CN113566815A

CN113566815A - Construction method and device of star map recognition navigation triangle library

Info

Publication number: CN113566815A
Application number: CN202110935560.6A
Authority: CN
Inventors: 江洁; 马岩; 张广军
Original assignee: Beihang University
Current assignee: Beihang University
Priority date: 2021-08-16
Filing date: 2021-08-16
Publication date: 2021-10-29
Anticipated expiration: 2041-08-16
Also published as: CN113566815B

Abstract

The invention discloses a construction method and a device of a star map identification navigation triangle library. The method comprises the following steps: acquiring the size, shape and number of triangles in the field of view in different directions; constructing a triangular density distribution model according to the size, shape and quantity of the triangles; generating a triangular utilization rate model according to the triangular density distribution model and the fixed star and the star corresponding to the triangle; and eliminating the triangles with low utilization rate in the triangle utilization rate model to obtain a triangle library. The invention screens the triangles in the celestial sphere, improves the use efficiency of the storage space, has high matching degree, enhances the recognition rate and improves the search efficiency.

Description

Construction method and device of star map recognition navigation triangle library

Technical Field

The invention relates to the technical field of star sensors, in particular to a method and a device for constructing a star map identification navigation triangle library.

Background

Along with the continuous development of intelligent science and technology, more and more intelligent equipment has been used among people's life, work, study, uses intelligent science and technology means, has improved the quality of people's life, has increased the efficiency of people's study and work.

The star sensor is a high-precision attitude measuring instrument. The instrument realizes measurement and calculation of the posture by observing the fixed star through the camera. The method has the advantages of high measurement precision, no accumulation of measurement errors along with time and the like, and is widely applied to the aerospace fields of satellites, airplanes, missiles and the like. In order to realize high-precision measurement of the attitude, the star sensor not only needs a high-performance lens and an image sensor, but also cannot leave a high-performance star map recognition algorithm and a navigation star library matched with the high-performance star map recognition algorithm. In recent years, star map recognition algorithms based on triangle patterns have been developed, such as Hash table star map recognition method (Gangyi Wang, Jian Li, and Xinguo Wei. Star identification based on hash map. IEEE Sensors journal,18(4): 1591-. Different from a triangle algorithm which searches for the star diagonal distances of three sides respectively, the method takes the length of three sides of a triangle as a characteristic vector and directly searches for the whole triangle. The method has higher search efficiency and robustness than the triangular algorithm as long as the proper search algorithm such as Hash is adopted. However, the three-side length of the triangle is very many, and it is obviously not feasible to put all the triangles in the celestial sphere into a warehouse. Therefore, the triangle in the celestial sphere is screened. If the number of triangles stored in the library is too small, the triangles in the library are insufficient, and the recognition rate is reduced. If the number of triangles included in the library is too large, the requirement on storage space is increased, the search efficiency is reduced, more probability of mismatching is caused, and the recognition rate is reduced. Even if the number of triangles in the library is moderate, if the selection is not proper, the situation that partial triangles are too dark and cannot be observed in the library or partial triangles are difficult to appear in a view field due to the problems of the shapes and the sizes of the triangles can also occur. This situation can cause a large number of invalid navigation triangles to be stored in the library, which are rarely used or even never detected, resulting in wasted storage space.

At present, screening methods for navigation star libraries mainly focus on screening of star point libraries. The traditional navigational star map method can be mainly divided into star class (Yan Ma, Jie Jiang, and Guangjun Zhang. Stellar Instrument timing in Infinite-Dimensional space. IEEE SENSORS JOURNAL, 20(3):1422 and 1432,2020.) and geometric class (Xinlu Li, Jinhua Yang, Liu Zhang, Shuang Li, and Guang Jin. A new sApplied selection OF the guide stand category for a stand sensor, JOURNAL OF NAVIGATION,67(6): 984-. No special research is currently available on the screening method of the triangle library. Wang Congyi et al used a spiral uniform distribution method in their paper to construct a navigation triangle library (Gangyi Wang, Jian Li, and Xinguo Wei. Star identification based on hash map. IEEE Sensorsjournal,18(4): 1591-. They sample the whole celestial sphere to generate 10 ten thousand spiral uniformly distributed points, and generate a simulated star map for each point, and take n from each star map_sAnd combining the brightest star points to obtain a plurality of triangles. The method uses a parameter n under the conditions of 25-degree field of view and 5.5 equi-stars_sThe 140K triangles generated at 10 occupy about 2MB of memory and can be applied to actual star sensor devices. And for a 5.4 ° field of view, 7 equi-star condition, the parameter n is set using this method_sThe number of the 10 generated triangles is about 1209K, which cannot meet the storage space requirement of practical application. If the parameter n of the brightest star point in each star map is adjusted_sThe number of triangles is reduced to 224K at 5, but the recognition rate of the triangles is also reduced under various noise interferences, and the overall performance cannot meet the measurement requirement.

In view of the above problems, no effective solution has been proposed.

Disclosure of Invention

The embodiment of the invention provides a construction method and a device of a star map recognition navigation triangle library, which are used for at least screening triangles in celestial spheres in a storage space, improving the use efficiency of the storage space, having high matching degree, enhancing the recognition rate and improving the search efficiency. If the number of triangles stored in the library is too small, the triangles in the library are insufficient, and the recognition rate is reduced. If the number of triangles in the storage is too large, the requirement on the storage space is increased, the search efficiency is reduced, more probability of mismatching is caused, and the recognition rate is reduced. Even if the number of triangles in the library is moderate, if the selection is not proper, the situation that partial triangles exist in the library, the star points are too dark and cannot be observed, or partial triangles are difficult to appear in a view field due to the problems of the shapes and the sizes of the triangles can also occur. This situation can cause a technical problem in that a large number of invalid navigation triangles are stored in the library, and these triangles are rarely used or even never detected.

