JPS63304113A - Star sensor - Google Patents

Abstract

PURPOSE:To suppress in a small way an error at the time of conversion, and to calculate with high accuracy the incident direction of star light beams by amplifying a signal of a low level, and a signal of a high level, by a high gain and a low gain, respectively, and thereafter, bringing them to an A/D conversion, and executing a calculation processing containing a correction of a non-linear amplification effect. CONSTITUTION:A signal level of the peripheral part of a star image is small, therefore, in order to execute a calculation processing, the signal is quantized by an A/D converter 4. Therefore, before it is converted by the converter 4, an output signal from a star light detecting CCD 2 is received and amplified by a non-linear amplifier for amplifying a signal of a low level and a signal of a high level by a high gain and a low gain, respectively. In such a way, a signal of a low level of the periphery of a star image is amplified greatly, and an error which enters at the time of A/D conversion is suppressed in a small way. This A/D-converted signal is inputted to a star light incident direction calculating means, a calculation processing containing a correction for returning a non-linear effect given by the non-linear amplifier 9 to the original state is executed, and the incident direction of the star light beams is calculated.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to an improvement of a star sensor mounted on a triaxially stabilized artificial satellite and used for detecting the direction of incidence of starlight and for use in attitude control, etc.

(Prior Art) A star sensor is mounted on, for example, an artificial satellite, and determines the attitude of the artificial satellite with high precision by measuring the direction of incidence of starlight. It is often used as an element that plays a very important role in controlling the posture of the body.

A conventional example of such a star sensor will be explained. FIG. 9 is a configuration diagram of a conventional star sensor.

1 is an optical system, 2 is a CCD (Charged Coup)
ledDevice (charge coupled device), 3 is a signal processing unit that processes the output of the ccD and outputs data representing the direction of the star. 4 is an A/D converter, 5 is a calculation processing section, and the signal processing section 3 is composed of both.

The starlight that enters the optical system 1 forms an image on the CCD 2. The size and diameter of this star image is generally several times the size of a C0D2 pixel (one pixel cell is simply called a pixel). It will be adjusted so that

This situation is shown in the star map on the CCD in Figure 2.

In Fig. 2, 6 is a CCD pixel, and 7 is a star image formed by the optical system. 8 indicates a pixel divided for the sake of explanation. In this case, the signal level excited on the CCD 2 by one star has a form as shown in FIG.

FIG. 3 is a COD output intensity distribution diagram, which shows the intensity distribution for the pixel 8 surrounded by the bold line in FIG.

For the signal of the intensity distribution of the star image obtained as above, S is the signal level of each pixel, and X and Y
If l is the coordinate of each pixel, the coordinates XC and yc of the center position of the star image are xc=river general L,,,
, (1) ΣS1.

Regarding the above prior art, r 35-THC0NGRE
SSOF TIIE INTERNATIONAL
ASTRONAUTICAL FEDERATI
ONPREPRINT ” DEVELOPMENT O
F MICRO-PROCESSOR CONTROLL
ED CCD 5 TART RACKER FORUS
E ON ASTRO-C''.

The signal processing unit 3 performs calculations shown by equations (1) and (2) on this signal to calculate the center position of the star image. The center position of this star image is data representing the direction of the star.

To make the star image a few pixels in size as mentioned above,
This is to obtain village direction data with higher accuracy by processing according to equations (1) and (2) shown here.

If the star image is made smaller and focused within one pixel, the direction of incidence of the starlight can be determined without the need for calculation processing, but in this case, the accuracy of the direction of the star cannot be increased beyond the size of one pixel. On the other hand, in the method described above in which the star image is formed into a size that spans many pixels and the center position is determined by calculation, the center position can be determined with a dimensional accuracy smaller than one pixel dimension.

You can find the direction of the star using the method described above, but
As shown in FIG. 3, the intensity distribution of the star image in the output of C0D2 is strong at the center, and the signal level drops sharply at the periphery. This is due to the characteristics of the optical system. In the signal processing section 3, the output of the COD is converted into digital data by the A/D converter 4, and then the calculations of equations (1) and (2) are performed, but since the signal level in the peripheral section is small, Relative quantization errors during A/D conversion of peripheral signals become large. For this reason, the calculation accuracy of the star center position (Xc, YJ) deteriorates.

