Date of Award
Doctor of Philosophy in Computer Engineering - (Ph.D.)
Electrical and Computer Engineering
Qing Gary Liu
The physical intelligence, which emphasizes physical capabilities such as dexterous manipulation and dynamic mobility, is essential for robots to physically coexist with humans. Much research on robot physical intelligence has achieved success on hyper robot motor capabilities, but mostly through heavily case-specific engineering. Meanwhile, in terms of robot acquiring skills in a ubiquitous manner, robot learning from human demonstration (LfD) has achieved great progress, but still has limitations handling dynamic skills and compound actions. In this dissertation, a composite learning scheme which goes beyond LfD and integrates robot learning from human definition, demonstration, and evaluation is proposed. This method tackles advanced motor skills that require dynamic time-critical maneuver, complex contact control, and handling partly soft partly rigid objects. Besides, the power of crowdsourcing is brought to tackle case-specific engineering problem in the robot physical intelligence. Crowdsourcing has demonstrated great potential in recent development of artificial intelligence. Constant learning from a large group of human mentors breaks the limit of learning from one or a few mentors in individual cases, and has achieved success in image recognition, translation, and many other cyber applications. A robot learning scheme that allows a robot to synthesize new physical skills using knowledge acquired from crowdsourced human mentors is proposed. The work is expected to provide a long-term and big-scale measure to produce advanced robot physical intelligence.
Zhao, Leidi, "Data-driven learning for robot physical intelligence" (2021). Dissertations. 1549.