According to an aspect of the embodiments of the present invention, there is provided a method for constructing a star atlas identification navigation triangle library, including: acquiring the size, shape and number of triangles in the field of view in different directions; constructing a triangular density distribution model according to the size, shape and quantity of the triangles; generating a triangular utilization rate model according to the triangular density distribution model and the fixed star and the like corresponding to the triangle; and eliminating the triangles with low utilization rate in the triangle utilization rate model to obtain a triangle library.

Optionally, the constructing a triangular density distribution model according to the size, shape and number of the triangles includes: obtaining effective area D of star point_iIt is defined as follows:

wherein i is the star number, s_iThe vector of the star vector is used as the vector of the star,

is a set of directional vectors, θ, in three-dimensional space_fovIs the angle of view; generating a fixed star density distribution model according to the effective star point areas:

in the formula

M is lower than the limit star_lmtCard (·) is an operator that takes the number of elements in the set; generating a triangular effective area E according to the star density distribution model_i,j,k：

Wherein i, j, k are asterisks constituting a triangle, c_τ(i, j, k) and r_eq(i, j, k) is the sum of the centers of the circles of the trianglesA radius; generating a triangular density distribution model according to the triangular effective area: rho_t(v)＝Card({(i,j,k)∈ Δ_g|v∈E_i,j,k) } in which, Δ_gIs a set of triangles.

Optionally, the generating a triangle utilization model by the triangle density distribution model and the star and the like corresponding to the triangle includes: calculating the occurrence probability P of the triangle according to the size and the shape of the triangle and the star point information_tProbability of detection of a sum Star point P_s(ii) a According to the occurrence probability P of the triangle_tProbability of detection of a sum Star point P_sAnd generating the triangular utilization rate model, wherein the triangular utilization rate model is described as follows by a formula:

,

in the formula of U_i,j,kIs the utilization ratio of a triangle consisting of three stars with the asterisks i, j, k, r_eq(i, j, k) is the effective area radius of the triangle, (. phi.). cndot._lmt，

δ_mThe estimation errors of the extreme star and the instrument star and the star of the p-th star point are respectively.

Optionally, the removing the low-utilization triangles in the triangle utilization model includes: calculating the utilization rates of all triangles in the triangle utilization rate model, and sequencing from low to high; and sequentially judging whether the density in each triangular effective area is greater than a triangular density threshold value, if so, deleting, and otherwise, reserving.

According to another aspect of the embodiments of the present invention, there is also provided a star atlas identification navigation triangle library construction apparatus, including: the acquisition module is used for acquiring the size, shape and number of the triangles in the fields of view in different directions; the density module is used for constructing a triangular density distribution model according to the size, shape and quantity of the triangles; the utilization rate module is used for generating a triangular utilization rate model according to the triangular density distribution model and the fixed stars and the like corresponding to the triangles; and the elimination module is used for eliminating the triangles with low utilization rate in the triangle utilization rate model to obtain a triangle library.

Optionally, the density module includes: an acquisition unit for acquiring a star point effective region D_iWhich is defined by the formula:

wherein i is an asterisk, s_iThe vector of the star vector is used as the vector of the star,

is a set of directional vectors, θ, in three-dimensional space_fovIs the angle of view; the generating unit is used for generating a star density distribution model according to the star effective area:

in the formula

M is lower than the limit star_lmtCard (·) is an operator that takes the number of elements in the set; a generating unit, further used for generating a triangular effective area E according to the constant star density distribution model_i,j,k：

Wherein i, j, k are asterisks constituting a triangle, c_τ(i, j, k) and r_eq(i, j, k) is the circle center and radius of the excircle of the triangle; the generating unit is further used for generating a triangular density distribution model according to the triangular effective area: rho_t(v)＝Card({(i,j,k)∈Δ_g|v∈E_i,j,k) } in which, Δ_gIs a set of triangles.

Optionally, the utilization module includes: a calculating unit for calculating the occurrence probability P of the triangle according to the size and shape of the triangle and the star point information_tProbability of detection of a sum Star point P_s(ii) a A generating unit for generating the probability P of the triangle_tAnd star pointIs detected with probability P_sGenerating the triangular utilization rate model, and describing the triangular utilization rate model as follows by using a formula:

in the formula of U_i,j,kIs the utilization ratio of a triangle consisting of three stars with the asterisks i, j, k, r_eq(i, j, k) is the effective area radius of the triangle, (. phi.) is the cumulative function of the normal distribution of positive and negative phases, m_lmt，

δ_mThe estimation errors of the limit star and the instrument star and the star of the p-th star point are respectively.

Optionally, the eliminating module includes: the sorting unit is used for calculating the utilization rates of all triangles in the triangle utilization rate model and sorting the triangles from low to high; and the judging unit is used for sequentially judging whether the density in the effective area of each triangle is greater than the triangle density threshold value, if so, deleting the density, and otherwise, keeping the density.

According to another aspect of the embodiment of the invention, a nonvolatile storage medium is further provided, and the nonvolatile storage medium includes a stored program, wherein the program controls a device in which the nonvolatile storage medium is located to execute a star map identification navigation triangle library construction method when running.

According to another aspect of the embodiments of the present invention, there is also provided an electronic device, including a processor and a memory; the memory is stored with computer readable instructions, and the processor is used for executing the computer readable instructions, wherein the computer readable instructions execute a star map identification navigation triangle library construction method when running.

