JP4133571B2 - Star sensor - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a star sensor that is mounted on a satellite and measures the attitude of the satellite by detecting the position of a star such as a star.
[0002]
[Prior art]
In a conventional star sensor, image data output from a two-dimensional image sensor (CCD) is stored in a memory without any processing, and after storage, the memory is accessed from a control device to process image information. By doing so, the attitude of the satellite was calculated (for example, see Patent Document 1). Further, the image is divided into a plurality of partial areas, and when the image data is stored in the memory, it is detected whether or not there is a star for each partial area, and the control device performs memory access for performing image processing. The attitude of the satellite was calculated at a certain high speed by reducing only the detected partial area (see, for example, Patent Document 2).
[0003]
In addition, with conventional star sensors, if the sun enters the field of view or out-of-field sun interference angle, stars cannot be seen and attitude determination cannot be performed. Attitude sensors such as sensors and rate sensors are also required. Alternatively, a system has been devised in which an ND filter is inserted only when the sun enters the field of view to attenuate input light and a posture is determined by detecting a sun image (see, for example, Patent Document 3).
[0004]
[Patent Document 1]
JP-A-7-270177 (first page, FIG. 1)
[Patent Document 2]
JP-A-11-291996 (2nd page, FIG. 1)
[Patent Document 3]
Japanese Patent No. 3331890 (page 4, FIG. 1)
[0005]
[Problems to be solved by the invention]
However, the prior art has the following problems. When the control device processes image data, in order to detect the presence and coordinates of a star, all of the memory containing the image data or all of the memory in the partial area determined to have a star are accessed. It was necessary and it took time to process the image information.
[0006]
Further, when using a CCD having a large number of pixels, there is a problem that if the drive signal speed is slow, noise components increase in the CCD transmission path or the control cycle becomes long. Conversely, when the drive signal speed is high, there are problems such as an increase in signal noise due to high-speed rectangular waves, etc., and an expensive board design that requires special care so that the signal does not deteriorate even at high speeds. It was.
[0007]
Furthermore, as a countermeasure when the sun enters the field of view, mounting an attitude sensor in addition to the star sensor increases costs and weight. Alternatively, when an ND filter is added to the star sensor as a countermeasure when the sun enters the field of view, a drive part for inserting and removing the ND filter is required, which is problematic in terms of cost, weight, and reliability. was there.
[0008]
The present invention has been made to solve the above-described problems, and an object of the present invention is to obtain a star sensor that shortens the image processing time for calculating the attitude of a satellite.
[0009]
[Means for Solving the Problems]
The star sensor according to the present invention includes an image input unit that is mounted on a satellite and captures a star image, an image memory that stores image data captured by the image input unit, and images that are sequentially transmitted from the image input unit. A coordinate information detection unit that sequentially detects coordinate value data corresponding to a maximum value in continuous image data that is equal to or greater than a preset threshold value from the data, and the coordinate value data detected by the coordinate information detection unit sequentially The information memory to be stored and the information on the distribution of the star are stored in the database, the coordinate value data stored in the information memory are sequentially extracted, and the image data in the vicinity of the coordinate value data is acquired from the image memory, Star coordinates are calculated by detecting the maximum value from the acquired image data, and the calculated star coordinates and the distribution information of the stars in the database are calculated. The image input unit is one that was a orientation calculation unit for calculating the attitude of the satellite on board by matching and.
[0010]
DETAILED DESCRIPTION OF THE INVENTION
Embodiment 1 FIG.
FIG. 1 is a block diagram showing a configuration of a star sensor according to Embodiment 1 of the present invention. The image input unit 1 includes an optical system, a two-dimensional image sensor (CCD or the like), an A / D converter, and the like, and outputs image data obtained by imaging a star as pixel data in order pixel by pixel. The output pixel data is recorded in the image memory 2 as it is and also sent to the coordinate information detection unit 3.
[0011]
The coordinate information detection unit 3 detects the coordinates of a stellar in a short time by an algorithm or the like described later based on the pixel data sequentially sent from the image input unit 1 and records the detected coordinate value data in the information memory 4. To do. Coordinate value data may be detected multiple times when a single star image spans multiple partial areas, or may not be the center coordinate of the star image. Accurate star coordinate detection will be described later. Here, it is meaningful to take out the reference information of the stellar coordinates in a short time in accordance with the speed at which the pixel data is recorded in the image memory 2.
