CN101034408A - Star map matching recognizing method based on ant colony algorithm - Google Patents

Abstract

The invention is a kind of star map matching identification method based on ant colony algorithm, makes the angle information striking from between the information of Star point as input of the ant colony algorithm, and makes full use of the advantages of ant path optimization algorithm to achieve rapid real-time identification of the star map matching. Specific steps to achieve: First of all, initialize the ant colony algorithm and parameters of star map matching recognition algorithm; Second, strike the angle between Star points from data of the basic Star Sheet, set path optimization of the angular result stroken by the ant colony algorithm, to build recognition navigation satellite library according to results of optimization; Again, the star point coordinates stroken by each time, combining with focal length of optical systems to strike the angular distance between them, also make optimization of path to a set of optimization results through the ant colony algorithm; Finally, search satellite navigation library by using the set of optimized results, according to the set matching recognition threshold,to achieve speed star map matching identification of each moment.

Description

A kind of star map matching recognizing method based on ant group algorithm

Technical field

The present invention relates to a kind of star map matching recognizing method based on ant group algorithm, can be used for to Star Sensor in the celestial navigation system the quick coupling identification of responsive fixed star star chart.

Background technology

The celestial navigation technology is meant and does not rely on uphole equipment, does not carry out information transmission and exchange with the external world, one of autonomous positioning airmanship of being used widely in space flight, aviation and navigation field; It obtains space fixed star information by star sensor, independently finish aircraft location navigation task, be survey of deep space, manned space flight and the requisite gordian technique of ocean navigation, or the important assisting navigation means of satellite, long-range missile, carrier rocket, high-altitude long-range reconnaissance airplane etc.; Its key factor is the accurate measurement to the aircraft flight attitude.At present, CMOS star sensor of new generation is measured sensor because of its pointing accuracy height as attitude of flight vehicle, no attitude cumulative errors, and failover capability is subjected to people's favor with intellectuality fast.It can be aircraft the flight attitude that is accurate to several rads is provided, and need not any priori.Autonomous star chart coupling recognizer is not only wanted to realize obtaining fast of attitude as its core, when causing attitude loss for a certain reason, and can also be realized Fast Reconstruction.Therefore, recognition speed and recognition success rate just become the key index of weighing recognizer performance quality.Aspect the identification of star chart coupling, more popular algorithm has the triangle matching algorithm at present, angle of polygon is apart from matching algorithm, the primary method of identification that Bezooijen proposes etc., they have all reached comparatively good effect, and recognition speed is slow, the shortcoming of data redundancy but the triangle matching algorithm exists, and angle of polygon has the deficiency that recognition success rate is low, recognition speed is slow apart from matching algorithm, though but primary method of identification quick identification, the sky coverage rate is low; Though many improved algorithms are also arranged, as improve triangle method of identification, primary constellation method of identification etc., but the former still has data redundancy and the low shortcoming of triangle character dimension, though the latter has improved the sky coverage rate by the constellation feature, but increased the capacity of nautical star database, caused recognition speed to reduce.

Summary of the invention

Technology of the present invention is dealt with problems and is: overcome that existing star chart coupling recognizer recognition success rate is low, recognition speed slow and the deficiency of information redundancy, and a kind of star map matching recognizing method based on ant group algorithm is provided.

Technical solution of the present invention is: a kind of star map matching recognizing method based on ant group algorithm, the angular distance information that this method is asked for asterism information between any two is as the input of ant group algorithm, make full use of the advantage of ant group algorithm aspect optimum path search, realize star chart coupling identification real-time.The specific implementation step is:

(1) at first initialization ant group algorithm maximum cycle Cmax, the heuristic factor-alpha of information, expect heuristic factor-beta, pheromones intensity Q, pheromones volatility coefficient ρ, ant number m and asterism data scale N parameter, and the resolution sizes of fundamental catalog asterism number M, optical system focal distance f, pixel size (high dh * wide dw) and the sensitive area of star chart coupling recognizer (horizontal pixel number X * vertical pixel count Y) parameter.