In the embodiment of the invention, the size, shape and number of triangles in the field of view in different directions are obtained; constructing a triangular density distribution model according to the size, shape and quantity of the triangles; generating a triangular utilization rate model according to the triangular density distribution model and the fixed star and the star corresponding to the triangle; the method for eliminating the triangles with low utilization rate in the triangle utilization rate model to obtain the triangle library solves the problem of screening the triangles in the celestial sphere of the storage space, improves the service efficiency of the storage space, has high matching degree, enhances the recognition rate and improves the search efficiency. If the number of triangles stored in the library is too small, the triangles in the library are insufficient, and the recognition rate is reduced. If the number of the triangles in the storage is too large, the requirement on the storage space is increased, the searching efficiency is reduced, more probability of mismatching is caused, and the recognition rate is reduced. Even if the number of triangles in the library is moderate, if the selection is not proper, the situation that partial triangles exist in the library, the star points are too dark and cannot be observed, or the partial triangles are difficult to appear in a view field due to the problems of the shapes and the sizes of the triangles can also occur. This situation can cause a technical problem in that a large number of invalid navigation triangles are stored in the library, and these triangles are rarely used or even never detected. The invention screens the triangles in the celestial sphere of the storage space, improves the use efficiency of the storage space, has high matching degree, enhances the recognition rate and improves the search efficiency.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention and not to limit the invention. In the drawings:

FIG. 1 is a schematic diagram of star density calculation according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of a triangle outer circle definition according to an embodiment of the invention;

FIG. 3 is a schematic diagram of a triangular effective area according to an embodiment of the invention;

FIG. 4 is a star density map according to an embodiment of the present invention;

FIG. 5 is a graph of a full celestial sphere triangle density distribution according to embodiments of the present invention;

FIG. 6 is an example of a screening algorithm according to an embodiment of the present invention;

FIG. 7 is a local density distribution diagram of a navigation triangle library according to an embodiment of the present invention;

FIG. 8 is a histogram of the navigation triangle density distribution in different directions in a celestial sphere in accordance with an embodiment of the present invention;

FIG. 9 is a histogram of triangle utilization distribution in a library according to an embodiment of the present invention;

FIG. 10 is a graph illustrating a comparison of recognition rate with the number of triangles in a triangle library under the condition of 1.0 star-like noise according to an embodiment of the present invention;

FIG. 11 is a graph comparing the recognition rate with the number of triangles in the triangle library under the condition of 1.0 position noise according to the embodiment of the present invention;

FIG. 12 is a flowchart of operations according to an embodiment of the present invention;

FIG. 13 is a flowchart of a method for constructing a star atlas identification navigation triangle library according to an embodiment of the present invention.

Detailed Description

In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

In accordance with an embodiment of the present invention, there is provided a method embodiment of a method for constructing a star atlas identification navigation triangle base, it is noted that the steps illustrated in the flowchart of the figure may be performed in a computer system such as a set of computer executable instructions, and that while a logical order is illustrated in the flowchart, in some cases the steps illustrated or described may be performed in an order different than here.

Example one

Fig. 13 is a flowchart of a method for constructing a star atlas identification navigation triangle library according to an embodiment of the present invention, and as shown in fig. 13, the method includes the following steps:

step S1302, the size, shape and number of triangles in the field of view in different directions are obtained.

And step S1304, constructing a triangular density distribution model according to the size, shape and quantity of the triangles.

And step S1306, generating a triangular utilization rate model according to the triangular density distribution model and the star and the like corresponding to the triangle.

Step S1308, removing the low-utilization triangles in the triangle utilization model to obtain a triangle library.

in the formula

Wherein i, j, k are asterisks constituting a triangle, c_τ(i, j, k) and r_eq(i, j, k) is the circle center and the radius of the excircle of the triangle; generating a triangular density distribution model according to the triangular effective area: rho_t(v)＝Card({(i,j,k)∈ Δ_g|v∈E_i,j,k) } in which, Δ_gIs a set of triangles.

Optionally, the generating a triangular utilization model according to the triangular density distribution model and the star corresponding to the triangle includes: calculating the occurrence probability P of the triangle according to the size and the shape of the triangle and the star point information_tProbability of detection of a sum Star point P_s(ii) a According to the occurrence probability P of the triangle_tProbability of detection of a sum Star point P_sGenerating the triangular utilization rate model, and describing the triangular utilization rate model by using a formula as follows:

in the formula of U_i,j,kIs the utilization ratio of a triangle formed by three stars with the star numbers i, j, k, r_eq(i, j, k) is the effective area radius of the triangle, (. phi.) is the cumulative function of the standard regular distribution, m_lmt，

As shown in fig. 7, the implementation process of the embodiment of the present invention is as follows:

1. establishing an all-celestial sphere star density and triangle density distribution model:

the star sensor typically operates near the earth, while other star points are quite distant from the earth. Therefore, the star point can be considered to be on the infinite plane in the star-sensitive camera system no matter where the star sensor is located near the earth. Therefore, in the J2000.0 celestial coordinate system, the position of any star point can be represented by a three-dimensional direction vector:

wherein

Is a unit sphere in three-dimensional space:

in the visual band, the vast majority of the brighter stars in the celestial sphere are recorded in the ebavalley table. The star table can be regarded as a collection of star point labels

If the limit star of the star sensor is assumed to be m_lmtThen can be according to the limit star, etc

Get the subset of visible stars

Since the stars are mostly concentrated on the silvery surface, the stars in the celestial sphere are not uniformly distributed over the entire celestial sphere, but are more dense closer to the silvery surface and more sparse closer to the poles. In order to quantitatively describe the distribution of stars on the celestial sphere, if m is given as the limit star and the like_lmtAngle of view theta_fovThe present invention defines the star density at a certain direction (direction vector v) in the celestial sphere as the number of stars in a circular field of view centered on that direction. The definition formula is as follows:

where rho_sFor sidereal density, Card (-) indicates the number of collection elements,

angle of view theta at direction v_fovHas a star equal to or less than m in the circular field of view_lmtThe set of stars of (a), namely:

in order to facilitate statistics of star density, the invention introduces the concept of effective area. Each star point corresponds to an active area and is observed when the visual axis of the field of view is within the active area. For convenience of calculation, the method replaces a square view field with an inscribed circle view field, and judges whether the star point is in the view field or not. If the effective area of the ith star point is marked as D_iThen it satisfies:

obviously, the effective area of a star point is a circle having the star point as its center and the angle of view as its diameter. If the celestial sphere is regarded as a spherical image, each star point will have its density value in the effective area increased by one. Therefore, the density of stars at any point on the celestial sphere is equal to the number of active areas containing that point, i.e.:

as shown in fig. 1, if there are two star points, a circular area with a radius of half field angle, i.e., an effective area of the star, can be obtained on the unit spherical surface with the direction vector of each star point as the center. The active areas are shown in grey in the figure. The dark gray area in the figure is the intersection of the effective areas corresponding to the two star points, when the center of the field of view is located in the area, the two stars can be seen at the same time, and the density is 2; when the field of view is in a light gray area, only one star is visible, with a density of 1.

