Abstract: The maneuvering control of an underactuated axisymmetric spacecraft with the symmetry axis as unactuated axis was investigated under the assumption that the angular velocity component along the unactuated axis was zero. The generalized dynamic inverse (GDI) method was utilized in conjunction with backstepping method to design a switched control logic yielding global convergence of the attitude error. Singularities inherent in the control law, which may arise at the beginning or during the steady phase of the control process, were wiped out by introducing a saturated function and properly choosing control gains. Simultaneously, the large transient amplitude, which may occur during the initial maneuver phase, was avoided. Strict proof shows that by this control law the attitude error globally converges to zero and there is at most one switching during the whole control process, thus eliminating the possibility of chattering. Numerical simulation results demonstrate the effectiveness of the proposed control law.

Key words: Attitude control, Dynamic inverse method, Global convergence, Underactuated spacecraft