(2) the fundamental catalog data are asked for interasteric in twos angular distance, utilized ant group algorithm that the angular distance result set of asking for is carried out optimum path search, make up the identification navigation star database according to optimizing the result.

Ask in twos according to the fundamental catalog data that the step of angular distance is between asterism:

1. from fundamental catalog, choose a pair of unlabelled asterism, ask for the angular distance between them.

If this a pair of asterism i and j be the direction vector l of unit in celestial coordinate system _iAnd l _jBe respectively:

$l_i=\left[\begin{array}{c}{\cos \mathrm{\α}}_i{\cos \mathrm{\δ}}_i\\ {\sin \mathrm{\α}}_i{\cos \mathrm{\δ}}_i\\ {\sin \mathrm{\δ}}_i\end{array}\right],l_j=\left[\begin{array}{c}{\cos \mathrm{\α}}_j{\cos \mathrm{\δ}}_j\\ {\sin \mathrm{\α}}_j{\cos \mathrm{\δ}}_j\\ {\sin \mathrm{\δ}}_j\end{array}\right]$ i，j＝1......M

Wherein, (α _i, δ _i) be asterism i right ascension, declination value, (α _j, δ _j) be asterism j right ascension, declination value; Then the angular distance expression formula of asterism i and j is:

${\mathrm{\θ}}_{\mathrm{ij}}=\frac{2\arcsin \left[\sqrt{(\cos {\mathrm{\α}}_i\cos {\mathrm{\δ}}_i-\cos {\mathrm{\α}}_j\cos {\mathrm{\δ}}_j{)}^2+{(\sin {\mathrm{\α}}_i\cos {\mathrm{\δ}}_i-\sin {\mathrm{\α}}_j\cos {\mathrm{\δ}}_j)}^2+{(\sin {\mathrm{\δ}}_i-\sin {\mathrm{\δ}}_j)}^2}\right]}{2}$

2. this a pair of asterism of mark, it is right to judge whether unlabelled asterism is arranged in the fundamental catalog, if having then jump to 1., otherwise finishes.

Utilize ant group algorithm to be to the step that the right angular distance of asterism in the fundamental catalog carries out optimum path search:

A. m ant placed on N the asterism data, make every couple of asterism angular distance θ _IjInitialization information amount τ _Ij(0) be constant, and initial time Δ τ _Ij(0)=0; τ _Ij(t) expression t moment angular distance θ _IjQuantity of information, Δ τ _Ij(t) represent that all ants stay angular distance θ constantly at t _IjOn quantity of information and, t=1,2...;

B. cycle index Count=Count+1, (the Count initial value is made as 0), the taboo table tabu of ant _kCall number k=1, wherein tabu _kThe current asterism of passing by of expression ant k;

C. ant number k=k+1, the probability that the ant individuality is calculated according to following state transition probability formula is selected asterism j and is advanced;

$p_{\mathrm{ij}}^k\left(t\right)=\left\{\begin{array}{cc}\frac{{\left[{\mathrm{\τ}}_{\mathrm{ij}}\left(t\right)\right]}^{\mathrm{\α}}\·{\left[{\mathrm{\η}}_{\mathrm{ij}}\left(t\right)\right]}^{\mathrm{\β}}}{\underset{s\⋐\{N-\mathrm{tab}{u}_k\}}{\mathrm{\Σ}}{\left[{\mathrm{\τ}}_{\mathrm{is}}\left(t\right)\right]}^{\mathrm{\α}}\·{\left[{\mathrm{\η}}_{\mathrm{is}}\left(t\right)\right]}^{\mathrm{\β}}},& j\∈\{N-\mathrm{tab}{u}_k\}\\ 0,& j\∉\{N-\mathrm{tab}{u}_k\}\end{array}\right.$

Wherein, η _Ij(t) be heuristic function, the expression ant is transferred to the expected degree of asterism j, η from asterism i _Ij(t)=1/ θ _Ij, θ _IjAngular distance between expression asterism i and the asterism j;

D. after choosing ant is moved on the new asterism, and this asterism is recorded in the taboo table of this ant individuality;