In order to traverse any direction of the whole celestial sphere, the celestial sphere is divided into 6 areas according to an inscribed cube star chart dividing method, and each area is further divided into 90 multiplied by 90 parts. At this point, the entire celestial sphere was divided into 48600 small regions, each of which was approximately 1 ° × 1 °. Calculating the star density rho by taking the central direction of each small region as v_s(v) The star density distribution of the whole celestial sphere in any direction can be calculated.

In the traditional triangle star map identification algorithm, a star diagonal distance library is used as a lookup table, and the angular distance between any two star points which is not more than a view field angle is stored in the table. Different from a triangle algorithm, the hash table star map identification algorithm uses a navigation triangle library as a lookup table, and directly carries out matching identification on triangles in the star map in the navigation triangle library. Let the navigation triangle library be Δ_g：

Δ_g＝{(i₁,j₁,k₁),(i₂,j₂,k₂),…} (13)

In the formula, (i, j, k) records the numbers of three stars constituting the triangle. Star sensitive camera stationThe triangle existing in the photographed picture is noted as

If any noise interference is not considered, under a perfect condition, at least one navigation triangle exists in the view field of any posture required to be met by the navigation triangle library for matching. The conditions can be expressed as:

when the actual observation is considered, the existence of interferences such as false stars, missing stars, position noise, star and the like exists, and triangles in the original star map triangle set have a certain probability of being mixed into the false triangles, missing or deformed. If only one star map triangle exists in the navigation triangle library and is missing or deformed due to interference, the star map cannot be identified. Therefore, when building a triangle library, a certain degree of redundancy is usually set, and at least rho exists in any posture_thOne navigation triangle is used for matching. The present invention defines p_thIs a triangular density threshold. At this time, the condition may be expressed as:

the invention defines the triangular density rho at a certain direction v in a celestial sphere_t(v) The number of triangles within a circular field of view centered on that direction. The definition formula is as follows:

similar to the calculation method of star density, any one of the trianglesThere is also an equivalent effective area for the shapes so that the triangle is visible when the center of the field of view is inside the area. Let the effective area of triangle as E_i,j,k. The difference from the star point is that the triangle occupies some space, and the condition is met only when the entire triangle is within the field of view. The invention takes the excircle of the triangle as the approximation of the occupied space of the triangle. The invention defines the outer circle of the triangle as the circle of minimum radius containing the triangle. As shown in fig. 2, when the triangle is an acute triangle or a right angle, the outer circle is an circumscribed circle, and when the triangle is an obtuse triangle, the outer circle is a circle having the longest side as the diameter. If the direction vector of the three stars is given as s_i,s_j,s_kThen the center c of the excircle_τAnd radius r_τThe following can be found:

when the triangle is an acute triangle:

wherein d is_p,q＝s_p-s_q,p,q∈{i,j,k}。

When the triangle is obtuse triangle, s is not set_kObtuse vertex:

in order to facilitate calculation, the triangle itself is replaced by the triangle excircle, the square field is replaced by the inscribed circle field, and whether the triangle is in the field is judged. In this case, the effective area of the triangle is represented by c_τA circle with a center having a radius r_eqCan be calculated as:

r_eq(i,j,k)＝r_fov-r_τ(i,j,k) (20)

wherein

The radius of the inscribed circle field of view. FIG. 3 showsIllustrating the relation of the triangle equivalent area to the triangle and the field of view. In the figure, the light circle is the external circle of a triangle, the dark circle is the effective area of the triangle, and the dotted line circles list a plurality of fields of view containing the triangle.

At this time, the triangular effective area E_i,j,kIt can be calculated that:

similar to the star density calculation, the density of triangles at any point on the celestial sphere is equal to the number of active areas containing that point, i.e.:

ρ_t(v)＝Card({(i,j,k)∈Δ_g|v∈E_i,j,k}) (22)

2. building a triangle utilization rate model:

when the star sensor operates in an orbit, the visual axis directions of the star sensor are uniformly distributed on the whole celestial sphere. However, because the triangles are different in size and brightness, for random visual axis directions, the probability of whether each triangle appears in the visual field is obviously different. The invention defines the existence probability of the triangle in the visual field as the utilization rate of the triangle.

First, the size of the triangle is considered. The larger the size of the triangle, the smaller its effective area. Conversely, the smaller the triangle, the larger its effective area. The probability of a triangle appearing within the field of view is proportional to the area of its active area. The triangle occurrence probability can be expressed as:

second, the star point detection probabilities that make up the triangle are considered. If the star points that make up a triangle cannot be detected, the triangle cannot appear within the field of view. Because the star table only provides star data of some fixed wave bands, even if the star in the star table is subjected to instrument star conversion in constructing the navigation star library, the star scatters noise and sensingInterference of factors such as additive noise and background noise, and certain errors still exist between actual stars observed by a fixed star in a star sensor camera and estimated instrument stars. Whether the fixed star can be detected or not cannot be directly judged according to the star of the instrument and the like. The invention considers that the star equal noise follows the positive space distribution, and the actual observation star equal m of the ith star_iCan be expressed as:

in the formula

The instrument star, delta, estimated for that star_mThe standard deviation of the star-like noise. The detection probability of stars can be found as:

where Φ (-) is the cumulative function of a standard normal distribution.