To solve this problem, improve the intensity distribution of the star image in the output of the CCD 2 to increase the signal level in the peripheral area, or use a larger bit length for the A/D converter 4 to reduce the quantization error. This method has traditionally been used.

(Problem to be solved by the invention) However, in order to improve the intensity distribution of the star image in the output of C0D2 and increase the signal level in the peripheral area, it is necessary to improve the characteristics of the optical system, and the weight increase of,
This will lead to an increase in the production cost of the optical part, and
The use of a D converter with a large bit length also has the disadvantage of increasing weight and cost.

An object of the present invention is to solve the above problems by using an optical system 1.
Utilizes a nonlinear amplifier that has the characteristics of amplifying the output signal of the intensity distribution of the star image obtained by the characteristics with a high gain for low-level signals, and amplifying high-level signals with a low gain. By doing so, the present invention aims to provide a star sensor that can reduce errors in A/D yR conversion and improve calculation accuracy of star directions.

(Means for Solving the Problems) The present invention has the following means configuration to achieve the above object. That is, the star sensor of the present invention is used in an attitude control device for an artificial satellite or the like, and includes: a starlight detection means including a CCD disposed in a predetermined positional relationship with an optical system that takes in incident light from a star; and an output from the starlight detection means. a nonlinear amplifier that receives a starlight detection output signal and amplifies a low level signal with a high gain, and amplifies a high level signal with a low gain; A/
A star sensor comprising: a D converter; and starlight incident direction calculation means for receiving the digital signal and calculating the incident direction of the starlight with the nonlinear effect corrected due to the nonlinear amplification.

(Function) Hereinafter, the function of the star sensor of the present invention having the above means configuration will be described. Incident light from a star is taken in by an optical system and subjected to predetermined optical processing to form an image of the star on the CCD of the starlight detection means. is converted into a signal, and an output signal corresponding to the starlight intensity distribution is taken out from the starlight detection means.

This output signal corresponding to the starlight intensity distribution is given by the characteristics of the lens of the optical system, but the signal level at the periphery of the star image is low, so the signal is output at A for calculation processing.
When performing /D conversion and quantization, the error becomes relatively large compared to the signal portion with a large level. Therefore, before A/D conversion, a low level signal is amplified with a high gain, and a high level signal is amplified with a low gain.The output signal from the starlight detection means is received and amplified by a nonlinear amplifier. In this way, the low-level signals around the star image are greatly amplified, and the A/
To suppress errors introduced during D conversion.

This A/D converted signal is input to the starlight incident direction calculation means, and calculation processing of equations (1) and (2) including correction to restore the nonlinear effect given by the nonlinear amplifier is performed. The direction of incidence of starlight is calculated.

(Example) Hereinafter, an example of the present invention will be described based on the drawings. 1st
The figure is a diagram showing the configuration of an embodiment of the star sensor of the present invention.

In FIG. 1, 1 is an optical system and 2 is a CCD. Starlight is taken in by the optical system 1 and a star image is formed on the CCD 2. 9 is a nonlinear amplifier, 4 is an A/D converter, and 5 is a calculation processing unit that corrects the nonlinear effect caused by the nonlinear amplifier and calculates equations (1) and (2).

The characteristics of the nonlinear amplifier 9 are shown in FIG. 4, and the characteristics of the nonlinear effect correction process performed by the calculation processing section 5 are shown in FIG. 5. In FIG. 4, the horizontal axis represents the input to the nonlinear amplifier 9, and the vertical axis represents the output. The falling eQp+Rp are the input and output saturation values, Q o , and R, respectively.

is a turning point parameter of the nonlinear amplifier 9.

In FIG. 5, the horizontal axis represents the input to the calculation processing unit 5, and the vertical axis represents the output. By performing A/D conversion after performing the correction shown in FIG. 4, quantization errors at low signal levels are reduced. The data obtained in this manner is subjected to calculations according to equations (1) and (2) in the calculation processing section 5, and the direction of incidence of starlight is calculated.

The characteristics and nonlinear effect correction characteristics of the nonlinear amplifier 9 shown here are just examples, and the nonlinear amplifier 9 may have a high gain for low level inputs and a low gain for high level inputs. , the characteristics are not limited to those shown in FIG.

