Data processing: vehicles – navigation – and relative location – Navigation – Employing position determining equipment
Reexamination Certificate
2000-08-01
2001-10-16
Cuchlinski, Jr., William A. (Department: 3661)
Data processing: vehicles, navigation, and relative location
Navigation
Employing position determining equipment
C244S158700
Reexamination Certificate
active
06304822
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates generally to spacecraft attitude control systems and, more particularly, to spacecraft attitude determination systems.
2. Description of the Related Art
Spacecraft attitude control is essential because a spacecraft must be properly oriented to perform the functions for which it was designed. Attitude control is generally realized with the aid of torque-generation elements (e.g., thrusters, momentum wheels and/or magnetic torquers) which apply torque inputs U(t
n
) to a spacecraft body so as to change its state X(t
n
) (e.g., its attitude and attitude rate).
To determine the current state, an attitude determination system typically receives attitude measurements Y(t
n
) from various instruments (e.g., star trackers) and, in response, generates attitude estimates X*(t
n
) The difference between these attitude estimates X*(t
n
) and a commanded attitude defines torque command signals and, in response to these torque command signals, the torque-generation elements apply corrective torque inputs U(t
n
) that urge the spacecraft body towards the commanded attitude.
As indicated by the symbol t
n
, the measurements Y(t
n
) and estimates X*(t
n
) are generally performed at successive times. Between each of these successive times, the attitude determination system typically receives attitude rate measurements Y(t
n
) from gyroscopes which enable it to extrapolate each attitude estimate to the time of the next attitude measurement. The difference between the extrapolated estimate and its corresponding measurement forms a residue which is processed with a variable gain factor to form a correction of the previous estimate and update it to a current estimate. These actions realize attitude estimates whose variances are significantly reduced from those of the measurements and they are generally performed with a recursive estimator process (e,g., a Kalman filter process) that is programmed into data processors of the attitude determination system.
FIG. 1
shows an exemplary arrangement in which a stellar inertial attitude determination (SIAD) system
20
receives star tracker signals
22
from one or more star trackers
24
. Star trackers are complex semiconductor systems which generally include a) an array of light sensitive elements that collect charge in response to incident light, b) an arrangement of charge-transfer elements that transfer (i.e., readout) the collected charges and c) an output structure that converts the transferred charges to corresponding voltage or current signals. The charge-transfer elements are generally realized with charge-coupled devices (CCDs) that are formed with metal-oxide semiconductor capacitors. Star trackers are thus sometimes referred to as CCD arrays and each array element is often referred to as an array pixel.
The collected charges of the array are processed into star centroids and each transfer of the processed charges to the star tracker output structure is typically referred to as a data frame. At a frame rate, therefore, the star tracker output signal provides frame data in the form of vertical coordinates C
v
, horizontal coordinates C
h
and star magnitudes M
s
for respective stars in the star tracker's field of view. The SIAD system
20
uses the star magnitudes M
s
and the vertical and horizontal coordinates C
v
and C
h
to identify respective stars and the vertical and horizontal coordinates C
v
and C
h
to determine the spacecraft attitude with respect to known stellar positions.
Star tracker signals generally include noise which comprise temporal noise (e.g., circuit-generated noise and background thermal noise) and spatial noise (e.g., errors induced by a less than perfect charge transfer efficiency). In order to reduce this noise, star trackers typically average the frame data of multiple data frames to form their output signals.
As indicated within the broken-line ellipse
26
of
FIG. 1
, for example, the star tracker
22
internally generates successive data frames
28
(each indicated by a frame enclosing a letter D) at a frame rate. The star tracker
22
then averages (as indicated by bracket-and-arrow
30
) the frame data of a plurality (e.g., four) of the most current data frames
28
to generate averaged data frames
32
(each indicated by a frame enclosing a letter D with an overhead line that symbolizes the averaging process) at the frame rate. In addition, the valid flags of the averaged data frames are anded to provide a corresponding valid flag
33
for each averaged data frame
32
. The averaged data frames
32
and corresponding valid flags
33
thus form the star tracker signals
22
as indicated by the broken-line ellipse
34
.
Because star trackers do not operate in a benign world, the charge collection at array pixels can be corrupted by spurious inputs such as the impinging protons
36
of FIG.
1
. The major source of impinging protons are solar flares of the sun that ebb and flow in various rhythms (e.g., with an eleven year cycle). The operation of the SIAD system
20
is degraded if it processes signals from frame pixels (often referred to as “hot pixels”) whose collected charges have been corrupted. Therefore, the star trackers
24
provide, for each averaged data frame
32
, a flag that is set invalid if the collected charges differ excessively from their previous values and is set valid otherwise. In response, the SIAD system
20
only processes data frames that are accompanied by a valid flag.
The probability of obtaining valid data frames with the static processing of
FIG. 1
decreases as the influx of impinging protons
26
increases. When this influx is high, the lack of valid data frames significantly degrades operation of the SIAD system
20
so that the accuracy of the spacecraft attitude control decreases. Under extreme influx conditions, spacecraft missions may be placed in jeopardy (especially long-term missions whose operational time frames include that of a solar flare).
SUMMARY OF THE INVENTION
The present invention is directed to methods for dynamically processing successively-generated star tracker data frames and associated valid flags to generate processed star tracker signals that have reduced noise and a probability greater than a selected probability P
slctd
of being valid. These methods maintain accurate spacecraft attitude control in the presence of spurious inputs (e.g., impinging protons) that corrupt collected charges in spacecraft star trackers.
In one method embodiment, valid data frames are identified, from the valid flags, in each of successive sets S of M data frames. The frame data in the valid data frames is then averaged to generate a processed star tracker signal at a rate 1/M of the rate of the data frames and with reduced noise. In this method embodiment, data frames are not statically selected for averaging but, rather, data frames of each set S are dynamically selected on the basis of their validity.
Another method embodiment comprises the process steps of
a) selecting a plurality of different frame combinations C
frm
in each of successive sets S of M data frames wherein each frame combination C
frm
is formed with a respective number N of the M data frames;
b) calculating, in response to the valid flags and for each frame combination C
frm
, a combination probability P
cmb
of being valid; and
c) from all frame combinations C
frm
whose combination probabilities P
cmb
exceed a selected probability P
slctd
, averaging the frame data of a valid one of those frame combinations that have the largest number N of the data frames to thereby obtain the processed star tracker signal at a rate 1/M of the rate of the data frames and with reduced noise.
The combination probabilities P
cmb
are calculated from a frame probability P
frm
which can be continuously tracked. The methods of the invention enhance the probability of generating valid star tracker signals because they respond to the current frame probability P
frm
by dynamically selecting the largest valid frame combination whose
Li Rongsheng
Liu Yong
Wu Yeong-Wei Andy
Cuchlinski Jr. William A.
Gudmestad T.
Hernandez Olga
Hughes Electronics Corporation
LandOfFree
Spacecraft attitude control systems with dynamic methods and... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Spacecraft attitude control systems with dynamic methods and..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Spacecraft attitude control systems with dynamic methods and... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2588527