Finally, the utilization of triangle (i, j, k) can be calculated as:

3. construction of a navigation triangle library:

the invention constructs the navigation triangle library according to the density distribution utilization rate of the triangles. First, the field angle, the limit star, and the like, and the triangle density threshold ρ are set according to the application scene_th. Then, 10 ten thousand spiral uniformly distributed points are generated for the whole celestial sphere sampling, a simulation star map is generated for each point, all star points in each star map are combined to obtain a plurality of triangles, the utilization rate of the triangles is calculated, and the triangles are placed into a navigation triangle library. After traversal, a triangle density map of the navigation triangle library is computed while preserving the utilization and effective range of each triangle. Then, put the library inThe triangles are sorted from low to high according to the utilization rate, and whether the densities in the effective range of the triangles are all larger than the set triangle density threshold value is judged in sequence. If the triangle is larger than the preset triangle, the triangle can be eliminated, otherwise, the triangle is reserved. And finally, obtaining a screened navigation triangle library. Since the remaining triangles are all triangles with high utilization rate, the library should have the smallest number of triangles and the highest recognition rate under the precondition that the triangle density threshold is met.

In addition, in practical application of the embodiment of the present invention, the angle of view θ is_fovUnder the condition of 5.4 degrees, the density of stars in any direction of the whole celestial sphere is calculated, and the density data is projected to a two-dimensional plane according to the Robinson projection method, as shown in figure 4. The image is a Robinson projection plane development image of a celestial sphere inscribed square, the density of white parts is 0, and the density of black parts is 43. It can be seen that the stars are unevenly distributed in the celestial sphere, the density of the stars is higher when the stars are closer to the silver road surface, and the density of the stars is lower when the stars are closer to the two poles.

Same as the calculation of star density, m is equal to the limit star_lmtAngle of view θ of 7_fovThe density of triangles in any direction of the whole celestial sphere was calculated under the condition of 5.4 ° as shown in fig. 5. The figure is a Robinson projection plane expansion of a cube inscribed in the celestial sphere, where the image is colored using a logarithmic scale because the triangle density scale changes too much, with a density of 0 at white and 8984 at black. It can be seen that the distribution of the triangles in the celestial sphere is more uneven because the stars are unevenly distributed in the celestial sphere. The maximum number of triangles that can appear within the field of view is 8984, which is marked by dots in the figure.

FIG. 6 is an example of the star bank screening algorithm of the present invention. The numbers in the square matrix are the triangular density distribution in the library. After all triangles of the all celestial sphere are put into a warehouse by using a spherical spiral method and the effective area and the utilization rate of each triangle are calculated, the density distribution of the triangles in the whole warehouse is calculated. The utilization is then used to order the triangles in the library. From the lowest utilization, it is checked whether each triangle can be deleted. In the first step of the loop, the triangle with the lowest utilization is examined. The circle in the figure is the effective area of the triangle. It can be seen that the utilization rate of the triangles at all positions in the effective area is greater than or equal to the set threshold. Thus, the triangle is deleted from the library and the triangular density distribution within the active area is also reduced by one accordingly. In the second step of the loop, the triangles with the second lowest utilization are examined. Since there are cases where the density is below the threshold value within the active area of the triangle, the triangle must be preserved. By analogy, all triangles are checked in sequence, and finally a navigation triangle library with high utilization rate, low capacity and uniform density distribution is obtained.

In order to test the effect of the navigation triangle library generated by the invention, a series of navigation triangle libraries are generated by using different parameters to form an experimental group and a control group. M at the limit star_lmtAngle of view θ of 7_fovUnder 5.4 °, a series of different navigation triangle libraries were generated, and a portion of the similarly-sized libraries were selected as experimental and control groups.

Firstly, the capacity of the navigation triangle library, the density distribution of the navigation triangles and the distribution of the utilization rate of the triangles are analyzed and compared. The method uses the parameter rho_thNavigation triangle library and spherical spiral method generated by 9 with parameter n_sThe number of triangles in the triangle library generated by 4 is almost equal. And the two navigation triangle libraries are used as representatives to analyze and compare the navigation triangle density distribution and the triangle utilization rate distribution of the algorithm and the spherical spiral method.

In the experiment, according to an inscribed cube star catalogue dividing method, 48600 small sky areas in different directions are uniformly divided in a celestial sphere, and an included angle between central direction vectors of adjacent sky areas is about 1 degree. The triangle density of the navigation triangle library in any direction of the celestial sphere can be calculated according to a formula. And taking the triangle density obtained by calculating the vector of the central direction of the sky area as the triangle density of the sky area. Fig. 7 is a graph of the local density distribution of the triangle. Comparing the method of the invention with the spherical spiral method, the navigation triangle library constructed by the method of the invention has the average triangle density of 11.4 which is higher than 5.2 of the spherical spiral method in the local density map. The standard deviation of the triangular density of the method is 2.3, which is lower than 3.6 of the spherical spiral method. Obviously, the navigation triangle library constructed by the method is more uniform, and the density of the triangle is higher.

And counting to obtain the proportion of the area of the celestial region in each triangular density interval in the whole celestial sphere. The statistical results are shown in FIG. 8, where the dark color represents the method of the invention and the light color is the spherical spiral method. It can be seen that the number of low-density triangles in the navigation triangle library generated by the method of the present invention is less than that of the spherical spiral method, while the number of high-density triangles is more than that of the spherical spiral method. Macroscopically, under the condition that the number of triangles in the library is certain, the navigation triangle library generated by the method has higher triangle density. The average triangle density for the all celestial sphere of the spherical spiral method in the figure is 4.6, while the average triangle density for the all celestial sphere of the method of the present invention is 8.5. Although the two triangle bins are of the same size, the number of triangles present in the field of view is greater and the triangle density is greater because the triangles selected by the algorithm of the present invention are generally smaller.

