catalogue |
| CN111024063A (en) * | 2019-12-23 | 2020-04-17 | 上海交通大学 | Star map recognition algorithm under large-mobility condition based on star point re-extraction |
| CN111174776A (en) * | 2020-01-22 | 2020-05-19 | 中国科学院西安光学精密机械研究所 | Radial and dynamic circumferential mode-based star map identification method |
-
2020
- 2020-12-29 CN CN202011590104.4A patent/CN112729277B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5935195A (en) * | 1997-03-27 | 1999-08-10 | Matra Marconi Space Uk | Autonomous star identification |
| CN101004349A (en) * | 2007-01-22 | 2007-07-25 | 北京航空航天大学 | Method for recognising not gauged star map |
| CN103148852A (en) * | 2013-03-01 | 2013-06-12 | 国家测绘地理信息局卫星测绘应用中心 | Star pattern recognition method based on directed loop |
| CN109813303A (en) * | 2019-03-08 | 2019-05-28 | 北京航空航天大学 | A kind of method for recognising star map for not depending on calibrating parameters based on the ballot of angle mould formula cluster |
| CN110715656A (en) * | 2019-11-01 | 2020-01-21 | 北京航天自动控制研究所 | Autonomous star map identification method based on dimensionality reduction star catalogue |
| CN111024063A (en) * | 2019-12-23 | 2020-04-17 | 上海交通大学 | Star map recognition algorithm under large-mobility condition based on star point re-extraction |
| CN111174776A (en) * | 2020-01-22 | 2020-05-19 | 中国科学院西安光学精密机械研究所 | Radial and dynamic circumferential mode-based star map identification method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112729277A (en) | 2021-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | Full-sky autonomous star identification based on radial and cyclic features of star pattern | |
| CN112729277B (en) | Star sensor star map identification method based on dynamic included angle matching | |
| CN110132263B (en) | Star map identification method based on representation learning | |
| CN103136525B (en) | A kind of special-shaped Extended target high-precision locating method utilizing Generalized Hough Transform | |
| Zhao et al. | Star identification algorithm based on K–L transformation and star walk formation | |
| CN108917753B (en) | Aircraft position determination method based on motion recovery structure | |
| Wei et al. | A star identification algorithm based on radial and dynamic cyclic features of star pattern | |
| Samirbhai et al. | A hamming distance and spearman correlation based star identification algorithm | |
| CN103148852A (en) | Star pattern recognition method based on directed loop | |
| Dor et al. | Visual SLAM for asteroid relative navigation | |
| CN109724591B (en) | Star sensor false star screening method based on angular distance screening | |
| Mehta et al. | A robust star identification algorithm with star shortlisting | |
| Zhao et al. | Real-time star identification using synthetic radial pattern and its hardware implementation | |
| CN113532438B (en) | Improved ICCP terrain matching method under large initial positioning error | |
| Du et al. | Star identification based on radial triangle mapping Matrix | |
| Yang et al. | Robust and efficient star identification algorithm based on 1-D convolutional neural network | |
| CN110942473A (en) | Moving target tracking detection method based on characteristic point gridding matching | |
| CN104833359B (en) | A kind of star chart mode identification method based on discrete Markov characteristic sequence model | |
| CN106767841A (en) | Vision navigation method based on self adaptation volume Kalman filtering and single-point random sampling | |
| CN111174776B (en) | Radial and dynamic circumferential mode-based star map identification method | |
| CN116659485A (en) | Quick robust global star map identification method | |
| Wei et al. | Star identification algorithm based on log-polar transform | |
| Yoon | Autonomous star identification using pattern code | |
| CN112287906A (en) | Template matching tracking method and system based on depth feature fusion | |
| CN115290100A (en) | Partitioned fast star map identification method based on attitude information |
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 |