EP4204765A1

EP4204765A1 - Method, device and computer program product for determining the position of a spacecraft in space

Info

Publication number: EP4204765A1
Application number: EP21769899.2A
Authority: EP
Inventors: Uwe Schmidt; Falk Ahlendorf; Rolf Hartmann
Original assignee: Jena Optronik GmbH
Current assignee: Jena Optronik GmbH
Priority date: 2020-08-31
Filing date: 2021-08-23
Publication date: 2023-07-05
Also published as: DE102020122748B3; US20230331403A1; JP2023539615A; WO2022043247A1

Abstract

The invention relates to a method for determining the position of a spacecraft in space, wherein the following steps are cyclically repeated: capturing distorted star images (4); processing the distorted star images (4) to form distorted star group data (6); storing the distorted star group data (6); determining a current rotation rate (8) by comparing the distorted star group data (6) of two consecutive cycles; transmitting the current rotation rate (8) to a position control system (9); and/or the following steps are carried out: processing the distorted star images (4) of a current cycle to form rectified star group data (11); determining position information (13) by matching the rectified star group data (11) with star group catalog data (14) which is carried along; transmitting the position information (13) to the position control system (9). The invention also relates to a method for determining the position of a spacecraft in space, wherein, on the basis of known system parameters of an optical system, the following steps are carried out: coding star group catalog data with n = 3...4 stars [x _n , y _n , z _n], which are visible in a field of view, into representative focal-plane coordinates; forming a scaling-, translation- and rotation-invariant star group code on the basis of [xPiX,yPiX]n; or the following step is carried out: coding star group catalog data with n = 3...4 stars [x _n , y _n , z _n], which are visible in a field of view, into representative tangent and/or angular coordinates [tan(α),tan(β)]_n. The invention further relates to a device for carrying out such methods and to a computer program product for carrying out such methods.

Description

Method, device and computer program product for determining the attitude of a spacecraft in space

description

The invention relates to methods for determining the attitude of a spacecraft in space. The invention also relates to a device for carrying out such a method. The invention also relates to a computer program product for executing such a method.

Document US 2011/0024571 A1 relates to a system and method for gyroless transmission of sun detection from transfer orbit using only blade current measurement feedback. The transfer orbit serves to transport the spacecraft into its operational orbit. The spacecraft is set in a targeted rotation, with the axis of rotation and rotation speed being optimally determined solely from the measurement of the available solar power. No other sensors such as gyro sensors, star sensors or sun sensors are used in transfer orbit. As a result, this mission phase, which is risky due to the low availability of resources and fatal consequences of incorrect controls, can be made very safe. However, the low accuracy of the yaw rate measurement and the necessary stable rotation do not generally allow gyro sensors to be dispensed with in all mission phases.

Document US 2014/0231589 A1 relates to an attitude estimator that uses the sun sensor outputs as the only attitude determination measurements to provide three-axis attitude information. The underlying task is that, in principle, only two of the three components of the position can be measured with sun sensors. The solution is to estimate the third attitude component using measurements from multiple sun sensors and the missile's moment of inertia. Again, there is the restriction on a stable rotation of the spacecraft and the limited measurement accuracy of sun sensors, which is not sufficient for all mission phases.

Document US 8,380,370 B2 relates to a system and method for controlling a spacecraft to perform a three-axis maneuver based purely on "position" (ie attitude) measurements. The goal is to replace gyro sensors with star sensors replace. The necessary maneuvers of the spacecraft are to be carried out solely with the angular position measurements without using angular rate measurements. In order to achieve this, the angular rates used in the maneuvers must remain below the upper limit of the usability of conventional star sensors of a few angular degrees per second. For this purpose, the Control of the spacecraft - called Sun Catch Mode, Safe Mode and Normal Mode - generates turn commands consisting of successive phases of specific rotations.The turn commands are defined using a gimbal model of rotation in fractional rotations about each Euler angle turn rates in each phase held. Under these circumstances, the readings from conventional star sensors are sufficient to determine the commands with the respective phases. The command generation contains protective measures against the failure of the star sensors involved. During a sortie, command phases are aborted and no turns are performed. After restoring the star sensor measurement, an extra phase is inserted, which restores the state before the failure with a specified rotation rate. The given solution cannot completely replace the gyro sensors, since it does not work at high rotation rates. For this reason, it is only intended to be used as an emergency solution in the event that data from the gyro sensors that are in principle available are not available.

The document US 7,410,130 B2 relates to a method for determining the attitude of a spinning spacecraft. It is about the use of star trackers to control rotation-stabilized spacecraft without the goal of completely replacing gyro sensors. A rotation rate of 3 angular degrees per second is specified as the upper limit for the usability of conventional star trackers. In order to in case of failure of the spacecraft's engines still under this To stay within the limit, only rotation rates in the range of 0.3 to 1.5 angular degrees per second are used for rotation stabilization, although higher rotation rates would result in better stability. In the event of unforeseen errors, for example if the spacecraft's engine not only fails but also accelerates impermissibly due to errors in the control, rotation rates of up to 20 angular degrees per second occur. This value is well above the application limit of conventional star sensors and has so far excluded them from being used as emergency sensors instead of gyro sensors. According to US Pat. No. 7,410,130 B2, the spacecraft's attitude measurement is initially initialized using star sensor data. For the initialization of the attitude measurement, the spacecraft must be put into a stable rotation with a rate of rotation below the operating limit of conventional star sensors. A large number of sensor and control components go into the attitude estimation following the initialization: earth sensors, sun sensors, gyro sensors, solar current sensors, Kalman filters, attitude controllers and ground-based attitude measurements. By linking the large number of components, it should be achieved that there does not have to be two different variants of the sensors for the transfer orbit and for the deployment orbit. For example, a rotating sun sensor has been used for transfer orbit and a wide-angle sun sensor for deployment orbit. This leads to an advantageous reduction in mass, costs and probability of failure. If the position information is lost, however, the method from document US Pat. No. 7,410,130 B2 requires a new initialization to be carried out by the star sensor in connection with the corresponding stabilization of the rotation. The star camera is also not available as a sensor for this stabilization; an additional gyro sensor must be used for measurements above the application limit of conventional star sensors.

