Dynamics of Nonlinear Orientation Control of Two Space Objects Bunch with the Accounting of Construction Flexibility

 

V.M. Glumov, S.D. Zemlyakov, V.Yu. Rutkovsky, V.M. Sukhanov

Institute of Control Sciences by V.A. Trapeznikov, Russian Academy of Sciences

Profsoyuznaya 65, 117806, Moscow, Russia

e-mail: zeml@ipu.rssi.ru

 

The synthesis and computer simulation of a control system for a free-flying space robotic module (SRM) is considered. It is assumed that the control system includes relay actuators. The mode of flexible payload transportation is analyzed. The control system is designed on the base of an adaptive co-ordinate-parametric control principle. The procedure of computer synthesis for a three-dimensional modal-physical dynamics model of the SRM plus a non-rigid payload (SRM-P) is described. The synthesis is made in MATLAB's and MAPLE's computer systems. On the basis of numerical analysis of the space module's dynamic portrait we have obtained the intensity dependence of the oscillations excited by relay control moments on the co-ordinates of the payload capture point. This dependence is represented by a multi-extremal function. Its global extremum (minimum) in the domain of admissible values for the argument is considered as the criterion in the task of choosing the optimal values of the payload capture point co-ordinates. The results of this task are used in the tuning of the capture point that makes it possible to avoid the resonance swing of the flexible payload and SRM-P as a whole.

Our results show that it is possible to solve the main problems connected with orientation of such complicated objects as free-flying space robotic modules with non-rigid payloads. The considered methods are not particularly simple. In the future it would be very useful for solving the problems of the type mentioned in the paper to obtain results in the following domains: 1) to improve identification algorithms for estimation of non-stationary parameters of flexible space object's with respect to increasing of the speed of acting without degrading of the accuracy achieved up to now; 2) to use the robustness property of adaptive loop to improve both the "rigid" motion () and the elastic one () with the help of the output of an adjustable model; 3) to develop more simple and reliable methods of non-exciting attitude control that would not demand overriding of the base algorithm.