The utilization of each triangle in the library can be calculated according to a formula. And counting the proportion of the triangle quantity in the navigation triangle library in different utilization rate intervals to obtain a triangle utilization rate distribution histogram, as shown in fig. 9. The dark colors in the figure represent the process of the invention and the light colors are the spherical spiral process. As can be seen from the figure, under the condition of the same library capacity, the triangle library generated by the method has a larger proportion of triangles with high utilization rate. The average triangle utilization rate of the spherical spiral method is 4.1 multiplied by 10^-5The average triangle utilization ratio of the method is 6.3 multiplied by 10^-5。

The analysis can prove that the triangle library generated by the method has higher triangle density and higher triangle utilization rate under the condition that the quantity of the triangles in the library is the same.

Then, a star atlas identification experiment was performed using the triangle library of the present invention. Using the table of Epighur as data source in m of extreme stars and the like_lmtAngle of view θ of 7_fov1000 random samples were taken over the entire celestial sphere at 5.4 °And attitude points are generated, a simulation star map is generated, and interference factors such as false stars, missing stars, position noise, star and the like are added to the simulation star map. And then, using a hash table star map identification algorithm to bring different navigation triangle libraries, identifying the simulation star map, and counting the identification rate. The simulated star map generated by the experiment has a small view field, so that the number of star points in the map is less than three, and the average number of star points in the star map is 10.

The first is the pseudostellar experiment. In the experiment, on the basis of a test set without noise, pseudolites with different proportions are added. The star of the fake star is random in size and position. Under the condition that the size of the library is the same, the library generated by the algorithm has higher recognition rate. The navigation triangle library generated by the algorithm of the invention is only 30k, and the recognition rate of the navigation triangle library under the condition of different pseudostellar proportions is almost the same as that of 364k in the spherical spiral method. Therefore, the algorithm of the invention greatly eliminates useless triangles with low utilization rate, and the remaining small triangles with high utilization rate can still meet the requirement of recognition rate.

Followed by a position noise experiment. In the experiment, zero-mean Gaussian noise is added to the position of each star point in the x and y directions on the basis of a test set without noise. The recognition rate of the 30k navigation triangle library and the spherical spiral homogenizing method 364k navigation triangle library are basically level. Under the interference of larger position noise, a 30k navigation triangle library with 10% capacity of the spherical spiral homogenizing method is only slightly inferior, but the difference is less than 0.5%, and the recognition rate of the 104k navigation triangle library completely exceeds that of a spherical spiral homogenizing method 364 k.

This is followed by a missing star experiment. In the experiment, a certain proportion of star points are randomly deleted on the basis of a test set without noise. The probability of each star point being deleted is the same regardless of the brightness of the star point. Under the condition that the size of the library is the same, the library generated by the algorithm has higher recognition rate, and under the condition of missing stars in a larger proportion, the algorithm has more obvious advantages. Under the condition of 60% missing satellites, the recognition rate of the navigation triangle library generated by the algorithm is about 15% -20% higher than that of a navigation triangle library generated by a spherical spiral uniform method with the same volume.

And finally, performing a star equal noise experiment. In the experiment, zero-mean Gaussian noise is added to the star of each star point on the basis of a test set without noise. After adding the noise, the stars below 7, etc. and now above 7, etc. are deleted and the stars above 7, etc. and now below 7, etc. are added to the graph. The star and other noise experiments are actually the synthesis of the pseudo star experiments and the missing star experiments, but the proportion of addition and deletion is small. Under the condition that the size of the library is the same, the library generated by the algorithm has higher recognition rate. The volume of the 104k navigation triangle library of the invention is less than 30% of that of the spherical spiral uniform method 364k, but the recognition rate is basically equal to that of the spherical spiral uniform method.

The triangular library constructed by the method can reduce the library capacity by 70-90% under the same recognition rate.

Fig. 10 and 11 analyze all the navigation triangle libraries tested in this experiment, where the abscissa is the number of triangles in the navigation triangle library and the ordinate is the recognition rate, the solid line is the method of the present invention, and the dotted line is the spherical spiral homogenization method. Wherein the recognition rate in fig. 10 is the test result under the condition that the standard deviation of the star-like noise is 1.0, and the recognition rate in fig. 11 is the test result under the condition that the standard deviation of the position noise is 1.0. It can be seen that the triangular star library generated by the method has higher recognition rate under the condition that the number of the navigation triangles is the same. It can be seen that when the number of triangles in the library exceeds 200k, the ascending trend of the recognition rate is already very smooth and tends to be saturated.

Through the embodiment, the problem of screening triangles in celestial spheres of the storage space is solved, the service efficiency of the storage space is improved, the matching degree is high, the recognition rate is enhanced, and the searching efficiency is improved. If the number of triangles stored in the library is too small, the triangles in the library are insufficient, and the recognition rate is reduced. If the number of triangles in the storage is too large, the requirement on the storage space is increased, the search efficiency is reduced, more probability of mismatching is caused, and the recognition rate is reduced. Even if the number of triangles in the library is moderate, if the selection is not proper, the situation that partial triangles exist in the library, the star points are too dark and cannot be observed, or partial triangles are difficult to appear in a view field due to the problems of the shapes and the sizes of the triangles can also occur. This situation can cause a technical problem in that a large number of invalid navigation triangles are stored in the library, and these triangles are rarely used or even never detected.

Example two

The invention embodiment of a star map recognition navigation triangle library construction device, the device includes:

an obtaining module 1400 for obtaining the size, shape and number of triangles in different directional fields of view.