Document US 9,073,648 B2 relates to an attitude estimator using star tracking measurements and enhanced Kalman filtering with or without attitude data to provide three-axis rate estimates. The yaw rate can also be estimated directly from the star positions without determining the yaw position. Estimating the rate of rotation from stellar positions is described here as very unreliable unless appropriate estimation algorithms are used will. The estimation algorithms proposed in document US 9,073,648 B2 include Kalman filters and forward and backward averaging over a time interval. Only after this improvement can the filtered rotation rates be used to calculate the rotation position and to compensate for the failures of gyro sensors. The unreliability of the angular rate measurement from star positions without position determination results from the fact that the star positions have to be tracked in the image sequence of the star camera. The tracking consists of anticipating the coming star position and finding the star again near the predicted position. At high spin rates and spin accelerations, it is easy to confuse the tracked star. Therefore, in the position estimator according to document US Pat. No. 9,073,648 B2, in addition to Kalman and mean filtering, a step of assigning star pairs is also provided. This assignment takes place without identifying the stars and without calculating the position. The success of the method disclosed in US Pat. No. 9,073,648 B2 depends on the inclusion of many stars and the consideration of spacecraft attitude control parameters. The improved reliability of the angular rate estimation is not sufficient to generally save on gyro sensors on the spacecraft. According to US Pat. No. 9,073,648 B2, the angular rate information from the star sensor can only be used for gyro sensor calibration and for limited functions in the event of unexpected gyro sensor failure.

The publication WHITE PAPER "Global Shutter, Rolling Shutter - Functionality and characteristics of two exposure methods (shutter variants)", Dominik Lappenküper, www.baslerweb.com, 2018, deals with the characterization of the global shutter and the rolling shutter for exposure control with modern CMOS cameras. With the global shutter, all cells of the matrix are exposed once at the same time, whereas the roller shutter exposes the individual image lines one after the other line by line. Depending on the exposure time, overlapping can occur with the rolling shutter if the exposure of the next line occurs before the Completion of the exposure of the previous image line starts Global shutter cameras require 2 to 3 times as many transistors on the sensor chip for control and Charge transport like cameras with roll-top closure. This results in comparatively high image noise, more heat and a high susceptibility to the radiation loads in orbit during the global shutter.

[0008] A disadvantage of the rolling shutter, especially on fast-moving platforms, is image distortion, which can exceed the level acceptable for star identification. The advantages of the roll-top closure outweigh the disadvantages for use in space. A rolling shutter is also a prerequisite for special optimizations to increase image quality. A method for preventing overexposure in CMOS cameras is specified in the document US 2006/0238632 A1. When used with star sensors, overexposure occurs when very bright objects, such as sun reflections from parts of the spacecraft, lead to an overexposure in which not all photons can be converted into measurable electrons. The excess electrons then spill over into neighboring cells and rows, causing false readings. The solution according to document US 2006/0238632 A1 prevents overflow by special, multiple resetting of the image lines, which is only possible with a rolling shutter.

[0009] Document US Pat. No. 9,503,653 B2 relates to a method for determining the position of a star sensor based on a rolling shutter image. The strong restriction of the dynamics of the spacecraft to the low rate of rotation for tolerable image distortions due to the roller shutter is described in document US Pat. No. 9,503,653 B2 as the most important barrier to the further development of conventional star sensors. According to document US Pat. No. 9,503,653 B2, high rotation rates are the decisive bottleneck for high-resolution earth observation missions. According to US Pat. No. 9,503,653 B2, the rotational position measurement for star cameras with a rolling shutter under conditions of high dynamics is to be improved in a three-stage process. The star sensor is in window mode, in which not the entire image is read out, but only the relevant image windows around the predicted star positions. The three stages run in the sequence of extraction of the individual navigation stars and not in the usual star sensor cycle, which results from the evaluation of the entire field of view. the The first step involves optimizing the time parameters of the rolling shutter for the extraction of the next navigation star, i.e. optimizing the exposure time, the readout time, the line transition time and the processing time. In the second step, the position of the next navigation star from the star catalog is estimated in the sensor matrix for the optimized parameters and the constellation is extracted there. In the third, final step, the angular position and angular rate are recursively re-estimated using the new star and the previous estimate. The optimization of the time parameters of the rolling shutter serves, among other things, to maintain a constant star sensor cycle for compatibility with conventional star sensors and to clearly define the point in time of the next star extraction. The advantages of the solution according to document US Pat. No. 9,503,653 B2 are considered to be that the measured values are estimated and output an order of magnitude faster with the single star cycle than with the star sensor cycle. In addition, shorter star sensor cycles should be achieved, since the good predictability of the star windows saves the exposure and reading of the entire image matrix. The method described in document US Pat. No. 9,503,653 B2 only provides updates of angular position and angular rate and cannot be used for the first acquisition if initial values for angular position and angular rate are missing. Another disadvantage is the dynamic setting of the time parameters of the rolling shutter. This setting is not possible with all cameras and complicates the camera setup.

