Control of Humanoid Systems


Luis Sentis, Stanford University



Abstract:

Research on humanoid robotics addresses challenging problems in mechatronics
and artificial intelligence and provides a unique platform to explore new
methodologies and techniques in related areas such as biomechanics,
rehabilitation and training, animation and entertainment. The focus of my
research is on the modeling and control of complex human-like robotic
structures that are capable of performing advanced interactive behaviors in
evolving dynamic environment and to cooperate and work with humans. In this
talk, I will discuss our ongoing effort in whole-body control of humanoid
systems involving multi-contact manipulation, gait locomotion, and other
related applications to the analysis of human physiological functions. Numerous
examples will be shown to illustrate the capabilities of our realtime control
framework.

Bio:

Dr. Luis Sentis is a postdoctoral fellow in the Department of Computer Science
at Stanford University. He obtained his Ph.D. and M.Sc Degrees in Electrical
Engineering from Stanford University in 2007 and 2000 respectively and his B.S.
in the School of Telecommunications from the Technical University of Catalonia
in 1996. From 1996 to 1998 he worked as a control engineer for Phase 2
Automation, a clean room automation company in Silicon Valley. During his
Ph.D., Dr. Sentis developed control methods for humanoid robots to allow them
to engage in advanced interactive behaviors needed to operate in human
environments. As part of a collaboration with Honda Motor Co., he is currently
implementing realtime control methods into the humanoid robot Asimo. In another
area of research, Dr. Sentis is implementing his framework to study
multi-vehicle group navigation in a project that he proposed to the Boeing
Company. More recently, he has started a collaboration with the Stanford School
of Medicine as well as with the Laboratoire d'Informatique, de Robotique et de
Microélectronique de Montpellier in France to develop artificial walking
methods for paraplegic individuals. In 2005 he was finalist for the best ICRA
(Internatinal Conference of Robotics and Automation) student paper award. In
1998, he was awarded La Caixa Foundation fellowship to pursue his Ph.D. studies
at Stanford. Dr. Sentis has 16 publications and has been an invited speaker to
more than 16 workshops and seminar talks. He is currently engaged in several
teaching activities for the courses on advanced robotic manipulation and
experimental robotics at Stanford.