E. if k＜m then jumps to the c step, otherwise arrives next step;

F. upgrade quantity of information on every paths according to following formula:

τ _Ij(t+N)=(1-ρ) τ _Ij(t)+Δ τ _Ij(t), cross angular distance θ when k ant _IjThe time, ${\mathrm{\Δ\τ}}_{\mathrm{ij}}\left(t\right)=\frac{Q}{L_k};$

Wherein, L _kRepresent the total length of k ant at this circulation angular distance of passing by;

G. as if Count＞Cmax, then loop ends forwards the b step to otherwise empty the taboo table.

(3) each is extracted constantly the asterism coordinate that obtains, ask between any two angular distance, carry out optimum path search by ant group algorithm equally, form and optimize result set in conjunction with the optical system focal length;

By the asterism coordinate that extraction obtains, asking between any two in conjunction with the optical system focal length, the step of angular distance is:

(a) it is right to choose a pair of unlabelled asterism coordinate from all asterism coordinates that extract, and asks for the angular distance between them.

Be located in the star sensor coordinate system projection vector m of unit of two asterism i and j _iAnd m _jCan be write as:

$m_i=\frac{1}{p}\left[\begin{array}{c}{x}_i\mathrm{dh}\\ y_i\mathrm{dw}\\ f\end{array}\right],m_j=\frac{1}{q}\left[\begin{array}{c}{x}_j\mathrm{dh}\\ y_j\mathrm{dw}\\ f\end{array}\right]$

Wherein, (x _i, y _i) be the coordinate of asterism i on sensitive area, (x _i, y _i) be the coordinate of asterism j on sensitive area, ${(x_i^2{\mathrm{dh}}^2+y_i^2{\mathrm{dw}}^2+f^2)}^{1/2},q={(x_j^2{\mathrm{dh}}^2+y_j^2{\mathrm{dw}}^2+f^2)}^{1/2}$ Be the normalizing coefficient;

Then the angular distance between these two asterisms is:

θ _ij″＝arccos(m _im _j)＝arccos[(x _ix _jdh ²+y _iy _jdw ²+f ²)/(pq)]

(b) this a pair of asterism coordinate of mark is right, judges whether to also have unlabelled asterism coordinate right, if having then jump to (a), otherwise finishes.

(4) utilize optimization result set retrieval navigation star database to finish the quick coupling identification of star chart, at first pass through the binary tree search mode of linear list, realize optimizing the quick retrieval of result set in navigation star database, the back is according to result for retrieval, and, finish each quick star chart coupling identification constantly in conjunction with the coupling recognition threshold of setting; Wherein, the setting of coupling recognition threshold depends primarily on the current size of extracting the right angular distance mean value of all asterism coordinates, and general threshold value just is taken as 0.1 times that is averaged.

Principle of the present invention is: at first, and initialization ant group algorithm and star chart coupling recognizer parameter; Secondly, the fundamental catalog data are asked for interasteric in twos angular distance, utilize ant group algorithm that the angular distance result set of asking for is carried out optimum path search, make up the identification navigation star database according to optimizing the result; Once more, each is extracted the asterism coordinate that obtains constantly, ask between any two angular distance, carry out optimum path search by ant group algorithm equally, form and optimize result set in conjunction with the optical system focal length; At last, utilize and optimize result set retrieval navigation star database,, finish each quick star chart coupling identification constantly according to the coupling recognition threshold of setting.

The present invention's advantage compared with prior art is: the present invention has overcome traditional star map matching recognizing method at recognition success rate, recognition speed, the database size, deficiency on the noise resisting ability, ant group algorithm is incorporated in the star chart coupling identification of celestial navigation system, fundamental catalog data and each asterism coordinate data of extracting are constantly asked for the input of two angular distances set of angular distance formation as ant group algorithm respectively in twos, make full use of the advantage of ant group algorithm aspect optimum path search, to these two angular distance set carrying out fast path optimizing, remove relatively poor, the data of distortion, thus realize in real time, fast, the identification of the break-even star chart coupling of data.