[0012]
The posture calculation unit 5 first acquires coordinate value data from the information memory 4, acquires pixel data in the vicinity of the coordinate value data from the image memory 2, performs image processing, and is captured by the image input unit 1. Calculate the exact coordinates of the star from the image. If the position of a star in the image is known, the satellite attitude can be calculated by collating it with a previously recorded database of star distributions.
[0013]
Next, a method of detecting coordinate value data corresponding to the maximum value for each image data continuous by a coordinate information detection unit 3 that is equal to or greater than a preset threshold will be described with reference to FIG. Here, a case where a CCD is used as the two-dimensional imaging element of the image input unit 1 will be described. FIG. 2 is an explanatory diagram for detecting the coordinate value data of the star from the image data for one row of the CCD according to the first embodiment of the present invention. Here, the X-th pixel data in the Y-th row can be expressed as d (X, Y). For example, X and Y when d (X, Y) is maximized are coordinate value data. . In FIG. 2, two locations are detected: the maximum value of the star 1 and the maximum value of the star 2.
[0014]
FIG. 3 is a flowchart of an algorithm for detecting the coordinate value data of the star by the coordinate information detection unit 3 according to Embodiment 1 of the present invention. The X coordinate direction data is sequentially searched from the start point x0 to the end point x1 and in the Y direction from the start point y0 to the end point y1 (see FIG. 2). In general, in order to prevent pixel data from becoming zero and the influence of noise even in a region without a star, a threshold value is set in advance, and only pixel data exceeding the threshold value is a target for detecting coordinate value data of the star. When pixel data exceeding the threshold continues, the maximum value dmax and the X coordinate pX and Y coordinate pY at that time are recorded in a temporary register. Then, the coordinate value data remaining in the register when the pixel data falls below the threshold value is sequentially recorded in the information memory 4 as MX and MY (corresponding to process A in FIG. 3). Such an algorithm can be realized with a relatively small circuit, has a high processing speed, and can detect coordinate value data in a short time according to the speed at which the image data is recorded in the image memory 2. It becomes possible.
[0015]
As another example of the algorithm, consider a case where the maximum number V of coordinate value data detected for each row and the maximum number W of rows are set in advance. If such a limitation is added, an area of V × W as the number of data may be held in advance as a storage area in the information memory 4. Furthermore, since the Y coordinate corresponds to a predetermined address in the information memory 4, only the X coordinate is required as the coordinate value data to be recorded, and the memory size and circuit scale can be reduced.
[0016]
According to the first embodiment, the attitude calculation unit 5 takes out only the data in the vicinity of the coordinate value data from the image memory 2 based on the coordinate value data of the star acquired from the information memory 4. Can be calculated. Therefore, the number of memory accesses for only a limited area based on the coordinate value data in the information memory 4 is sufficient, and the processing time can be greatly shortened. Furthermore, in accordance with the speed at which the image data is recorded in the image memory 2, the coordinate information detection unit 3 can detect the coordinate value data in a short time, and can detect the coordinates of a fast and accurate star. As a result, the satellite attitude can be calculated accurately at high speed.
[0017]
Embodiment 2. FIG.
FIG. 4 is a block diagram showing the configuration of the star sensor according to Embodiment 2 of the present invention. The image input unit 1 includes a two-dimensional image sensor (CCD or the like) 11 and an A / D converter 12. Further, the two-dimensional image sensor 11 and the A / D converter 12 have a function of making the clock signal variable by an external variable drive signal.
[0018]
The drive circuit unit 15 a supplies variable drive signals for performing synchronous control with variable clock signals to the two-dimensional imaging device 11, the A / D converter 12, the image memory 2, the coordinate information detection unit 3, and the information memory 4. Respectively. The two-dimensional imaging device 11 acquires an image readout signal such as a horizontal transfer signal and a vertical transfer signal from the drive circuit unit 15a as a variable drive signal, and extracts pixel data at a timing based on the variable drive signal. Further, the A / D converter 12 acquires a timing signal for performing A / D conversion in synchronization with the timing of the pixel data sent from the two-dimensional image sensor 11 from the drive circuit unit 15a as a variable drive signal, A / D conversion is performed.
[0019]
The timing signal acquired by the A / D converter 12 is further transmitted to the image memory 2, the coordinate information detection unit 3, and the information memory 4 as a variable drive signal. The image memory 2 writes image data from the A / D converter 12 based on the variable drive signal. The coordinate information detection unit 3 detects coordinate value data from the image data from the A / D converter 12 based on the variable drive signal. The information memory 4 writes the coordinate value data from the coordinate information detector 3 based on the variable drive signal.
