学术报告第201326期

Takayuki  Matsuno  graduated from Nagoya University, Japan in 1998 and received  the Master of Engineering degree from Nagoya University in 2000. He received his Ph.D  from Nagoya  University in 2005. He was  Research Associate in the Department of Micro-Nano Systems Engineering and  Department of Mechano-Informatics and Systems, Nagoya University by Mar. 2006. He was Research  Associate or Lecturer in Intelligent Systems Design Engineering, Toyama Prefectural University by Sep. 2011. He is currently a  Lecturer in Graduate School of Natural Science and Technology, Okayama  University. He is mainly engaging in the research fields of flexible object  manipulation, automation of assembly task in factories and dynamical system  control.

报告题目：

Fault Detection Algorithm for External Thread  Fastening by Robotic Manipulator

-----------------------------

摘要：Fault detection functions of a robotic manipulator are very useful  for factory automation. All production has the possibility to fail due to  unexpected accidents, particularly producing them in new lines. Currently, human  workers should observe assembly robots in order to recovery them from small  accident. To reduce the fatigue of human workers, small errors automatically  should be corrected by a robot system.  In this paper, an external thread fastening  task by a robotic manipulator is investigated. To discriminate the four states  of a task, linear support vector machine methods (Linear SVM) with two feature  parameters are introduced. One is the estimated insertion length, and the other  is the maximum reaction force. The effectiveness of the proposed algorithm is  confirmed through an experiment and recognition  examination.

地点：南一楼中311室

时间：2013年12月11日上午9点30

邀请方：“多谱信息处理技术”国家级重点实验室