So-called autonomous star sensors have been in use for many years. They are the further development of the star tracker (English Star Tracker). Autonomous star sensors not only track the stars in the image sequence of the star camera, they also determine the absolute rotational position of the star sensor from the star positions. In the case of star trackers, the attitude calculation is carried out in the satellite's mainframe and not in the sensor. The identification of the tracked stars is necessary for the calculation of the orientation. Star identification means that the stars in the star camera image are clearly assigned to the stars on a global star map. The star map is carried in the sensor in the form of a star catalogue. Modern autonomous star sensors can perform star identification without any prior information, such as the camera's approximate line of sight, determine. This ability is called "Lost in Space" identification. The methods of "Lost in Space" identification essentially determine the quality of a star sensor and are therefore the subject of permanent further development. The study "A Survey on Star Identification Algorithms", Benjamin B. Spratling et al. Algorithms 2009, 2, 93-107, gives a good overview of the changing methods of identification. According to this study, the first autonomous star sensors need about the same number Search steps for the identification of a star, as there are stars in the star catalogue. The consequence of this is that with such star sensors, the initial acquisition of the position cannot take place in the same rhythm as the image acquisition. The star sensor cycle of the image acquisition is usually between 1/5 seconds and 1/16 seconds. The first acquisition, on the other hand, takes an order of magnitude longer. In the mid-1990s, there was a decisive breakthrough in improving the identification algorithms. For the search for the star to be identified or a star group in the star catalogue, methods of graph theory with index trees were used instead of sequential algorithms used for searching.The star catalog contains now additionally an index tree and possibly a catalog of the asterisms, which leads to an increase in memory requirements. On the other hand, the number of search steps in the catalog is reduced dramatically. The number of steps is no longer proportional to the catalog length, typically around 5000, but corresponds to the binary logarithm of the catalog length, i.e. around 12 search steps. With this improvement, it becomes possible to perform "Lost in Space" star identification in each cycle of image acquisition.

[001 1 ] Document US Pat. No. 5,935,195 A describes a variant of star identification with index trees. According to document US Pat. No. 5,935,195 A, star groups consisting of 3 stars are formed from the stars in the camera image. Each asterism is then characterized by at least two angular distances between stars in the group. The angular distances are chosen using geometric ordering criteria in such a way that they clearly identify the star group. Using the same procedure, asterisms are formed from the stars in the star catalogue. The star groups are used as a database together with the characterizing angular distances in the star sensor as an extension of the star catalog saved. The selection of the stars for a group can be done according to the geometric aspects of the neighborhood and according to the brightness of the stars. According to document US Pat. No. 5,935,195 A, a binary index tree is used to search for the appropriate star group in the database.

As from the publication "Making the Sky Searchable: Fast Geometric Hashing for Automated Astrometry", Sam Roweis, Dustin Lang & Keir Mierle, University of Toronto, David Hogg & Michael Blanton, New York University, http://astrometry. net, 2005-2019, shows that vector index trees are used for the search in more recent developments.In addition to accelerating the search in the catalogue, there are approaches to simpler and more robust methods of calculating the star group from the image of the star camera.The calculation effort for the parameters of the star groups from the image is proportional to the doubled number of stars in the image in document US Pat. No. 5,935,195 A, and is therefore typically around 100 steps.

According to the study "A Survey on Star Identification Algorithms", Benjamin B. Spratling et al. Algorithms 2009, 2, 93-107, the calculation of the star groups was based on the number of stars in the group, i. 5. The study "A Survey on Star Identification Algorithms", Benjamin B. Spratling et al For this purpose, parameters of star groups are used that are independent of the imaging scale of the camera. For example, instead of the angles between the stars of a 3-star group, the inner angles of the triangle spanned by the 3 stars are used. For a 4-star group, the publication "Making the Sky Searchable: Fast Geometric Hashing for Automated Astrometry", Sam Roweis, Dustin Lang & Keir Mierle, University of Toronto, David Hogg & Michael Blanton, New York University, http://a According to stromemetry.net, 2005-2019, a normalization of the stellar separation angles to the largest angular separation of the group was applied to achieve independence from the imaging scale.

In the international project of astronomers "Astrometry.net" a method was developed which any images of the starry sky without knowledge of the Parameters of the imaging system can be identified. The method described in the source encodes the entire starry sky up to magnitude 22mag in (scale, translation and rotation) invariant star groups of four. With sufficiently high computing power, constellations from any source are possible without system knowledge of the recording system.

The star four-group catalogs used in this method, as well as the computing power required, exceed the resources of a star sensor by several orders of magnitude. This is due to the fact that the above method works without knowledge of the parameters (field of view, sensitivity, etc.) of the imaging system.

