Neural optimal control of flexible spacecraft slew maneuver

This paper deals with the problem of optimal large-angle single-axis maneuvers of a flexible spacecraft with simultaneous vibration suppression of elastic modes. A spacecraft model with a cylindrical hub and one flexible appendage and tip mass is considered. Gravity gradient torque is considered as...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Acta astronautica Ročník 55; číslo 10; s. 817 - 827
Hlavní autori: Nayeri, M.Reza Dehghan, Alasty, Aria, Daneshjou, Kamran
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Elsevier Ltd 01.11.2004
ISSN:0094-5765, 1879-2030
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:This paper deals with the problem of optimal large-angle single-axis maneuvers of a flexible spacecraft with simultaneous vibration suppression of elastic modes. A spacecraft model with a cylindrical hub and one flexible appendage and tip mass is considered. Gravity gradient torque is considered as a disturbance torque. Multilayer perceptron neural networks are used to design a Neural Optimal Controller (NOC) for this multivariable non-linear maneuver. For NOC training, an off-line training procedure based on backpropagation through time algorithm is developed to minimize the general quadratic cost function in forward and backward pass stages. The proposed controller is also applicable to simultaneous multi-axis reorientation of a flexible spacecraft. Simulation results are presented to show that very fast reference pitch angle trajectory tracking and vibration suppression are accomplished.
Bibliografia:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 23
ISSN:0094-5765
1879-2030
DOI:10.1016/j.actaastro.2004.04.002