Design and Evaluation of State and Disturbance Observers for a Multivariable Magnetic Levitation System

Abstract

For inherently unstable magnetic levitation (maglev) systems, state observation is required for feedback control, and disturbance estimation is needed to compensate for unknown disturbances. Previous designs for simultaneously observing state and disturbance can lead to aggressive disturbance compensation and an unstable levitation. Aiming at these problems, this paper presents a state and disturbance observer that has a cascade structure for a maglev system. The proposed design allows the dynamics of disturbance observation to be separately tuned from those of state observation, avoiding aggressive disturbance compensation and enabling a stable levitation of a maglev system. In this paper, the proposed and previous state and disturbance observers are designed and realized for a maglev system that features wide ranges of translation and rotation motions in five directions. Comparative studies are reported in this paper, and experimental results show feasibility and efficiency of the proposed scheme.

Keywords:

• Luenberger observer
• Magnetic levitation
• Multi-degree-of-freedom system
• Sliding mode
• State and disturbance estimation
• State and disturbance observer

ACKNOWLEDGMENTS

The authors are grateful to the Ministry of Science and Technology, Taiwan, for supporting this research under Grant MOST 109-2221-E-003-022. The supports from the University of Hawaii College of Engineering are also gratefully acknowledged.

Additional information

Funding

This work was supported by the Ministry of Science and Technology, Taiwan [grant number MOST 109-2221-E-003-022].

Notes on contributors

Yu-Sheng Lu

Yu-Sheng Lu received the BS degree in mechanical and electrical engineering from National Sun Yat-Sen University in 1990, and the PhD degree in engineering from National Tsing Hua University in 1995. From 1997 to 1998, he was at the Opto-Electronics & Systems Laboratories, ITRI. From 1998 to 2000, he was with the Institute of Robotics and Mechatronics, DLR, Germany. From 2000 to 2008, he was with the Dept. Mechanical Engineering, National Yunlin University of Science and Technology. In 2008, he moved to the Dept. Mechatronic Engineering, National Taiwan Normal University, Taipei, Taiwan, where he is currently a professor.

Peter Berkelman

Peter Berkelman received the BS and MS degrees in mechanical engineering from the Massachusetts Institute of Technology, Cambridge, MA, in 1992, and the PhD degree in robotics from Carnegie Mellon University, Pittsburgh, PA, in 1999. He has completed engineering internships at Philips, Fujitsu, and Toshiba Laboratories. He was a postdoctoral researcher at The Johns Hopkins University during 1999–2001 and the TIMC-IMAG Laboratory at the University of Grenoble during 2001–2004. He is currently an associate professor in the Department of Mechanical Engineering, University of Hawaii-Manoa, Honolulu. Email: [email protected]