Star sensors work on the principle of detecting point objects in the raw image data (by means of detection threshold), discriminating (star-like object against pixel defects), following (tracking), interpolating, transforming (pixel coordinates -> angle coordinates), identifying (against a star catalogue) and finally to calculate the 3-axis position. This process of the initial acquisition of the position can take several seconds and is very uncertain in the case of high movement dynamics (rotation rate, acceleration). By using a system-adapted star group catalog coding using the known star sensor properties (optics, detector), extremely short star identification times (real time!) can be achieved in connection with efficient on-board star catalogues.

The object of the invention is to improve the method mentioned at the outset. In addition, the invention is based on the object of providing a device as mentioned at the outset. In addition, the invention is based on the object of providing a computer program product as mentioned at the outset.

In particular, the invention is based on the object of making gyro sensors in the attitude determination system of highly dynamic spacecraft completely replaceable by improvements to the star sensors. For this purpose, the limitations in the use of star cameras caused by image distortion are to be overcome with a roll-up shutter and angular position and angular rate measurements far be made possible beyond the previous rotation rate limit. In particular, the rotation rate limit is to be improved to such an extent that a gyro-free measurement of rotation position and rotation rate is possible for all essential applications, including emergency situations. The solution according to the invention should not require any intervention in the hardware of available star cameras, in particular not in their sensor-related control and pre-processing electronics, so that existing cameras can continue to be used. The yaw rate limit should not only be improved for yaw rate measurements with the aim of replacing gyro sensors, but also for measuring the absolute yaw rate at very high yaw rates. The solution according to the invention should manage without additional delays with the minimum number of star sensor cycles, i.e. with one cycle in the continuation of the measurement in a continuous sequence and with one or two cycles under "lost in space" conditions.

In particular, the invention is based on the object of a process "discriminate ... track ... interpolate ... transform ... identify ... calculate position", which takes a few seconds, typically about 10s, into one To implement a real-time process with, for example, <100ms at a 10Hz measurement cycle. Real-time star identification does not require interference-prone star tracking ("tracking"), especially at high rotation rates and accelerations. In each individual measurement cycle, independent position measurements are possible, which then also enable real-time determination of rotation rate and acceleration in subsequent cycles. Especially at very high rotation rates and accelerations, a possibility can be provided to determine the rotation direction and a rate estimate with a star sensor, precisely when suitable star groups can be detected in at least two consecutive cycles (bright star fields), but not necessarily in every cycle must be present (when sweeping over dark star fields). A classically implemented position tracking (attitude tracking) at high accelerations and yaw rates can be supported by a parallel real-time identification with the method described. [0020] The object is achieved with a method having the features of claim 1. The task is also solved with a method having the features of claim 2. The task is also solved with a device having the features of claim 12. The task is also solved with a computer program product with the features of claim 15. Advantageous embodiments and/or Developments are the subject of the dependent claims.

The method can be performed without using gyroscopes or gyroscopes. In this respect, the method or the device can also be referred to as gyroless. The spacecraft can be a highly dynamic spacecraft.

[0022] A basic aspect of the invention can be that image distortions caused by a roller shutter are corrected to such an extent that star identification is also possible for very high rotation rates.

[0023] An availability of a star identification in one sensor cycle for an entire image with modern search methods can then mean that no window mode is required for star tracking anymore. This can prevent star sensors from failing at high rotation rates and in particular at high rotation accelerations due to an uncertain prediction of the window position. For an identification that is carried out again and again, the need for a prediction of a next star position can be omitted. It always works as long as an image geometry is still good enough to allow identification.

[0024] In order to correct an image distortion caused by a roll shutter, on the one hand a rotation rate is required for a roll correction and on the other hand an identification and thus a correction is required for a rotation rate. This conflict of objectives can be resolved in that the identification is not carried out immediately with the geometrically correct star groups in a catalogue, but first in an intermediate step with the star groups distorted according to a current rate of rotation. In the intermediate step, the roll-warped star groups from the current image can be found again in a set of the roll-warped star groups of the last image. This is less sensitive to yaw rates and spins than catalog identification because only the change in distortion in one star tracker cycle is relevant and not the total distortion.

[0026] The current rate of rotation can then be initially estimated sufficiently well from the roll-distorted groups of stars found again in two consecutive images. A location cannot be determined from this mapping. This initial rate of rotation is good enough for correcting the roll distortion in the current distorted star groups to identify the groups in the catalog. With this identification in relation to the catalog data stored in inertial coordinates, the final determination of the angular position and fine angular rate of the star sensor can then be made.

With knowledge of the system parameters, in particular focal length, pixel size, number of pixels of a star sensor or detector, a star group coding can be carried out. Catalog star groups with n = 3 ... 4 stars [x _n ,y _n , z _n ], which can be seen simultaneously in one field of view, are encoded in representative focal plane coordinates, in particular in pixel coordinates. Due to the knowledge of the system, there is no need for further simultaneously visible stars. Based on [x _pix , y _P tx ] _n scale, translation and rotation invariant star group codes can be formed. With this coding, the coordinates supplied by the detector can be used directly without further processing. Alternatively or additionally, asterism catalog data with n = 3 ... 4 stars [x _n ,y _n , z _n ] visible in a tile can be converted to representative tangent and/or angular coordinates [tan(a) ,tan( /?)] _n are encoded.

