Seminar: Robot learning in unstructured environments

Title: Robot learning in unstructured environments Abstract: Contemporary robots have become exceptionally skilled at achieving specific tasks in structured environments, but often fail when faced with the limitless permutations of real-world unstructured environments. This motivates robotics methods which learn from experience, rather than follow a pre-defined set of rules. Machine learning can be applied to a diverse range of components throughout the autonomy stack, allowing the robot to learn both about and from its surroundings. In this talk, we present a range of learning-based methods aimed at enabling robots, operating in dynamic and unstructured environments, to better understand their surroundings, anticipate the actions of others, and take informed actions accordingly. Bio: William Zhi is a PhD candidate at the University of Sydney supervised by Prof. Fabio Ramos and Dr. Lionel Ott (now at ETH Zurich). His research lies at the intersection of machine learning and robotics. He has worked on a broad range of problems, which include motion planning and control, motion prediction, and learning from demonstration. He has published 14 peer-reviewed publications at leading robotics and machine learning venues, including ICML, CoRL, L4DC, IEEE RA-L, ICRA, IROS. He was recognised as a Robotics: Science and Systems (RSS) Pioneer in 2020, and was awarded the Annual Learning for Dynamics and Control Conference (L4DC) 2022 best paper award. During his PhD, he interned at NVIDIA's robotics lab, led by Prof. Dieter Fox. He will be joining the Robotics Institute at Carnegie Mellon University as a postdoctoral research fellow in early 2023.