Title: Trait-Based Modeling for Multi-Robot Coordination

Date: Septemeber 30, 2022
 

Time: 1:30PM - 3:30PM EST
Location (Virtual): https://gatech.zoom.us/u/a3m8YouaS

Video Conferencing, Web Conferencing, Webinars, Screen Sharing

Zoom is the leader in modern enterprise video communications, with an easy, reliable cloud platform for video and audio conferencing, chat, and webinars across mobile, desktop, and room systems. Zoom Rooms is the original software-based conference room solution used around the world in board, conference, huddle, and training rooms, as well as executive offices and classrooms. Founded in 2011, Zoom helps businesses and organizations bring their teams together in a frictionless environment to get more done. Zoom is a publicly traded company headquartered in San Jose, CA.

gatech.zoom.us

Location (In-Person): C1008 Bolton in CODA

Glen Neville
Robotics PhD Student
School of Electrical/Computer Engineering
Georgia Institute of Technology

Committee:
Dr. Sonia Chernova(Advisor) - School of Interactive Computing, Georgia Institute of Technology
Dr. Harish Ravichandar- School of Interactive Computing, Georgia Institute of Technology
Dr. Seth Hutchinson - School of Interactive Computing, Georgia Institute of Technology
Dr. Magnus Egerstedt- School of Engineering, University of California Irvine
Dr. Nicholas Roy - Department of Aeronautics & Astronautics, Massachusetts Institute of Technology

Abstract:
 Heterogeneous multi-agent systems offer the potential to solve complex problems in various domains, that would otherwise be infeasible for a single agent. To effectively deploy multi-robot teams, researchers need to reason about several interdependent problems at varying levels of abstraction. In particular, there are four important questions that heterogeneous multi-robot systems must address: task planning (what), motion planning (how), task allocation (who), and scheduling (when). These problems are complex, interconnected, and flexible to various team compositions and agent types. To solve such complex problems, researchers require accurate models of agent capabilities that help teams effectively leverage the individual agents' relative strengths. One methodology for modeling agents in a multi-agent team is through the use of traits (capabilities).
        This thesis examines the use of trait-based models for representing individual agents in the context of multi-agent teaming applications and how trait-based modeling can be leveraged to enable more robust and efficient solutions to multi-agent coalition formation.  Specifically, we examine how these techniques can be used in coalition formation algorithms to answer the four problems of task allocation, scheduling, motion planning, and task planning.