Similarly, regarding the nonlinear effect correction characteristics, if the gain is low for low level inputs and high for high level inputs so as to correct the characteristics of the nonlinear amplifier, then the characteristics shown in Figure 5 are limited. It's not a thing.

In order to actually demonstrate the significance of this method, we will consider only the part of the COD surrounded by the thick line in Figure 2, and use equation (1) from the signal value of the pixel in the thick line. XC
We will evaluate the calculation accuracy of the star image center position in the X direction by calculating only the equation.

As shown in equations (1) and (2), the equation for calculating XC and the equation for calculating yc are independent (the signal distribution in the Y direction has no effect when calculating XC), so such an evaluation without loss of generality.

During the evaluation, the intensity distribution of the star signal was assumed as shown in Figure 6. In FIG. 6, each line on the horizontal axis corresponds to the length of one pixel. The origin is the center of the star image, and the vertical axis represents the signal strength at a certain distance from the center.
The A/D converter 4 has a length of 4 bits.

Shows the deviation between the true star image position and the star image center calculated by computer simulation using equations (1) and (2) when the star center position takes various positions within 1 pixel. Figures 7 and 8 are 6
In both figures, the horizontal axis represents the deviation of the star image center, if any, from the center of the xel, and the vertical axis represents the star image center calculation error when the star image center is at the position indicated on the horizontal axis. Figure 7 shows the error when calculating the center position of a star using the conventional method, and Figure 7 shows the error when calculating the center position of a star using the method of the present invention. It is.

In the example shown here, Qo=0.5. R,=0.8. By comparing both figures, it can be seen that this method improves the accuracy by a maximum of about 3 times or more.

(Effects of the Invention) The star sensor of the present invention has a conventional star sensor that produces small signals around the intensity distribution of a star image.
Considering that the error that occurs during /D conversion and quantization is the cause of poor accuracy in determining the star image center, low-level signals are amplified with high gain using a nonlinear amplifier, and high-level signals are The signal is A/D converted after being amplified with a low gain, and the A/D converted signal is subjected to calculation processing including correction of nonlinear amplification effects.
This has the advantage that the error during conversion can be kept small and the direction of incidence of starlight can be calculated with high accuracy.

[Brief explanation of drawings]

Figure 1 is a configuration diagram of an embodiment of the star sensor of the present invention, Figure 2 is a configuration diagram of an embodiment of the star sensor of the present invention.
The figure is a star image map on the CCD, and Figure 3 is a star signal intensity distribution map.
Figure 4 is a characteristic diagram of the nonlinear amplifier, Figure 5 is a characteristic diagram of nonlinear effect correction, Figure 6 is a star signal intensity distribution diagram assumed in accuracy evaluation, and Figure 7 is a star image center calculation using the star sensor of the present invention. FIG. 8 is a diagram showing the error in star image center calculation by a conventional star sensor, and FIG. 9 is a configuration diagram of the conventional star sensor. 1...Optical system, 2...CCD, 3
... Signal processing section, 4 ... A/D converter, 5 ... Calculation processing section, 6 ...
・CCD vixel, 7...Star image, 8...
...Vixel region used to display the signal strength of the star signal intensity distribution in Figure 3, 9...Nonlinear amplifier. Agent: Yoshi Yahata, Hiroshi Hiroshi's Zutter 7) Suno's Sanskrit Plane Example 7th Edition / Diagram CCD's Star Map No. 2 Diagram E/l Kosuke Fuma No. 3 Diagram No. 5 Illustrated prayer] Hayabusa Funoma who received the star signal

Claims (1)

[Claims] In attitude control devices for artificial satellites, C is placed in a predetermined positional relationship with the optical system that captures incident light from stars.
A starlight detection means including a CD; upon receiving a starlight detection output signal output from the starlight detection means, a low level signal is amplified with a high gain, and a high level signal is amplified with a low gain. a nonlinear amplifier; an A/D converter that receives an output signal from the nonlinear amplifier and converts it into a digital signal of a predetermined number of bits; receives the digital signal;
A star sensor comprising: a starlight incident direction calculation means for calculating an incident direction of the starlight with the nonlinear effect due to the nonlinear amplification corrected;