Speaker: Siavash Rezazadeh, Ph.D.
Research Scientist
Locomotor Control Systems Laboratory
University of Texas at Dallas
Title:
Using Reduced-Order Models for Design and Control of Bipedal Robotic Systems
Abstract:
Understanding the principles governing bipedal locomotion can lead to design and control of legged robots for applications such as disaster recovery, package delivery, and domestic assistance. From a different perspective, this understanding can also facilitate the development and control of wearable robots such as exoskeletons and multi-joint prosthesis. Despite the perplexing complexity of the dynamics of biological locomotion, it has been shown that their main characteristics can usually be described by simple underlying models. In particular, animals and humans extensively use the passive dynamics of their bodies to save energy. In this talk, I will present the use of reduced-order models in control of two different bipedal systems, namely, an underactuated legged robot and a multi-joint powered prosthetic leg. First, I present the blending of biological inspiration with nonlinear control theories in the formulation of a novel controller for stable and robust walking of ATRIAS for participation in DARPA Robotics Challenge. Next, I show how the ideas from legged robots control can inspire the design of a new multi-joint prosthesis controller for both walking and volitional tasks. Finally, using the results of these controllers, I propose a methodology to improve the leg design process for greater energy efficiency and better control, based on which the next generation of robotic legs can be developed.
Biography
Dr. Siavash Rezazadeh received his B.Sc. and M.Sc. from Sharif University of Technology, and his PhD from University of Alberta. Currently, he is a Research Scientist at the Locomotor Control Systems Laboratory, University of Texas at Dallas, working on design and control of prosthetic legs. Before joining Locolab, he worked at Dynamic Robotics Laboratory, Oregon State University on control of ATRIAS, a bipedal robot, for DARPA Robotics Challenge (DRC). His research interests include using fundamental concepts of mechanics for design and control of novel robots and mechanisms.
