Human motion intent description based on switching mechanism for rehabilitation robot

Prof. Steven W. Su

[Abstract]

For a complicated rehabilitation task, the major difficulty of the EMG-based decoder modeling is estimating the needed force from the EMG signals with high accuracy. Our recent study presented a switching mechanism for carving up the complex task into simple subtasks for easy and accurate estimation. However, it was observed that the "bump" behavior (the discontinuity) during the switching between the two models is unacceptable during real rehabilitation experiments. The bump of the outputs in real-time may cause unexpected high forces in the affected limb and may be a cause of injury for patients. In order to improve this bad transient behavior, we attempted to maintain the continuity of the outputs during switches between multiple EMG-decoder models. A new modelling and switching method is proposed and applied in the control of rehabilitation robots. Numerical simulation and real-time experiments indicate the bumpless transfer is significantly better than the bump transfer in both testing and verifying phases, and thus can further avoid the risk to the limb caused by handling unpredictable loads for participants.

[Biography]

Steven Su received the Ph.D. degree from the Research School of Information Sciences and Engineering (RSISE), the Australian National University (ANU), Canberra, Australia, in 2002. He was also served as a Research Fellow at the University of New South Wales, Sydney, Australia, from 2002 to 2006. He is currently an Associate Professor with the School of Biomedical Engineering, the Faculty of Engineering, University of Technology Sydney (UTS). Dr. Su has been engaged in the research areas of biomedical system modeling and control, and has published over two hundred refereed research papers. Dr. Su is a senior member of IEEE. He served as an Associate Editor of "International Journal of Advanced Robotic Systems" and an Editorial Member of the Journal of “Applied soft computing”, and the associate editor and committee members of Australian Control Conference and Australian Biomedical Engineering Conference.