No abstract

No abstract

This paper investigates adaptive zero reaction motion control for free-floating space manipulators with uncertain kinematics and dynamics. The challenge in deriving the adaptive reaction null-space (RNS) based control scheme is that it is difficult to obtain a linear expression, which is the basis of the adaptive control. The main contribution of this paper is that we skillfully obtain such a linear expression, based on which, an adaptive version of the RNS-based controller (referred to as the adaptive zero reaction motion controller in the sequel) is developed at the velocity level, taking into account both the kinematic and dynamic uncertainties. It is shown that the proposed controller achieves both the spacecraft attitude regulation and end-effector trajectory tracking. The performance of the proposed adaptive controller is shown by numerical simulations with a planar 3-DOF (degree-of-freedom) space manipulator. Index TermsReaction null-space, adaptive control, uncertain kinematics and dynamics, free-floating space manipulator. is a qualified approach to handle parametric uncertainties [33].Among the control modes of space manipulators, free-floating space manipulators (FFSM) have their potential advantages, e.g., non-renewable fuel on the spacecraft can be saved and the safety of close-range manipulation can DRAFT 2 be ensured [8]. It is known that in a free-floating space manipulator, the motion of the spacecraft will evolve under the dynamic reaction due to that of the manipulator, and the evolution of the whole system is governed by the principle of momentum conservation. For the end-effector tracking problem without consideration of the spacecraft attitude, many adaptive control algorithms have been proposed (e.g., [6], [7]). Specifically, the tracking objective is realized by the prediction error based approach in [7] with the uncertainties of both the kinematics and dynamics being taken into consideration. However, in practice the spacecraft attitude maintenance is a major concern since the communication with the Earth can be carried out only when the spacecraft antenna points to the Earth (guaranteed by the attitude maintenance control) [9]. Hence, joint motion algorithms for space manipulators without reaction to the spacecraft are highly preferred. On the other hand, the manipulator end-effector is usually required to track some trajectory in Cartesian space when executing On-orbit servicing (OOS). Thus, it is meaningful to realize coordinated spacecraft/manipulator motion control.Many researchers have studied coordination control of a manipulator and its free-floating base. Vafa and Dubowsky proposed joint cyclic motion algorithm so that the spacecraft orientation is maintained [10]. Nakamura and Mukherjee presented an algorithm to achieve the regulation of both the spacecraft attitude and the manipulator joint angles simultaneously, where the stability of the system is analyzed by the Lyapunov method [11]. The motion planning for a system of coupled rigid bodies is investigated in [12], which is c...

No abstract