In particular, only the brightest stars are selected for catalog asterism formation. Using the brightest stars reduces the amount of stars and increases the significance/tolerance of the asterisms. Measurement inaccuracies can thus be tolerated and used to advantage. The measured star groups can be erroneous with regard to a single star position due to: measurement noise with small signals; Rolling Shutter (roller shutter) Distortion in motion dynamics. By combining pixels (binning), defectivity can be increased at very high rotation rates and thus star position uncertainties can be counteracted. Asterisks can be encoded in a binary search tree. A search subject to tolerances can be carried out in the search tree. The identification solution found can be verified with other available stars.

Known properties of a calibrated star sensor, such as optic focal length, detector pixel size and number, can be included in the star group encoding. By arranging star group codes in preferably binary search trees, very fast star group identifications are possible.

Star identifications can be made possible in real time (<100ms) in each individual and independent measurement cycle.

As soon as a star group (3...4 stars) is detected, an identification and position calculation can take place. If this takes place in directly consecutive cycles or with known time intervals, the yaw rate, direction and acceleration or a rate estimation can be carried out even for very high yaw rates. Conventional position tracking can be supported in particular in the case of high rates and accelerations, which leads to an increase in the robustness of the position measurement.

The asterism catalog data may include asterism, asterism, and/or vector index tree data. The star group data may include identification vectors and/or reference data. The vector index tree can refer to the identification vectors of the asterisms. Further star catalog data with additional star data can be carried and used. The other star catalog data can be used for expansion. The position information can be statistically filtered. The position information can be filtered by averaging and/or by Kalman filtering. At least one star camera, in particular a roll-up shutter camera, can be used to record the distorted constellations. The distorted constellations can use the at least one star camera to the distorted star group data are processed. The at least one star camera can have picture elements. Several adjacent picture elements can be combined to form a picture element module in order to increase a rotation rate limit. One star camera or multiple star cameras combined can be used to capture different fields of view. The method may be performed using at least one separate processor device and/or using a processor device on the spacecraft.

The processing blocks can be logically, functionally and/or structurally delimited. The processing blocks can be connected to one another in a signal-transmitting manner. The device can have several star cameras and a separate processor device for each star camera. The device can have several star cameras and a common processor device for a group of star cameras. The at least one processor device may be a separate processor device for performing the method or a processor device of the spacecraft. The device can have at least one program memory and/or at least one data memory.

The computer program product can be present on a computer-readable storage medium, on a computer-readable data carrier or as a data carrier signal.

Exemplary embodiments of the invention are described in more detail below with reference to figures, which show schematically and by way of example:

1 is a block diagram of a gyroless attitude determination system for highly dynamic spacecraft.

2 shows a block diagram of a gyroless attitude determination system for highly dynamic spacecraft with exclusive rotation rate measurement,

3 is a block diagram of the attitude determination system for low dynamic spacecraft. 4 is a block diagram of a gyroless attitude determination system for highly dynamic spacecraft with extra accurate attitude measurement,

Fig. 5 groups of stars distorted by a roller shutter,

Fig. 6 identification parameters in star group data,

Fig. 7 is a flow chart of processing in "lost in space" conditions,

8 shows an application of a robust determination of rotation rate and attitude with a star sensor using selected star group catalog coding and

9 shows a flow chart of a robust determination of rotation rate and position with a star sensor using selected star group catalog coding.

Figure 1 shows a block diagram of a gyroless attitude determination system 1 for high dynamic spacecraft. A star camera 2 embodied as a roll-up shutter camera of the gyroscope-less attitude determination system 1 records distorted star images 4 of the starry sky in time with the star sensor cycle and transfers them to a processor device 3 for evaluation. There, in a processing block 5 for star group generation, star vectors are determined from the star images 4 which characterize the position of all the stars in the image and which are further processed into star group data 6 . Similar to the usual constellations, asterisms consist of three, four or more stars and can be derived according to different principles. What is decisive is the property of the asterisms that they can be found unequivocally in the entire celestial sky with the help of their asterism parameters. The asterism data 6 is, like the image, distorted by scrolling and cannot yet be safely used for searching in asterism catalog data 14 . They can, however, be identified in the amount of asterism data from the previous cycle, which is also distorted. Therefore, the distorted asterisk data 6 of the current image is latched for the next cycle as the distorted asterisk data of the previous cycle. If it is the very first cycle, i.e. “Lost in Space” conditions are present, there is still no distorted star group data from the previous cycle. In this case, a current rotation rate 8 cannot be calculated immediately in processing block 7 will. In this case, a yaw rate of zero angular degrees per second is used as a fixed initial value. In subsequent cycles after the initialization, the current yaw rate can always be estimated in the processing block for generating the current yaw rate 8 . For this purpose, at least one star group from the current image is found in the previous image. The current rotation rate can be determined from the associated distorted star groups using various methods. With some variants of asterisms, a direct calculation of the rotation rate from the asterism parameters is possible. It is always possible to use the delta quaternion estimation through statistical optimization of the rotation matrix, which brings the two sets of star vectors of the associated star groups into congruence as well as possible. The well-known QUEST algorithm can be used for this optimization.

The current rate of rotation 8 determined in this way is used on the one hand for the geometric correction of the distorted star group data 6 in the processing block 10 for roll shutter correction and on the other hand is output directly to the attitude control system 9 of the spacecraft. From the second cycle onwards, the current rate of rotation 8 can also be reliably measured for very high rates of rotation with an unstable axis of rotation and is available to the spacecraft for applications with high rates of rotation and low demands on accuracy. An important such application is the emergency with a rapidly tumbling spacecraft, which can be stabilized again with the help of the current rotation rate 8.