[0020]
The variable drive signal output from the drive circuit unit 15a is generally a high-speed pulse train. The analog signal output from the two-dimensional image sensor 11 reaches the A / D converter 12, but both the two-dimensional image sensor 11 and the A / D converter 12 operate with the variable drive signal output from the drive circuit unit 15a. Therefore, the noise component tends to increase. In addition, the higher the frequency of the variable drive signal, the greater the noise component. However, if the frequency is slowed down as a whole, it takes time to read out pixel data for one screen. In general, the pulse period of the horizontal transfer signal input to the two-dimensional image pickup device 11 is constant throughout all pixels, but the drive circuit unit 15a can partially change the frequency of the drive signal including the horizontal transfer signal as a variable drive signal. .
[0021]
The light receiving area of a general CCD as the two-dimensional image pickup device 11 is a horizontally long rectangle in accordance with the television screen. However, the star sensor does not need to be horizontally long and does not necessarily require all pixel data. However, when reading out pixel data of the two-dimensional image sensor 11, there is a case where a transfer signal for reading out all pixels, such as a CCD, is required even if only a part of the pixel data is to be taken out. In this case, the pixel data transfer time can be made variable by outputting the variable drive signal using the drive signal frequency changing function of the drive circuit unit 15a. In other words, the pixel data transfer period in the required area is operated at a relatively low frequency to suppress an increase in noise components, and the pixel data transfer period in an unnecessary area is transferred by operating at a relatively high frequency. Time can be shortened. Since the frequency is not constant, it contributes to the reduction of electromagnetic radiation intensity.
[0022]
According to the second embodiment, by using the variable drive signal of the drive circuit unit 15a, the image data of a specific area of the two-dimensional image sensor 11 can be acquired at a high speed while reducing the influence of noise. As a result, the detection coordinates of the star and the attitude of the satellite can be calculated accurately at high speed. Furthermore, the effect of relaxing the design requirements for the noise substrate and reducing the electromagnetic radiation intensity can be obtained.
[0023]
Embodiment 3 FIG.
FIG. 5 is a block diagram showing a configuration of a star sensor according to Embodiment 3 of the present invention. With this configuration, the level of lightness and darkness of image data acquired from the image input unit 1 can be made variable. As an example in which the gray level is extremely high, a method for detecting the sun position when sunlight enters the field of view of the two-dimensional image sensor 11 will be described. In FIG. 5, the drive circuit unit 15 b has a function capable of variably controlling the exposure time of the two-dimensional image sensor 11 and the amplification degree of the amplifier 13 based on a command from the attitude calculation unit 5. When the sun enters the field of view of the two-dimensional image sensor 11, the image becomes too bright and the position of the star cannot be detected. Therefore, the drive circuit unit 15b attempts to detect the position of the sun by shortening the exposure time of the two-dimensional image sensor 11 and decreasing the amplification degree of the amplifier 13.
[0024]
Normally, as described in the first embodiment, by adopting the configuration of FIG. 1, the attitude calculation unit 5 obtains only the data in the vicinity of the coordinate value data based on the coordinate value data of the star acquired from the information memory 4. By taking out from the image memory 2, it becomes possible to calculate the exact coordinates of the star. On the other hand, when the sun enters the field of view of the two-dimensional image sensor 11, the image becomes extremely bright, and the coordinate information detection unit 3 cannot detect the coordinate information of the star by threshold determination. Even in such a case, the intensity level at which sunlight is captured as image data is reduced to a level equivalent to the intensity level of the star when the position of the star can be detected without being affected by sunlight. If it is possible, the position of the sun can be detected.
[0025]
Whether the sun has entered the field of view of the two-dimensional image sensor 11 can be determined as follows. First, the coordinate information detection unit 3 coordinates a predetermined specific numerical value which cannot be an actual coordinate value when the acquired image data is only a high value exceeding the threshold and the coordinate value data cannot be detected. Write to the information memory 4 as value data. The posture calculation unit 5 can determine that the image data is affected by sunlight when the coordinate value data acquired from the information memory 4 indicates a specific numerical value.
[0026]
As another method for determining that the sun has entered the field of view of the two-dimensional image sensor 11, the attitude calculation unit 5 can also determine that there is no corresponding coordinate value data in the information memory 4. is there. It is also possible for the posture calculation unit 5 to obtain an average value of pixel intensity for each partial area of the image memory 2 and to determine that the value is close to a saturation value. If these methods are adopted, the coordinate information detection unit 3 does not need to write the coordinate value data of a specific numerical value in the information memory 4.