And a density module 1402 for constructing a triangular density distribution model according to the size, shape and number of the triangles.

And a utilization rate module 1406, configured to generate a triangle utilization rate model according to the triangle density distribution model and a fixed star corresponding to the triangle.

A removing module 1408, configured to remove a low-utilization triangle in the triangle utilization model to obtain a triangle library.

in the formula

M is lower than the limit star_lmtThe star set of (1), Card (-) is taken from the setAn operator of the number of elements; a generating unit, further used for generating a triangular effective area E according to the constant star density distribution model_i,j,k：

Optionally, the utilization module includes: a calculating unit for calculating the occurrence probability P of the triangle according to the size and shape of the triangle and the star point information_tProbability of detection of a sum Star point P_s(ii) a A generating unit for generating the probability P of the triangle_tProbability of detection of a sum Star point P_sGenerating the triangular utilization rate model, and describing the triangular utilization rate model as follows by using a formula:

wherein

Is a unit sphere in three-dimensional space:

Get the subset of visible stars

Since the stars are mostly concentrated on the silver road surface, the stars in the celestial sphere are not uniformly distributedOn the whole celestial sphere, the closer to the silver road surface, the denser, the closer to the two poles, the sparser. In order to quantitatively describe the distribution of stars on the celestial sphere, if m is given as the limit star and the like_lmtAngle of view theta_fovThe present invention defines the star density at a certain direction (direction vector v) in the celestial sphere as the number of stars in a circular field of view centered on that direction. The definition formula is as follows:

Δ_g＝{(i₁,j₁,k₁),(i₂,j₂,k₂),…} (13)

In the formula, (i, j, k) records the numbers of three stars constituting the triangle. The triangle existing in the picture shot by the star sensor camera is recorded as

similar to the calculation method of the star density, an equivalent effective area exists in any triangle, so that the triangle can be seen when the center of the visual field is in the area. Let the effective area of triangle as E_i,j,k. The difference from the star point is that the triangle occupies some space, and the condition is met only when the entire triangle is within the field of view. The invention is triangularThe outer circle is used as an approximation of the triangular occupied space. The invention defines the outer circle of the triangle as the circle of minimum radius containing the triangle. As shown in fig. 2, when the triangle is an acute triangle or a right angle, the outer circle is an circumscribed circle, and when the triangle is an obtuse triangle, the outer circle is a circle having the longest side as the diameter. If the direction vector of the three stars is given as s_i,s_j,s_kThen the center c of the excircle_τAnd radius r_τThe following can be found:

when the triangle is an acute triangle:

wherein d is_p,q＝s_p-s_q,p,q∈{i,j,k}。

When the triangle is obtuse triangle, s is not set_kObtuse vertex:

r_eq(i,j,k)＝r_fov-r_τ(i,j,k) (20)

wherein

The radius of the inscribed circle field of view. Fig. 3 illustrates the relationship of the triangle equivalent area to the triangle, the field of view. In the figure, the light circle is the external circle of a triangle, the dark circle is the effective area of the triangle, and the dotted line circles list a plurality of fields of view containing the triangle.

At this time, the triangular effective area E_i,j,kIt can be calculated that:

ρ_t(v)＝Card({(i,j,k)∈Δ_g|v∈E_i,j,k}) (22)

2. building a triangle utilization rate model:

second, the star point detection probabilities that make up the triangle are considered. If the star points that make up a triangle cannot be detected, the triangle cannot appear within the field of view. Because the star table only provides data of stars and the like in certain fixed wave bands, even if instrument star and the like conversion is carried out on the star stars and the like in the star table when a navigation star library is constructed, due to the interference of factors such as star shot noise, sensor additive noise, background noise and the like, certain errors still exist between actual stars and the like observed by the stars in the star sensitive camera and estimated instrument stars and the like. Whether the fixed star can be detected or not cannot be directly judged according to the star of the instrument and the like. The invention considers that the star equal noise follows the positive space distribution, and the actual observation star equal m of the ith star_iCan be expressed as:

in the formula

where Φ (-) is the cumulative function of a standard normal distribution.

Finally, the utilization of triangle (i, j, k) can be calculated as:

3. construction of a navigation triangle library:

the invention constructs the navigation triangle library according to the density distribution utilization rate of the triangles. First, the field angle, the limit star, and the like, and the triangle density threshold ρ are set according to the application scene_th. Then, 10 ten thousand spiral uniformly distributed points are generated for the whole celestial sphere sampling, a simulation star map is generated for each point, all star points in each star map are combined to obtain a plurality of triangles, the utilization rate of the triangles is calculated, and the triangles are placed into a navigation triangle library. After traversal, a triangle density map of the navigation triangle library is computed while preserving the utilization and effective range of each triangle. And then, sorting the triangles in the library from low to high according to the utilization rate, and sequentially judging whether the densities in the effective range of the triangles are all larger than a set triangle density threshold value. If the triangle is larger than the preset triangle, the triangle can be eliminated, otherwise, the triangle is reserved. And finally, obtaining a screened navigation triangle library. Because the reserved triangles are all triangles with high utilization rateAnd therefore, the library should have the smallest number of triangles and the highest recognition rate under the premise that the triangle density threshold is met.

Then, a star atlas identification experiment was performed using the triangle library of the present invention. Using the table of Epighur as data source in m of extreme stars and the like_lmtAngle of view θ of 7_fovUnder the condition of 5.4 degrees, 1000 random attitude points are randomly sampled in the whole celestial sphere, a simulation star map is generated, and interference factors such as false stars, missing stars, position noise, star noise and the like are added to the simulation star map. And then, using a hash table star map identification algorithm to bring different navigation triangle libraries, identifying the simulation star map, and counting the identification rate. The simulation star map generated by the experiment is due toThe visual field is small, the number of star points in the map is less than three, and the average star point number of the star map is 10.