Description of drawings

Fig. 1 is a kind of star map matching recognizing method process flow diagram based on ant group algorithm of the present invention.

Fig. 2 is two asterism angular distance synoptic diagram on the unit celestial sphere.

Fig. 3 is the right angular distance synoptic diagram of two asterism coordinates.

Embodiment

As shown in Figure 1, specific implementation method of the present invention is as follows:

(1) at first, the maximum cycle Cmax (〉=200 of initialization ant group algorithm, present embodiment gets 200), cycle index is represented with Count, initial value is made as 0, the heuristic factor-alpha of information (0≤α≤5, present embodiment gets 1), expect heuristic factor-beta (0≤β≤5, present embodiment gets 1), pheromones intensity Q (10≤Q≤10000, present embodiment gets 2000), pheromones volatility coefficient ρ (0.1≤ρ≤0.99, present embodiment gets 0.2), ant number m (〉=20, present embodiment gets 50) and asterism data scale N (〉=20, present embodiment gets 50) parameter, and the fundamental catalog asterism number M (3000) of star chart coupling recognizer, optical system focal distance f (98.8mm), the resolution sizes of pixel size (high dh * wide dw) (27 μ m * 27 μ m) and sensitive area (horizontal pixel number X * vertical pixel count Y) (640 * 480) parameter.

(2) the fundamental catalog data are asked for interasteric in twos angular distance, utilized ant group algorithm that the angular distance result set of asking for is carried out optimum path search, make up the identification navigation star database according to optimizing the result.

Ask in twos according to the fundamental catalog data that the step of angular distance is between asterism:

1. from fundamental catalog, choose a pair of unlabelled asterism, ask for the angular distance between them.

If this a pair of asterism i and j be the direction vector l of unit in celestial coordinate system _iAnd l _jBe respectively:

$l_i=\left[\begin{array}{c}{\cos \mathrm{\α}}_i{\cos \mathrm{\δ}}_i\\ {\sin \mathrm{\α}}_i{\cos \mathrm{\δ}}_i\\ {\sin \mathrm{\δ}}_i\end{array}\right],l_j=\left[\begin{array}{c}{\cos \mathrm{\α}}_j{\cos \mathrm{\δ}}_j\\ {\sin \mathrm{\α}}_j{\cos \mathrm{\δ}}_j\\ {\sin \mathrm{\δ}}_j\end{array}\right]$ i，j＝1......M

Wherein, (α _i, δ _i) be asterism i right ascension, declination value, (α _j, δ _j) be asterism j right ascension, declination value;

As shown in Figure 2, the O point is the centre of sphere of unit celestial sphere among the figure, central angle θ _IjAngular distance for asterism i and j; Because θ _Ij=2 β, then the angular distance θ of asterism i and j _IjExpression formula is:

${\mathrm{\θ}}_{\mathrm{ij}}=\frac{2\arcsin \left[\sqrt{(\cos {\mathrm{\α}}_i\cos {\mathrm{\δ}}_i-\cos {\mathrm{\α}}_j\cos {\mathrm{\δ}}_j{)}^2+{(\sin {\mathrm{\α}}_i\cos {\mathrm{\δ}}_i-\sin {\mathrm{\α}}_j\cos {\mathrm{\δ}}_j)}^2+{(\sin {\mathrm{\δ}}_i-\sin {\mathrm{\δ}}_j)}^2}\right]}{2}$

2. this a pair of asterism of mark, it is right to judge whether unlabelled asterism is arranged in the fundamental catalog, if having then jump to 1., otherwise finishes.

Utilize ant group algorithm to be to the step that the right angular distance of asterism in the fundamental catalog carries out optimum path search:

A. m ant placed on N the asterism data, make every couple of asterism angular distance θ _IjInitialization information amount τ _Ij(0) be constant, and initial time Δ τ _Ij(0)=0; τ wherein _Ij(t) expression t moment angular distance θ _IjQuantity of information, Δ τ _Ij(t) the expression ant is stayed angular distance θ constantly at t _IjOn quantity of information and, t=1,2....