[0027]
When it is determined that the sun has entered the field of view of the two-dimensional image sensor 11, the posture calculation unit 5 samples image data from the image memory 2 and obtains an average value of pixel intensity. The location where the image data is sampled can be specified in advance as a set of partial areas within the area captured as image data. For example, five locations at the four corners and the central portion can be specified as partial areas. The posture calculation unit 5 obtains an average value of the intensity of the image data for each partial area.
[0028]
When a bright image is captured, the exposure time is shortened and the degree of amplification is reduced to reduce the intensity of the image data. Therefore, the actual image input unit 1 is used to calculate the average value of the gray level of the image data in the partial area at that time for various light intensity inputs, and the optimum exposure time and amplification level for the obtained average value of the gray levels. Collect data in advance. Based on the data collection result, exposure time data and amplification degree data corresponding to the average value of pixel intensity are stored in advance in the database of the attitude calculation unit 5.
[0029]
The posture calculation unit 5 extracts the maximum value from the average value of the intensities obtained for each partial area, and further extracts exposure time data and amplification degree data corresponding to the extracted values from the database. Next, the attitude calculation unit 5 outputs the extracted exposure time data and amplification degree data as a drive control signal change command to the drive circuit unit 15b.
[0030]
The drive circuit unit 15 b transmits exposure time data to the two-dimensional image sensor 11 and transmits amplification degree data to the amplifier 13 based on the drive control signal change command acquired from the attitude calculation unit 5. The two-dimensional imaging device 11 captures an image based on the exposure time data acquired from the drive circuit unit 15b, and the amplifier 13 amplifies the captured image based on the amplification degree data acquired from the drive circuit unit 15b. Thereby, the image input unit 1 can capture an image again in a state where the exposure time is shortened and the amplification degree is lowered. In order to remove the influence of sunlight, it is necessary to change the exposure time and the amplification degree by orders of magnitude, and the drive circuit unit 15b has a function capable of controlling the exposure time and the amplification degree over such a wide range. .
[0031]
In this way, by changing the exposure time and the amplification degree, even when the sunlight is within the field of view of the two-dimensional image sensor 11, it is equivalent to the level of star density when not affected by sunlight. The level of sunlight intensity can be reduced to the level of, and the position of the sun can be detected by a method equivalent to the detection of the position of a star.
[0032]
Although the case where the exposure time data and the amplification data corresponding to the average value of the gray level of the image data are included in the database in the attitude calculation unit 5 has been described, it is also possible to have a database in the drive circuit unit 15b. That is, the attitude calculation unit 5 sends the average value of the gray value or the corresponding code information to the drive circuit unit 15b as a drive control signal change command, and the drive circuit unit 15b outputs exposure time data associated with the drive control signal change command and Amplification data is stored in the database, and exposure time data and amplification data corresponding to the acquired drive control signal change command can be extracted.
[0033]
The image data is acquired again by changing the exposure time and the amplification degree based on the average value of the gray level by the method shown in the third embodiment, but by repeating this operation several times, it is more suitable for image processing. Image data having a lightness level can be acquired. For example, when the average value of the gray values is close to the maximum value and the amount of light received by the two-dimensional image sensor 11 is in a saturated state, the desired gray level is set by repeating the change of the exposure time and the amplification degree several times. The image data possessed can be acquired. In the third embodiment, the case where the exposure time and the amplification degree are changed in order to mitigate the influence of sunlight has been described. However, in the detection of the position of a bright star and the position of a dark star, the level of intensity is also determined. By adjusting, accurate position detection becomes possible.
[0034]
According to the third embodiment, even when sunlight enters the field of view of the two-dimensional image sensor 11, the posture calculation unit 5 determines this state from the image data. Then, by changing the exposure time of the two-dimensional imaging device 11 and the amplification degree of the amplifier 13 via the drive circuit unit 15b, it is possible to acquire sunlight image data with a lowered level of lightness. From the acquired image, the solar position can be detected by a method similar to the method of detecting the position of the star without being affected by sunlight. As a result, it is possible to detect the sun position without using a posture sensor such as a sun sensor, earth sensor, or rate sensor. Further, an ND filter insertion mechanism is not required. Further, even in the position detection of bright stars and the position of dark stars, accurate position detection can be performed by adjusting the level of lightness.
