Date: Tuesday, November 18th, 2025

Time: 11:00 am - 1 pm EST

Location: (In person) Klaus 2108, or (Virtual) Zoom

Jinwoo Park

Robotics Ph.D. Candidate
Institute for Robotics and Intelligent Machines

Daniel Guggenheim School of Aerospace Engineering
Georgia Institute of Technology

Committee:

Dr. Seth Hutchinson (Advisor) - Khoury College of Computer Sciences, Northeastern University

Dr. Harish Ravichandar - School of Interactive Computing, Georgia Institute of Technology

Dr. Sonia Chernova - School of Interactive Computing, Georgia Institute of Technology

Dr. Nicholas Roy - Department of Aeronautics and Astronautics, Massachusetts Institute of Technology  

Dr. Matthew Hale - School of Electrical and Computer Engineering, Georgia Institute of Technology

Abstract: 

Coordinating trait-based heterogeneous multi-robot systems requires addressing interdependent challenges such as task allocation, scheduling, and execution under spatiotemporal and resource constraints. 

These challenges become more complex when considering trait uncertainty, task priority, battery limitations, task and team compatibility, and guarantees of collision- and deadlock-free execution. 

Prior work has addressed these issues partially and often in isolation, but it typically does not consider multi-robot task allocation in time-extended settings, where planning occurs over longer horizons.

This dissertation examines multiple problem-specific frameworks for trait-based, time-extended multi-robot task allocation. 

Each framework addresses a distinct coordination challenge, including trait uncertainty, the provisioning of exhaustible traits, task prioritization, safe execution, and task and team compatibility in multi-human multi-robot settings.

Together, these frameworks improve the feasibility, efficiency, and adaptability of heterogeneous multi-robot teams in realistic and demanding scenarios.

The main contributions of this dissertation are as follows:
(i) a probabilistic task allocation framework that models uncertainty in robot and task traits using statistical hypothesis testing, thereby enabling tighter temporal constraints while providing theoretical guarantees on solution quality;
(ii) a dynamic trait modeling framework that captures time-varying and exhaustible traits under battery constraints; 

(iii) an integrated planning and execution framework that incorporates task priority into both high-level planning and low-level control, employing priority-aware control barrier functions to ensure timely and collision-free execution while permitting flexibility in path following; and 

(iv) a compatibility-aware teaming framework for multi-human, multi-robot coordination, which incorporates task and team compatibility into the problem formulation.