B. cycle index Count=Count+1, the taboo table tabu of ant _kCall number k=1, wherein taboo table tabu _kThe current asterism of passing by of expression ant k.

C. ant number k=k+1, the probability that the ant individuality is calculated according to following state transition probability formula is selected asterism j and is advanced;

$p_{\mathrm{ij}}^k\left(t\right)=\left\{\begin{array}{cc}\frac{{\left[{\mathrm{\τ}}_{\mathrm{ij}}\left(t\right)\right]}^{\mathrm{\α}}\·{\left[{\mathrm{\η}}_{\mathrm{ij}}\left(t\right)\right]}^{\mathrm{\β}}}{\underset{s\⋐\{N-\mathrm{tab}{u}_k\}}{\mathrm{\Σ}}{\left[{\mathrm{\τ}}_{\mathrm{is}}\left(t\right)\right]}^{\mathrm{\α}}\·{\left[{\mathrm{\η}}_{\mathrm{is}}\left(t\right)\right]}^{\mathrm{\β}}},& j\∈\{N-\mathrm{tab}{u}_k\}\\ 0,& j\∉\{N-\mathrm{tab}{u}_k\}\end{array}\right.$

Wherein, η _Ij(t) be heuristic function, the expression ant is transferred to the expected degree of asterism j, η from asterism i _Ij(t)=1/ θ _Ij, θ _IjAngular distance between expression asterism i and the asterism j;

D. after choosing ant is moved on the new asterism, and this asterism is recorded in the taboo table of this ant individuality;

E. if k＜m then jumps to the c step, otherwise arrives next step.

F. upgrade quantity of information on every paths according to following formula:

τ _Ij(t+N)=(1-ρ) τ _Ij(t)+Δ τ _Ij(t), cross angular distance θ when k ant _IjThe time, ${\mathrm{\Δ\τ}}_{\mathrm{ij}}\left(t\right)=\frac{Q}{L_k};$

Wherein, L _kRepresent the total length of k ant at this circulation angular distance of passing by;

G. as if Count＞Cmax, then loop ends forwards the b step to otherwise empty the taboo table.

(3) each is extracted constantly the asterism coordinate that obtains, ask between any two angular distance, carry out optimum path search by ant group algorithm equally, form and optimize result set in conjunction with the optical system focal length;

By the asterism coordinate that extraction obtains, asking between any two in conjunction with the optical system focal length, the step of angular distance is:

(a) it is right to choose a pair of unlabelled asterism coordinate from all asterism coordinates that extract, and asks for the angular distance between them.

As shown in Figure 3, in star sensor coordinate system O ' XYZ, the projection vector m of unit of two asterism i and j _iAnd m _j, being the unit vector of O ' A and O ' B direction, can be write as:

$m_i=\frac{1}{p}\left[\begin{array}{c}{x}_i\mathrm{dh}\\ y_i\mathrm{dw}\\ f\end{array}\right],m_j=\frac{1}{q}\left[\begin{array}{c}{x}_j\mathrm{dh}\\ y_j\mathrm{dw}\\ f\end{array}\right]$

Wherein, (x _i, y _i) be the coordinate of asterism i on sensitive area ox ' y ', (x _j, y _j) be the coordinate of asterism j on ox ' y ', OO ' is a focal distance f, ${(x_i^2{\mathrm{dh}}^2+y_i^2{\mathrm{dw}}^2+f^2)}^{1/2},q={(x_j^2{\mathrm{dh}}^2+y_j^2{\mathrm{dw}}^2+f^2)}^{1/2}$ Be the normalizing coefficient;

The angular distance θ between asterism i and the j then _Ij" be A and B asterism coordinate between angular distance ∠ AO ' B, that is:

θ _ij″＝∠AO′B＝arccos(m _im _j)＝arccos[(x _ix _jdh ²+y _iy _jdw ²+f ²)/(pq)]

(b) this a pair of asterism coordinate of mark is right, judges whether to also have unlabelled asterism coordinate right, if having then jump to (a), otherwise finishes.