[0035]
Embodiment 4 FIG.
In the third embodiment, the method of detecting the sun position when the sun is in the field of view of the two-dimensional image sensor 11 has been described. In the fourth embodiment, the sun position is detected when the sun does not exist within the field of view of the two-dimensional image sensor 11 but exists within the out-of-field sun interference angle that affects the image captured by the two-dimensional image sensor 11. How to do will be described.
[0036]
FIG. 6 is an explanatory diagram of a solar position detection method according to Embodiment 4 of the present invention. The configuration of the star sensor is the same as that of FIG. 5 described in the third embodiment. In FIG. 6, an imaging range of the two-dimensional imaging device 11 when the two-dimensional imaging device 11 faces the directivity direction 21 is indicated by a visual field 22. Furthermore, the position of the sun is represented by the solar direction 20. In FIG. 6, the sun direction 20 is outside the field of view 22, but is in a position that affects the image captured by the two-dimensional image sensor 11, and this area is represented by a diagonal line as an out-of-field sun interference angle 23.
[0037]
The image captured by the two-dimensional imaging device 11 when the solar direction 20 is within the out-of-view solar interference angle 23 is not as much as when the solar direction 20 is in the visual field 22, but is affected by reflected light (stray light) from the sun. Becomes brighter and the star position cannot be detected. With respect to the captured bright image, the exposure time and the amplification degree are adjusted by the method described in the third embodiment.
[0038]
The image data for which the adjustment of the exposure time and the amplification degree has been completed will show a two-dimensional pattern of shading according to the sun direction 20 within the out-of-view sun interference angle 23. Here, the two-dimensional pattern is a pattern showing a distribution of the lightness that can be obtained by connecting the same lightness level, and an example thereof is shown as the two-dimensional pattern 24 in FIG. Since the sun direction 20 is not within the field of view 22, the position of the sun cannot be determined simply from the maximum value of the light and shade values. It is possible to detect the position of the sun at.
[0039]
The relationship between the sun direction 20 within the out-of-field sun interference angle 23 and the two-dimensional pattern 24 is determined by the design of the optical system in the image input unit 1, and the relationship between the sun direction 20 and the two-dimensional pattern 24 is obtained by collecting data in advance. It can be made into a database as the solar position information and stored in the attitude calculation unit 5. For example, image data obtained by adjusting the exposure time and amplification when the sun is in a predetermined position within the out-of-view sun interference angle 23 is acquired, and the acquired data is stored in the database as one of the solar position information. I can keep it.
[0040]
The attitude calculation unit 5 can know the region where the sun exists from the exposure time data and the amplification degree data commanded to the drive circuit unit 15b as the drive control signal change command. The closer the sun direction 20 is to the directivity direction 21, the brighter the image data is captured, and accordingly, the exposure time data is set shorter, and the amplification degree data is set to a smaller value, so that the intensity suitable for position detection is set. Image data is obtained. Therefore, by dividing the magnitudes of the values set as the exposure time data and the amplification degree data in advance, the sun direction 20 is the field of view 22, the out-of-view sun interference angle 23, and the out-of-view sun interference angle 23. It will be possible to determine which is outside.
[0041]
When it is determined from the set exposure time data and amplification degree data that the sun direction 20 is present within the out-of-view sun interference angle 23, it is stored as sun position information in the actually captured image data and database. The degree of similarity can be obtained by collation with the image data that has been obtained, and the position of the sun within the out-of-view sun interference angle 23 can be detected from the image data that showed the highest degree of similarity. As described above, even when the sun position is detected based on the similarity, the portion with the highest gray level of the two-dimensional pattern 24 can be calculated in a short time by the method shown in the first embodiment, and the position of the obtained position can be calculated. As a result, the time required for detecting the position of the sun can be shortened by comparing the degree of similarity with neighboring image data.
[0042]
The three-axis attitude of the satellite can be determined from the positions of a plurality of stars, and the three-axis attitude of the satellite is not determined only by knowing the sun direction according to the third or fourth embodiment. However, a certain degree of posture prediction calculation is possible based on previous posture information before being affected by sunlight. Further, when the sun goes outside the outside-of-field-of-view sun interference angle 23, the attitude calculation unit 5 can start the star identification process with the attitude information for at least two axes as known by using the obtained solar direction. Therefore, it is possible to identify a star with a smaller amount of processing than when the star identification processing is started from a state where there is no attitude information.
