In this dissertation the solutions of the dynamics and real-time optimal control of
magnetic attitude control and formation flying systems are presented. In magnetic
attitude control, magnetic actuators for the time-optimal rest-to-rest maneuver with a
pseudospectral algorithm are examined. The time-optimal magnetic control is bang-bang
and the optimal slew time is about 232.7 seconds. The start time occurs when the
maneuver is symmetric about the maximum field strength. For real-time computations,
all the tested samples converge to optimal solutions or feasible solutions. We find the
average computation time is about 0.45 seconds with the warm start and 19 seconds with
the cold start, which is a great potential for real-time computations. Three-axis magnetic
attitude stabilization is achieved by using a pseudospectral control law via the receding
horizon control for satellites in eccentric low Earth orbits. The solutions from the
pseudospectral control law are in excellent agreement with those obtained from the
Riccati equation, but the computation speed improves by one order of magnitude. Numerical solutions show state responses quickly tend to the region where the attitude
motion is in the steady state.
Approximate models are often used for the study of relative motion of formation
flying satellites. A modeling error index is introduced for evaluating and comparing the
accuracy of various theories of the relative motion of satellites in order to determine the
effect of modeling errors on the various theories. The numerical results show the
sequence of the index from high to low should be Hill's equation, non- J2, small
eccentricity, Gim-Alfriend state transition matrix index, with the unit sphere approach
and the Yan-Alfriend nonlinear method having the lowest index and equivalent
performance. A higher order state transition matrix is developed using unit sphere
approach in the mean elements space. Based on the state transition matrix analytical
control laws for formation flying maintenance and reconfiguration are proposed using
low-thrust and impulsive scheme. The control laws are easily derived with high
accuracy. Numerical solutions show the control law works well in real-time
computations.
| Identifer | oai:union.ndltd.org:tamu.edu/oai:repository.tamu.edu:1969.1/4283 |
| Date | 30 October 2006 |
| Creators | Yan, Hui |
| Contributors | Alfriend, Kyle T. |
| Publisher | Texas A&M University |
| Source Sets | Texas A and M University |
| Language | en_US |
| Detected Language | English |
| Type | Book, Thesis, Electronic Dissertation, text |
| Format | 1056719 bytes, electronic, application/pdf, born digital |
Page generated in 0.0122 seconds