(4) utilize optimization result set retrieval navigation star database to finish the quick coupling identification of star chart, at first pass through the binary tree search mode of linear list, being about to navigation star database stores with the linear list mode, search by binary tree search mode in the data structure again, realize optimizing the quick retrieval of result set in navigation star database, the back is according to result for retrieval, and in conjunction with the coupling recognition threshold of setting, quick star chart of finishing each moment mates identification; Wherein, the setting of coupling recognition threshold depends primarily on the current size of extracting the right angular distance mean value of all asterism coordinates, and general threshold value just is taken as 0.1 times that is averaged.

The content that is not described in detail in the instructions of the present invention belongs to this area professional and technical personnel's known prior art.

Claims (6)

1, a kind of star map matching recognizing method based on ant group algorithm is characterized in that may further comprise the steps:

(1) initialization ant group algorithm and star chart coupling recognizer parameter;

(2) the fundamental catalog data are asked for interasteric in twos angular distance, utilized ant group algorithm that the angular distance result set of asking for is carried out optimum path search, make up the identification navigation star database according to optimizing the result;

(3) each is extracted constantly the asterism coordinate that obtains, ask between any two angular distance, carry out optimum path search by ant group algorithm equally, form and optimize result set in conjunction with the optical system focal length;

(4) utilize optimization result set retrieval navigation star database,, finish each quick star chart coupling constantly and discern according to the coupling recognition threshold of setting.

2, star map matching recognizing method based on ant group algorithm according to claim 1, it is characterized in that: initialization ant group algorithm and star chart coupling recognizer parameter has ant group algorithm maximum cycle Cmax in the described step (1), the heuristic factor-alpha of information, expect heuristic factor-beta, the pheromones intensity Q, pheromones volatility coefficient ρ, ant number m and asterism data scale N, the fundamental catalog asterism number M of star chart coupling recognizer, the optical system focal distance f, the resolution sizes of pixel size (high dh * wide dw) and sensitive area (horizontal pixel number X * vertical pixel count Y).

3, the star map matching recognizing method based on ant group algorithm according to claim 1 is characterized in that: ask in the described step (2) in the fundamental catalog in twos that the step of angular distance is between asterism:

1. establish any two asterism i and the direction vector l of unit of j in celestial coordinate system in the fundamental catalog _iAnd l _jBe respectively:

$l_i=\left[\begin{array}{c}\cos {\mathrm{\α}}_i\cos {\mathrm{\δ}}_i\\ \sin {\mathrm{\α}}_i\cos {\mathrm{\δ}}_i\\ \sin {\mathrm{\δ}}_i\end{array}\right],l_j=\left[\begin{array}{c}\cos {\mathrm{\α}}_j\cos {\mathrm{\δ}}_j\\ \sin {\mathrm{\α}}_j\cos {\mathrm{\δ}}_j\\ \sin {\mathrm{\δ}}_j\end{array}\right]$ i，j＝1……M

Wherein, (α _i, δ _i) be asterism i right ascension, declination value, (α _j, δ _j) be asterism j right ascension, declination value;

2. then the angular distance expression formula of asterism i and j is:

${\mathrm{\θ}}_{\mathrm{ij}}=\frac{2\arcsin \left[\sqrt{{(\cos {\mathrm{\α}}_i\cos {\mathrm{\δ}}_i-\cos {\mathrm{\α}}_j\cos {\mathrm{\δ}}_j)}^2+{({\sin \mathrm{\α}}_i\cos {\mathrm{\δ}}_i-\sin {\mathrm{\α}}_j\cos {\mathrm{\δ}}_j)}^2+{(\sin {\mathrm{\δ}}_i-\sin {\mathrm{\δ}}_j)}^2}\right]}{2}$

4, the star map matching recognizing method based on ant group algorithm according to claim 1 is characterized in that: the step of utilizing ant group algorithm to carry out optimum path search in described step (2) and the step (3) is:

