Comparative study of post-impact motion control of a flexible arm space robot

Abstract

Designing a controller to manipulate a flexible space robot is hampered by the inherent complex nonlinear dynamics. Furthermore, the collision of a space robot with an external object in the space environment often leads to degradation of the space robot system. Therefore, an effective control algorithm is necessary for a successful space robot mission to regulate the motion of the entire system after an impact. This paper presents a comparative study on the performance of two different composite controllers to control post-impact motion of the space robot with two-link flexible arms. A performance comparison study helps to choose a better controller to carry out specific tasks. The dynamics of a space robot with flexible manipulators are highly nonlinear and coupled, hence split into rigid and flexible motion to make the control design easier. In this paper, two different composite controllers, namely Nonlinear Model Predictive Control with a Linear Quadratic Regulator (NMPC-LQR) and Sliding Mode Control with Linear Quadratic Regulator (SMC-LQR) are developed. NMPC and SMC are used to control rigid motion of the system, while LQR is used to suppress the flexible motion. The objective of the proposed controllers is to regulate the motion of the entire space robot system, including the flexible manipulator, after impact with an external object. The effectiveness of the proposed composite controllers is demonstrated using numerical simulations. Simulation results confirm that, in the absence of uncertainties, the performance of the control system using the proposed NMPC-LQR is more accurate compared to using SMC-LQR. Furthermore, given uncertainties of mass and inertia of the system, we find NMPC-LQR provides a higher level of accuracy when compared to SMC-LQR.