Environmental Engineering Reference
In-Depth Information
Onboard control of attitude, by contrast, has been a necessary part of all
space missions other than a handful of generally low-cost missions utilizing
passive control methods such as gravity gradient stabilization. Using attitude
sensor measurements to determine the current attitude, the FSW compares
that attitude to the commanded (i.e., desired) attitude and determines an at-
titude error. For spacecraft employing gyros (an attitude sensor that measures
the change in spacecraft attitude during a set time period, as opposed to mea-
suring the spacecraft's absolute orientation with respect to inertial space), a
Kalman filter usually is utilized both to calculate the current attitude and to
calibrate the gyro's drift bias (which ramps with time) relative to an absolute
attitude sensor, such as a star tracker. The attitude error is estimated and fed
into a control law that calculates on each control cycle what attitude actuator
commands (e.g., reaction-wheel control torques) must be generated in order
to null the error. On the next control cycle, feedback from the attitude sensors
provides the information needed to determine how good a job of reducing at-
titude error the previous cycle's actuator commands did, as well as how much
new attitude error has been introduced this cycle by external perturbative
torques. Although this description of onboard attitude control has implicitly
addressed maintenance of a constant commanded attitude in the presence of
perturbative torques, it can equally well be applied to the execution of large
desired attitude changes, called
slews
. Slews can be dealt with two ways. First,
the FSW can calculate the amount of attitude change to be performed during
a given control cycle, and modify the previous commanded attitude to reflect
that change, which would then be used directly as part of that control cycle's
attitude error. A second approach is simply to make the commanded attitude
the slew target attitude. Although the control law would not be able to null
that very large error (say, 90
◦
) in one control cycle, by limiting the size of the
commanded control torques in a given control cycle the FSW could gradually
work off the error over a series of control cycles, eventually reaching the slew's
target attitude.
