Unifying Semantic and Physical Intelligence for Generalist Humanoid Robots
Georgia Tech Research via YouTube
Overview
Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Join a one-hour seminar presentation by Carnegie Mellon University's Assistant Professor Guanya Shi from The Robotics Institute, exploring the integration of semantic and physical intelligence in generalist humanoid robots. Delve into three critical aspects of humanoid intelligence: semantic understanding and reasoning, physical motion capabilities, and mechanical actuation/sensing. Learn about groundbreaking projects including H2O and OmniH2O, which create versatile interfaces for human control and autonomy methods, WoCoCo's framework for efficient locomotion-manipulation skill learning, and ABS's safety protocols for vision-based locomotion control. Discover how the unique advantages of humanoid robots - their generalist capabilities and human-aligned embodiment - can be leveraged through the combination of learning-based and traditional model-based control approaches to advance embodied intelligence in complex environments.
Syllabus
IRIM Seminar: Unifying Semantic and Physical Intelligence for Generalist Humanoid Robots
Taught by
Georgia Tech Research