A. m ant placed on N the asterism data, make every couple of asterism angular distance θ _IjInitialization information amount τ _Ij(0) be constant, and initial time Δ τ _Ij(0)=0; τ _Ij(t) expression t moment angular distance θ _IjQuantity of information, Δ τ _Ij(t) represent that all ants stay angular distance θ constantly at t _IjOn quantity of information and, t=1,2

B. cycle index Count=Count+1, the taboo table tabu of ant _kCall number k=1, wherein tabu _kThe current asterism of passing by of expression ant k;

C. ant number k=k+1, the probability that the ant individuality is calculated according to following state transition probability formula is selected asterism j and is advanced;

$p_{\mathrm{ij}}^k\left(t\right)=\left\{\begin{array}{cc}\frac{{\left[{\mathrm{\τ}}_{\mathrm{ij}}\left(t\right)\right]}^{\mathrm{\α}}\·{\left[{\mathrm{\η}}_{\mathrm{ij}}\left(t\right)\right]}^{\mathrm{\β}}}{\underset{s\⋐\left\{{N-\mathrm{tabu}}_k\right\}}{\mathrm{\Σ}}{\left[{\mathrm{\τ}}_{\mathrm{is}}\left(t\right)\right]}^{\mathrm{\α}}\·{\left[{\mathrm{\η}}_{\mathrm{is}}\left(t\right)\right]}^{\mathrm{\β}}},& j\∈\left\{{N-\mathrm{tabu}}_k\right\}\\ 0,& j{\∉\{N-\mathrm{tabu}}_k\}\end{array}\right.$

Wherein, η _Ij(t) be heuristic function, the expression ant is transferred to the expected degree of asterism j, η from asterism i _Ij(t)=1/ θ _Ij, θ _IjAngular distance between expression asterism i and the asterism j;

D. after choosing ant is moved on the new asterism, and this asterism is recorded in the taboo table of this ant individuality;

E. if k＜m then jumps to the c step, otherwise arrives next step;

F. upgrade quantity of information on every paths according to following formula:

τ _Ij(t+N)=(1-ρ) τ _Ij(t)+Δ τ _Ij(t), cross angular distance θ when k ant _IjThe time, $\mathrm{\Δ}{\mathrm{\τ}}_{\mathrm{ij}}\left(t\right)=\frac{Q}{L_k};$

Wherein, L _kRepresent the total length of k ant at this circulation angular distance of passing by;

G. as if Count＞Cmax, then loop ends forwards the b step to otherwise empty the taboo table.

5, the star map matching recognizing method based on ant group algorithm according to claim 1 is characterized in that: in conjunction with the optical system focus information, the step of asking for angular distance between the two asterism coordinates is in the described step (3):

(a) be located in the star sensor coordinate system projection vector m of unit of two asterism i and j _iAnd m _jCan be write as:

$m_i=\frac{1}{p}\left[\begin{array}{c}{x}_i\mathrm{dh}\\ y_i\mathrm{dw}\\ f\end{array}\right],m_j=\frac{1}{q}\left[\begin{array}{c}{x}_j\mathrm{dh}\\ y_j\mathrm{dw}\\ f\end{array}\right]$

Wherein, (x _i, y _i) be the coordinate of asterism i on sensitive area, (x _i, y _i) be the coordinate of asterism j on sensitive area, $p={(x_i^{2}d{h}^2+y_i^{2}d{w}^2+f^2)}^{1/2},q={(x_j^{2}d{h}^2+y_j^{2}d{w}^2+f^2)}^{1/2}$ Be the normalizing coefficient;

(b) then the angular distance between these two asterisms is:

θ _ij″＝arccos(m _im _j)＝arccos[(x _ix _jdh ²+y _iy _jdw ²+f ²)/(pq)]

6, the star map matching recognizing method based on ant group algorithm according to claim 1, it is characterized in that: utilize in the described step (4) and optimize the quick coupling identification that result set retrieval navigation star database is finished star chart, at first pass through the binary tree search mode of linear list, realize optimizing the quick retrieval of result set in navigation star database, the back is according to result for retrieval, and, finish each quick star chart coupling identification constantly in conjunction with the coupling recognition threshold of setting.

Images