[0043]
According to the fourth embodiment, even when the sun direction 20 exists within the out-of-view sun interference angle 23, image data in each sun direction actually stored in advance as the solar position information in the attitude calculation unit 5 is actually acquired. The sun direction can be detected by collating with the image data. Furthermore, by knowing the sun direction, the effect that the posture can be determined in a short time after the sun goes out of the sun interference field of view can be obtained.
[0044]
【The invention's effect】
As described above, according to the present invention, the coordinate value data of the star is obtained in accordance with the speed at which the image data acquired by the image input unit is stored in the image memory, and only the data in the vicinity of the coordinate value data is obtained from the image memory. The star sensor can be obtained in which the accurate coordinates of the star can be calculated in a short time and the image processing time for calculating the attitude of the satellite is shortened.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a star sensor according to Embodiment 1 of the present invention.
FIG. 2 is an explanatory diagram for detecting stellar coordinate value data from image data for one row of the CCD according to the first embodiment of the present invention;
FIG. 3 is a flowchart of an algorithm for detecting the coordinate value data of a star by the coordinate information detection unit according to the first embodiment of the present invention.
FIG. 4 is a block diagram showing a configuration of a star sensor according to Embodiment 2 of the present invention.
FIG. 5 is a block diagram showing a configuration of a star sensor according to Embodiment 3 of the present invention.
FIG. 6 is an explanatory diagram of a solar position detection method according to Embodiment 4 of the present invention.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Image input part, 2 Image memory, 3 Coordinate information detection part, 4 Information memory, 5 Posture calculation part, 11 Two-dimensional image sensor, 12 A / D converter, 13 Amplifier, 15a, 15b Drive circuit part, 20 Sun direction , 21 Directional direction, 22 field of view, 23 out of field sun interference angle, 24 two-dimensional pattern.

Claims (4)

An image input unit that is loaded on the satellite and captures the star image,
An image memory for storing image data captured by the image input unit;
A coordinate information detection unit for sequentially detecting coordinate value data corresponding to the maximum value among the continuous image data at a predetermined threshold value or higher from the image data sequentially sent from the image input unit;
An information memory for sequentially storing the coordinate value data detected by the coordinate information detector;
Information on stellar distribution is stored in a database, coordinate value data stored in the information memory is sequentially extracted, image data in the vicinity of the coordinate value data is acquired from the image memory, and The coordinates of the star are calculated by detecting the maximum value from the coordinates, and the attitude of the satellite on which the image input unit is mounted is calculated by comparing the calculated coordinates of the star with the information on the distribution of the stars in the database. A star sensor comprising: a posture calculating unit that performs The star sensor according to claim 1,
A drive circuit unit that outputs a variable drive signal for performing synchronous control with a transfer clock signal variable to each of the image input unit, the image memory, the coordinate information detection unit, and the information memory. ,
The image input unit captures a star image based on a variable drive signal from the drive circuit unit,
The image memory stores image data captured by the image input unit based on a variable drive signal from the drive circuit unit,
The coordinate information detection unit detects the coordinate value data based on a variable drive signal from the drive circuit unit,
The star memory, wherein the information memory sequentially stores the coordinate value data detected by the coordinate information detection unit based on a variable drive signal from the drive circuit unit. The star sensor according to claim 1,
A drive circuit unit that outputs a drive control signal to the image input unit based on a command from the attitude calculation unit;
The posture calculation unit has information of exposure time data and amplification degree data corresponding to the average value of the gray level in a database, calculates an average value of the gray level based on the image data acquired from the image memory, and calculates the gray level Information on exposure time data and amplification degree data corresponding to the average value of degrees is extracted from the database, and a drive control signal change command including information on the exposure time data and amplification degree data is output to the drive circuit unit,
The image input unit acquires exposure time data and amplification data as drive control signals from the drive circuit unit, captures image data based on the exposure time data, and captures based on the amplification data A star sensor characterized by acquiring image data whose intensity is changed by amplifying image data. The star sensor according to claim 3,
The posture calculation unit has, in a database, solar position information in which image data based on sunlight stray light and the position of the sun are associated in advance, and the exposure time data and the amplification degree data included in the drive control signal change command If it is determined that the position of the sun is out of the field of view but within the sun interference angle affected by sunlight, the image data obtained from the stray light of the sunlight acquired from the image memory, and the database A star sensor that detects a solar direction by collating with the solar position information.

Images