Speaker
Matthew Millard
Topic
Predicting Human Motion Using Optimal Control
Abstract
Computed predictions of human motion and muscle force patterns are valuable tools for studying biomechanics, designing wearable robotic systems, and improving subject-specific orthoses fitting. Optimal control is a promising means of predicting human motion given a sufficiently realistic musculoskeletal model, cost-function, and task description. This talk will focus on two exciting applications of optimal control: predicting the best ankle-foot-orthoses spring stiffness for a child with crouch gait, and predicting human lifting motions as a design aid for the design of a low-back exoskeleton.
Speaker's Biography
Matthew Millard is a Postdoctoral fellow in the Interdisciplinary Center for Scientific Computing in the Optimization in Robotics and Biomechanics research group at Heidelberg University. He is interested in the mechanics and control of human movement. Presently he is focused on applying optimal control to predict the motions and muscle forces of musculoskeletal models during walking and lifting motions.
Sponsored by the IEEE Computational Intelligence Society and the IEEE Control Systems Society of the KW Section. Refreshments will be served. This talk is part of the Centre for Pattern Analysis and Machine Intelligence Seminar Series.