Through the embodiment, the problem of screening triangles in celestial spheres of the storage space is solved, the service efficiency of the storage space is improved, the matching degree is high, the recognition rate is enhanced, and the searching efficiency is improved. If the number of triangles stored in the library is too small, the triangles in the library are insufficient, and the recognition rate is reduced. If the number of triangles in the storage is too large, the requirement on the storage space is increased, the search efficiency is reduced, more probability of mismatching is caused, and the recognition rate is reduced. Even if the number of triangles in the library is moderate, if the selection is not proper, the situation that partial triangles exist in the library, the star points are too dark and cannot be observed, or partial triangles are difficult to appear in a view field due to the problems of the shapes and the sizes of the triangles can also occur. This situation can cause a technical problem in that a large number of invalid navigation triangles are stored in the library, and these triangles are rarely used or even never detected. The invention screens the triangles in the celestial sphere of the storage space, improves the use efficiency of the storage space, has high matching degree, enhances the recognition rate and improves the search efficiency.

The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.

In the above embodiments of the present invention, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.

In the embodiments provided in the present application, it should be understood that the disclosed technology can be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units may be a logical division, and in actual implementation, there may be another division, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, units or modules, and may be in an electrical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.

The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.

The foregoing is only a preferred embodiment of the present invention, and it should be noted that those skilled in the art can make various improvements and modifications without departing from the principle of the present invention, and these improvements and modifications should also be construed as the protection scope of the present invention.

Claims

1. A star map identification navigation triangle library construction method is characterized by comprising the following steps:

acquiring the size, shape and number of triangles in the field of view in different directions;

constructing a triangular density distribution model according to the size, shape and quantity of the triangles;

generating a triangular utilization rate model according to the triangular density distribution model and the fixed star and the star corresponding to the triangle;

and eliminating the triangles with low utilization rate in the triangle utilization rate model to obtain a triangle library.

2. The method of claim 1, wherein constructing a triangular density distribution model based on the triangular size shapes and numbers comprises:

obtaining effective area D of star point_iIt is defined as follows:

is a set of directional vectors, θ, in three-dimensional space_fovIs the angle of view;

generating a fixed star density distribution model according to the effective star point areas:

in the formula

M is lower than the limit star_lmtCard (·) is an operator that takes the number of elements in the set;

generating a triangular effective area E according to the star density distribution model_i,j,k：

Wherein i, j, k are trianglesAsterisk of c_τ(i, j, k) and r_eq(i, j, k) is the circle center and radius of the excircle of the triangle;

generating a triangular density distribution model according to the triangular effective area: rho_t(v)＝Card({(i,j,k)∈Δ_g|v∈E_i,j,k) } in which, Δ_gIs a set of triangles.

3. The method according to claim 1, wherein the generating a triangle utilization model according to the triangle density distribution model and the star and the like corresponding to the triangle comprises:

calculating the occurrence probability P of the triangle according to the size and the shape of the triangle and the star point information_tProbability of detection of a sum Star point P_s；

According to the occurrence probability P of the triangle_tProbability of detection of a sum Star point P_sGenerating the triangular utilization rate model, and describing the triangular utilization rate model as follows by using a formula:

4. The method of claim 1, wherein the culling low-utilization triangles in the triangle utilization model comprises:

calculating the utilization rates of all triangles in the triangle utilization rate model, and sequencing from low to high;

and sequentially judging whether the density in the effective area of each triangle is larger than the triangle density threshold value, if so, deleting, and otherwise, reserving.

5. A star atlas identification navigation triangle library construction device is characterized by comprising the following components:

the acquisition module is used for acquiring the size, shape and number of the triangles in the fields of view in different directions;

the density module is used for constructing a triangular density distribution model according to the size, shape and quantity of the triangles;

the utilization rate module is used for generating a triangular utilization rate model according to the triangular density distribution model and the fixed stars and the like corresponding to the triangles;

and the elimination module is used for eliminating the triangles with low utilization rate in the triangle utilization rate model to obtain a triangle library.

6. The apparatus of claim 5, wherein the density module comprises:

an acquisition unit for acquiring a star point effective region D_iIt is defined as follows:

the generating unit is used for generating a star density distribution model according to the star effective area:

in the formula

Star equal to lower than limit starM is equal_lmtCard (·) is an operator that takes the number of elements in the set;

a generating unit, further used for generating a triangular effective area E according to the star density distribution model_i,j,k：

Wherein i, j, k are asterisks constituting a triangle, c_τ(i, j, k) and r_eq(i, j, k) is the circle center and radius of the excircle of the triangle;

the generating unit is further configured to generate a triangular density distribution model according to the triangular effective area: rho_t(v)＝Card({(i,j,k)∈Δ_g|v∈E_i,j,k) } in which, Δ_gIs a set of triangles.

7. The apparatus of claim 5, wherein the utilization module comprises:

a calculating unit for calculating the occurrence probability P of the triangle according to the size and shape of the triangle and the star point information_tProbability of detection of a sum Star point P_s；

A generating unit for generating the probability P of the triangle_tProbability of detection of a sum Star point P_sGenerating the triangular utilization rate model, and describing the triangular utilization rate model as follows by using a formula:

8. The apparatus of claim 5, wherein the culling module comprises:

the sorting unit is used for calculating the utilization rates of all triangles in the triangle utilization rate model and sorting the triangles from low to high;

and the judging unit is used for sequentially judging whether the density in the effective area of each triangle is greater than the triangle density threshold value, if so, deleting the density, and otherwise, keeping the density.

9. A non-volatile storage medium, comprising a stored program, wherein the program, when executed, controls an apparatus in which the non-volatile storage medium is located to perform the method of any one of claims 1 to 4.

10. An electronic device comprising a processor and a memory; the memory has stored therein computer readable instructions for execution by the processor, wherein the computer readable instructions when executed perform the method of any one of claims 1 to 4.