ID：843 563 856
报告人：Australia Zhihong Man 教授
Zhihong Man is currently the Professor of Engineering in the School of Science, Computing Engineering Technologies at Swinburne University of Technology, Melbourne, Australia. His research interests are in nonlinear control, signal processing, robotics, neural networks, engineering optimization and vehicle dynamics & control. He has published more than 250 research papers in refereed international journals and refereed international conferences proceedings, and his research results have widely cited about 13400 times by the researchers from more than 30 different countries.
In this talk, a new extended sliding mode observer for a class of second-order linear systems is discussed. It will be shown that, when the sampling period is small enough, the unmeasurable state estimation errors can be numerically approximated by using the measurable system output information and then used for constructing the switching functions of the sliding mode observer, to drive the observer error dynamics to reach and then slide across the sliding surfaces. The advantages of the proposed new extended sliding observer are in then following three points: (i) It has a solid theoretical background of stability and convergence ; (ii) The designed new observer can sufficiently approximate the observer-like system when the sampling period is small enough; (iii) The achieved fast error convergence is very helpful for practical implementations.