Structure and controller design of a piezo-driven orientation stage for space antenna pointing. (April 2020)

• Record Type: Journal Article
• Authors: Shao, Shubao; Shao, Yan; Song, Siyang; Xu, Minglong; Ma, Xiaofei
• Is Part Of: Mechanical systems and signal processing. Volume 138(2020)
• Publication Date: 2020-04
• Subjects: Piezo-driven -- Orientation stage -- Support vector regression -- Adaptive control; Structural dynamics -- Periodicals; Vibration -- Periodicals; Constructions -- Dynamique -- Périodiques; Vibration -- Périodiques; Structural dynamics; Vibration; Periodicals; 621
• DOI: 10.1016/j.ymssp.2019.106525
• Languages: English
• ISSNs: 0888-3270

Abstract

Highlights: A 2-DOF piezo-driven orientation stage(POS) for space antenna pointing is proposed. Support vector regression and finite element analysis are combined to implement the mechanical analysis of POS. Pointing accuracy is improved by the MIMO adaptive control that compensates the hysteresis and coupling motion between axes. Abstract: Space antenna pointing is key to establishing inter-satellite and earth-satellite communication links. Compared with the traditional electromagnetic driven mechanism for antenna pointing, the piezo-driven orientation stage (POS) is more compact in structure and has higher pointing accuracy. However, the main difficulties of designing a POS include the compromise between workspace and structural stiffness, as well as devising a controller that can compensate the hysteresis behavior of the piezo-driven unit and coupled motion between axes. This paper presents a 2-DOF POS in which the rhombic mechanism and 2D flexure hinge are used to enlarge the workspace and decouple the motion between axes. To improve the mechanical modeling efficiency, support vector regression (SVR) and finite element analysis (FEA) are employed in combination. Subsequently, the trade-off between the workspace and structural stiffness is achieved by implementing the structure optimization design. To compensate the hysteresis and remaining coupling motion between axes, multiple input multiple output (MIMO) adaptive control is used to improve the control accuracy, where the real-time nonlinear model of the controlled plant with hysteresis property is approached by the controlled auto regression and moving average (CARMA) model.