BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
X-WR-CALNAME:ISE Graduate Seminar: Jingang Yi
X-WR-TIMEZONE:Eastern Time (US & Canada)
BEGIN:VEVENT
DTSTAMP:20261014T225151Z
UID:tag:localist.com\,2008:EventInstance_47746903039445
DTSTART:20221111T191500Z
DTEND:20221111T201500Z
DESCRIPTION:Presenter: Jingang Yi\, Professor at Rutgers University\n\nSemi
 nar Title: Machine Learning-enabled Inertial Sensor-based Activity Detecti
 on and Pose Estimation for Legged Walkers\n\n \n\nAbstract:\n\nReal-time g
 ait activity detection and pose estimation are critical for enabling many 
 human healthcare automation and other industrial automation applications. 
 In this talk\, I will present a machine learning-enabled\, wearable inerti
 al measurement unit (IMU)-based design to provide effective and efficient 
 activity detection and pose estimation for legged walkers. The gait event 
 and activity detection are first built on a recurrent neural network (RNN)
  with long short-term memory (LSTM) cells with IMUs on lower limbs. A lear
 ned low-dimensional motion manifold is parameterized by a phase variable w
 ith a Gaussian process dynamic model. I will present two case studies to i
 llustrate the design. The first application is related to real-time human 
 walking gait detection and pose estimation for construction workers\, and 
 the second application is for horse limb lameness detection and pose estim
 ation. Experimental results show that the RNN-LSTM-based approach achieves
  gait activity detection with 96% accuracy\, the estimated human pose erro
 rs are within 8.30 degs\, and the detection latency is within 20 ms using 
 only a single IMU attached to human shank. For horse lameness detection\, 
 the design achieves 95% accuracy and the pose estimation scheme can predic
 t the 24 limb joint angles with average errors less than 5 and 10 degs und
 er normal and induced lameness conditions\, respectively. Because the desi
 gn is built on wearable inertial sensors\, it can be potentially used for 
 potential real-time applications in open field.\n\n \n\nBio: \n\nProfessor
  Jingang Yi received the B.S. degree in electrical engineering from Zhejia
 ng University in 1993\, the M.Eng. degree in precision instruments from Ts
 inghua University in 1996\, and the M.A. degree in mathematics and the Ph.
 D. degree in mechanical engineering from the University of California\, Be
 rkeley\, in 2001 and 2002\, respectively. He is currently a Full Professor
  in mechanical engineering and a Graduate Faculty member in electrical and
  computer engineering at Rutgers University. His research interests includ
 e human-robot interactions and assistive robotics\, autonomous robotic and
  vehicle systems\, dynamic systems and control\, mechatronics\, automation
  science and engineering\, with applications to biomedical\, transportatio
 n and civil infrastructure systems. Prof. Yi is a Fellow of American Socie
 ty of Mechanical Engineers (ASME) and a Senior Member of IEEE. He has rece
 ived several awards\, including the 2017 Rutgers Chancellor’s Scholars\,
  2014 ASCE Charles Pankow Award for Innovation\, the 2013 Rutgers Board of
  Trustees Research Fellowship for Scholarly Excellence\, and the 2010 NSF 
 CAREER Award. He serves as a Senior Editor for IEEE Robotics and Automatio
 n Letters and IEEE Transactions on Automation Science and Engineering\, an
 d an Associate Editor for International Journal of Intelligent Robotics an
 d Applications. He also served in editorial board of IFAC journals Control
  Engineering Practice\, Mechatronics\, IEEE/ASME Transactions on Mechatron
 ics\, IEEE Transactions on Automation Science and Engineering\, IEEE Robot
 ics and Automation Letters\, and ASME Journal of Dynamic Systems\, Measure
 ment and Control.
LOCATION:
SUMMARY:ISE Graduate Seminar: Jingang Yi
URL;VALUE=URI:https://calendar.utk.edu/event/ise-graduate-seminar-jingang-y
 i
CATEGORIES:Lectures & Presentations
END:VEVENT
END:VCALENDAR