The processing block 10 for correcting the roller shutter converts the star position of the stars of the asterisk data 6 into a coordinate system of the star sensor in such a way that they would occur in the case of a snapshot without the roller shutter. The conversion takes place in relation to a well-defined reference time that can be freely selected in the star sensor cycle. For example, the middle of the exposure time of the first image line can be chosen as the reference time. With the help of a model of the roller shutter, depending on the number of the image line of the star, the time of its recording in relation to the reference time can be determined as a time difference. Stars in the first line of the image have a difference of zero at the reference time mentioned. From the current rotation rate 8 results for each star of the groups a rotation of its star vector, which puts the star vector in the position it had at the reference time. Stars from the first image line are not rotated at all, the stars from the last image line are subjected to the maximum rotation corresponding to the maximum time difference. The resulting corrected asterism data 1 1 can now be successfully found in the asterism catalog data 14. This second assignment of star groups takes place in processing block 12 to generate the final position information. As a result of the assignment, star vectors are then available both in the coordinate system of the star sensor at the reference time and in the inertial coordinate system of the catalogue. This enables the calculation of the position information 13 consisting of the angular position and a fine angular rate using known methods. In this case, it is possible to calculate the position information 13 directly from the star group parameters or to use the associated star vectors. Depending on the actual current yaw rate, under "Lost in Space" conditions, the catalog identification can also function in the very first cycle without a measured current yaw rate. This means that the yaw position part of the length information is available from the first or second cycle. For turns below the previous yaw rate limit from conventional star sensors, the angular position is immediately available, so that known disadvantages can be avoided.

The gyroless attitude determination system can be adapted to specific mission requirements, with the required resources being optimized in accordance with the adaptation.

Fig. 2 shows a block diagram of a gyroless attitude determination system 1a for highly dynamic spacecraft with exclusive rotation rate measurement. All components of the system that are not required for the measurement of the current rotation rate 8 can be left out here. The remaining components do not require any modification, so that a high degree of modularity of the system is achieved.

Figure 3 shows a block diagram of the attitude determination system 1b for low dynamic spacecraft. The system components for determining the current rotation rate 8 and for the processing block 10 are omitted here Roll closure correction, also in a modular way. If the system is to be upgraded for higher requirements, corresponding extensions are possible, as shown in FIG. 4 for the case of the gyroless attitude determination system for highly dynamic spacecraft with extra accurate attitude measurement.

[0041] The measurement accuracy can be increased by using additional star catalog data 15. The accuracy of both the rotational position measurement and the rotational rate measurement improves with the number of associated stars. The number of stars assigned is no longer limited by the capacity of tracking stars in image gates, as with conventional star sensors. It can potentially be extended to all detectable stars in the image. This increases the number of stars measured, for example to around 50 to 100 compared to 16 stars tracked in windows. This leads to a significant increase in measurement accuracy.

Figure 5 shows two examples of star clusters distorted by the roll closure. The undistorted star group 17 is also shown for comparison. The undistorted asterism 17 consists of the three stars 16. The asterism 19 with the stars 18 demonstrates the distortion by a rotation around the optical axis of the star camera, while the asterism 21 with the stars 20 the distortion by a rotation around the vertical image axis of the star camera shows. The imaging ratios correspond to a typical case with a square star camera image field 22 of 25 angular degrees, a rolling shutter delay of 100 milliseconds between the first and the last image line in an 8 Hz star sensor cycle. The first line of the image is in the image above. Assuming the size of the image matrix is 1000 pixels by 1000 pixels, the last image line has a distortion of around 80 pixels for a rotation around the image axis and around 34 pixels for a rotation around the optical axis of the star camera, each with a rotation rate of 20 degrees per second. Figure 5 reflects the geometric proportions of the distortion for these cases. Depending on the rate of rotation and the axis of rotation, there are corrections for the star positions, which are only compensated for with the rotation corresponding to all three components of the current rate of rotation 8 as executed in processing block 10 for. Pure shifts are not enough.

Four stars are preferably used for the star groups. When using three stars per group, there is theoretically already a clear identification of the group, but this requires very small errors in determining the star position, which cannot always be guaranteed in practice. The asterisk data contains the identification parameters, summarized in an identification vector. In addition, they can contain reference data which, after successful identification, are used to calculate the angular position and angular rate. For large sets of asterism data, such as asterism catalog data 14, additional index trees may be included to speed up searches.

Figure 6 shows identification parameters in asterisk data. For stellar calculations, preferably, unit sphere data with unit vectors and solid angles are used. The star group considered consists of four stars 23, whose position is represented as a direction by the corresponding star vector on the unit sphere. The star vectors of the stars in the image of the star camera are determined by the position of the star in the image and the optical imaging parameters, especially the focal length and possibly other calibration parameters. In the example shown in Fig. 6, the identification parameters are calculated as follows: First, the two stars with the largest angular separation are selected, they are called primaries. The main spherical axis 24 of the asterism is the segment of a circle on the unit sphere that connects the two stars with the greatest angle of separation. The length of the main axis corresponds to the maximum angle of separation. The unit vector to the center of the major axis is used as the position vector 25 of the group. The spherical minor axis is perpendicular to the major axis and has its origin in the position vector of the group. With these two axes, the angular coordinates of the two remaining stars of the group, called secondary stars, are then defined in a local group coordinate system. The group coordinate system is two-dimensional with the angular coordinates perpendicular to the main axis 26 and the angular coordinates parallel to the main axis 27. In order to make the group identification independent of the image scale of the star camera, or to carry out a post-calibration of the focal length in orbit, the angular coordinates of the secondary stars can also be normalized to the angular size of the main axis . Changes in scale, for example due to thermal effects in the lens, can then be compensated for after identification has taken place by evaluating the non-standardized angles. The variant of identification with the angles normalized to the main axis, which is independent of the imaging scale, is to be used.

