Date: Friday, December 6th
Time: 11:30 am - 1 pm EST
Location: (In person) Klaus 2108,
(Virtual) https://gatech.zoom.us/my/jinwoop?pwd=OFJhVTcwNGI3ZW1YdERzWHgrUzZIZz09
Abstract:
Effective coordination of multi-robot teams is essential for achieving goals through improved efficiency
and robustness, demanding careful reasoning in task allocation, scheduling, and motion planning. However,
addressing these interdependent problems simultaneously increases complexity, making it challenging to find
an optimal solution. The proposed thesis examines a time-extended heterogeneous multi-robot task allocation
framework designed to achieve multiple objectives under numerous constraints in probabilistic and dynamic
settings. To that end, we introduce a framework that employs statistical tests to account for trait uncertainties
in both robots and tasks, offering a theoretical guarantee for the resulting solution while also accommodating
tighter temporal constraints. Furthermore, we propose an approach to handle dynamic traits which evolve
over time. By incorporating the notion of depletion and trait provision rates, the time required to complete
tasks varies based on coalition composition, thus enabling a more accurate estimation of makespan. The
proposed research seeks to integrate task prioritization and leverage control barrier functions to ensure the
collision-free execution of tasks by robot coalitions. We also aim to explore the impact of human preference on
human-robot collaboration, which has applications in smart factories and warehouse environments. Lastly, we
propose a reactive framework for a dynamic environment, balancing feasibility and time efficiency.
Committee:
Dr. Seth Hutchinson (Advisor)
School of Interactive Computing
Georgia Institute of Technology
Dr. Sonia Chernova
School of Interactive Computing
Georgia Institute of Technology
Dr. Harish Ravichandar
School of Interactive Computing
Georgia Institute of Technology
Dr. Matthew Hale
School of Electrical and Computer Engineering
Georgia Institute of Technology
Dr. Nicholas Roy
Department of Aeronautics and Astronautics
Massachusetts Institute of Technology
]]>