The four values of the angular coordinates (principal axes 26, 27) of the two secondary stars form the four-dimensional identification vector of the asterism. The indices of the four group stars in the index list of stars detected in the image are the reference data. The global asterism catalog with the asterism catalog data 14 is calculated using the same method on the ground and carried in flight. In the asterism catalog data 14, the indices of the four stars correspond to a continuous numbering of all group stars. The index tree, which is also calculated on the ground, is preferably designed as a vector index tree for the specified identification vector. For the variant of the gyroless attitude determination system 1c for highly dynamic spacecraft with extra-precise attitude measurement, additional star catalog data 15 are carried along in addition to the star group catalog data 14 .

For star detection with windows, the typical star sensor is not suitable for very high rotation rates for geometric reasons and possible modifications of the sensor configuration must be considered. An easy-to-implement improvement comes with increasing the field of view. With a star camera with a 40 angular degree square field of view, only about 1000 stars are needed in the catalog for the star groups. This results in a rotation rate limit of 14 angular degrees per second with an image exposure of 100 milliseconds. Even larger fields of view result in a further increase in the theoretical rotation rate limit. The approximate uniform distribution of the stars assumed here is not practical in front. Therefore, when designing the star catalog and the star group catalog for high rotation rates, special attention must be paid to the interstellar distance. Stars with close neighbors may not be used for the catalogue. Since all stars are detected in the image of the star camera, including the unsuitable ones, there is also a moderately increased processing power in processing block 7 of the star group generation. Overall, with the help of an enlarged field of view, an adapted catalog design and increased computing power, the desired rotation rate limit of 20 angular degrees per second for the emergency regime can be achieved with a simple, typical star sensor with window detection.

Newer star sensors no longer work with star windows but with star clusters. Clusters are groups of contiguous bright image elements that are extracted from the image as objects.

When continuing the measurement of angular position and angular rate with the method of the invention in a continuous sequence, a measurement is possible in each star sensor cycle, no matter how the angular rate changes below the limit. During the first measurement under "Lost in Space" conditions, it can happen that a measurement result is only available in the second cycle at high rotation rates. The reason for the possible failure of the very first cycle is the missing constellation from the previous cycle.

Fig. 7 shows a flowchart of processing in "Lost in Space" conditions in the very first star sensor cycle. In the very first cycle, the current rate of rotation is set to zero for roll correction. This can be done by assigning the zero value directly, or by instead of the asterisks of the previous cycle, the data of the current cycle are used. With a zero yaw rate, the roll shutter correction does not change the data. A measurement result in the first cycle will therefore only be available if the yaw rate leads to tolerable distortions. With a yaw rate of 1 Angular degrees per second and an extent of the star group over half the field of view, the typical star sensor produces a distortion of about 2 pixels. 2 pixels of geometric error usually lead to failure of the identification. Nevertheless, in practice rotation rates of 1 to 4 angular degrees per Second for star identification on a typical star sensor be tolerated. That's because the rolling distortion can act in part as rotation and magnification. These components of the distortions are compensated for by the identification algorithms.

In the flow chart of Figure 7 there is a success check of the assignment of the equalized star groups to the catalog. It is needed in particular for the very first processing cycle. A success check of the assignment of the star groups from the current and the last image is not required. The assignment of the distorted star groups of two consecutive images works for all relevant scenarios of use of the star sensor. When rotating around the image axes, the time differences in star detection do not change. In this best case, there is no change in rolling distortion from frame to frame. The greatest change in roll distortion occurs when rotating about the star camera's optical axis. At a rotation rate of 20 angular degrees per second, it is less than a third of the picture element and is therefore non-critical.

8 shows an application of a robust determination of rotation rate and orientation with a star sensor using selected star group catalog coding. FIG. 9 shows a flow chart for robust determination of rotation rate and orientation with a star sensor using selected star group catalog coding.

Derived from the method according to FIG. 8 and FIG. 9, the known properties of a calibrated star sensor such as optical focal length, detector pixel size and number are included in the star group coding. By arranging the asterisk codes in preferably binary search trees, very fast asterisk identifications are possible. The method was optimized and verified when applied to real measurement data. With this proposal, real-time star identifications are possible in each individual and mutually independent measurement cycle (<100ms). This means that as soon as a group of stars (3...4 stars) is detected, an identification and position calculation can take place. If this takes place in directly consecutive cycles or with known time intervals, the yaw rate, direction and acceleration or a rate estimation can be carried out even for very high yaw rates. Conventional position tracking can be supported with this method, especially at high rates and accelerations. which leads to an increase in the robustness of the position measurement. These properties are a valuable addition to the functionality of a star tracker.

In particular, optional features of the invention are referred to as “may”. Accordingly, there are also developments and/or exemplary embodiments of the invention which additionally or alternatively have the respective feature or features.

If necessary, isolated features can also be selected from the combinations of features disclosed here and used in combination with other features to delimit the subject matter of the claim, eliminating any structural and/or functional relationship that may exist between the features.

Reference sign

1 orientation system

1 a orientation system

1 b orientation system

1 c attitude determination system

2 star camera

3 processor device

4 constellation

5 processing block

6 Asterisk Dates

7 processing block

8 rotation rate

9 attitude control system

10 processing block

1 1 Asterisk Data

12 processing block

13 location information

14 asterism catalog data

15 star catalog data

16 star

17 asterism

18 star

19 asterism

20 star

21 asterism

22 field of view

23 star

24 main axis

25 position vector

26 main axis

ZI main axis

28 Start of measurement, cycle number=O Calculation of the star vectors from the image

Calculation of the star groups from the star vectors

cycle number=O? Yes No

Current turn rate=O

Assignment of the current and the previous star groups

Calculation of the current rotation rate

Roll-top correction of the star groups

Assignment of corrected and catalog asterisms

Successful mapping? no Yes

Reporting a measurement error to the satellites

Final calculation of angular position and angular rate and report to the satellite

Cycle number=cycle number+1

Execution of the method according to the invention, duration <100ms (real time)

detect (x,y)

Identify

location calculation

Detector, binning level 1, 2, 4

# objects?

Identify

rates exist?

Yes

Rolling shutter correction

no

Form Asterisk Code(s)

Catalog Asterisks Storage

Asterism ID?

No next cycle Yes Location storage cycle N Guide Star catalog locations N-1 exists? No Yes Calculate rate and direction

Claims

Claims Method for determining the position of a spacecraft in space, characterized in that the steps are cyclically repeated recording of distorted constellations (4), processing of the distorted constellations (4) to form distorted star group data (6), storage of the distorted star group data (6),

determining a current rotation rate (8) by comparing the distorted star group data (6) of two consecutive cycles,

Transmission of the current rotation rate (8) to an attitude control system (9) and/or the steps

Processing of the distorted constellations (4) of a current cycle to equalized star group data (11),

Determination of position information (13) by assigning the equalized asterism data (1 1) to entrained asterism catalog data (14), transmission of the position information (13) to the position control system (9). A method for determining the attitude of a spacecraft in space, characterized in that taking into account known system parameters of an optical system, the steps

encoding asterism catalog data with = 3 ...4 stars [x _n ,y _n ,z _n ] visible in an image field into representative focal plane coordinates,

Forming a scaling, translation and rotation invariant

Asterisk codes based on [x _PiX ,y _PiX ] _n or the step

Encoding of asterism catalog data with n = 3...4 stars [x _n ,y _n ,z _n ] visible in a tile into representative tangent and/or angular coordinates [tan(a) , tan(/? )] _n to be executed.

27 Method according to at least one of the preceding claims, characterized in that the star group catalog data (14) contains data on group stars, on star groups (17, 19, 21) and on a vector index tree, the data on the star groups (17, 19 , 21 ) contain identification vectors and reference data and the vector index tree relates to the identification vectors of the asterisk groups (17, 19, 21 ). Method according to at least one of the preceding claims, characterized in that star catalog data (15) are carried along with additional star data and used. Method according to at least one of the preceding claims, characterized in that the position information (13) is statistically filtered. Method according to Claim 5, characterized in that the position information (13) is filtered over a number of cycles and/or over a number of star cameras. Method according to at least one of the preceding claims, characterized in that at least one star camera (2), in particular a roll-up shutter camera, is used to record the distorted constellations (4). Method according to Claim 7, characterized in that the distorted star images (4) are processed using the at least one star camera (2) to form the distorted star group data (6). Method according to at least one of Claims 7 to 8, characterized in that the at least one star camera (2) has picture elements and a number of picture elements which are adjacent to one another are combined to form a picture element module in order to increase a rotation rate limit. Method according to at least one of claims 7 to 9, characterized in that for detecting different image fields (22) a Star camera (2) or several star cameras (2) can be used together. 1 . Method according to at least one of the preceding claims, characterized in that the method is carried out using at least one separate processor device (3) and/or using a processor device (3) of the spacecraft. 2. Device for carrying out a method according to at least one of Claims 1 to 11, characterized in that the device has at least one star camera (2), in particular a roll-up shutter camera, and at least one processor device (3), the at least one processor device (3 ) a first processing block (5) for cyclically repeated recording of distorted constellations (4), for processing the distorted constellations (4) into distorted constellation data (6) and for storing the distorted constellation data (6) and a second processing block (7) for determination a current yaw rate (8) by comparing the distorted star group data (6) of two consecutive cycles, and/or a third processing block (10) for processing the distorted star images (4) of a current cycle into corrected star group data (11) and a fourth processing block (12 ) for determining a position information (13) by assigning the ent dragged asterism data (1 1 ) to carried asterism catalog data (14); includes. 3. Device according to claim 12, characterized in that the device has several star cameras (2) and for each star camera (2) a separate processor device (3) or for a group of star cameras (2) a common processor device (3). Device according to at least one of Claims 12 to 13, characterized in that the at least one processor device (3) is a separate processor device (3) for carrying out the method according to at least one of Claims 1 to 11 or a processor device (3) of the spacecraft. Computer program product, comprising instructions which, when the program is executed, cause at least one processor device (3) to carry out the method according to at least one of claims 1 to 11.