Back to search

Cornell University

High-Level Collaborative Task Planning for Heterogeneous Multi-Robot Systems

Abstract

dc:description.abstract

Temporal logics serve as a framework for expressing complex, temporally-extended specifications in a mathematical precise manner. Using formal synthesis techniques, these specifications can be automatically translated into high-level, correct-by-construction controllers for robots to execute. This enables the provision of robust guarantees regarding robot behavior and task feasibility. The diverse expressivity of different logics enables them to be used in a wide variety of robotic systems. This dissertation focuses on synthesizing high-level controllers for heterogeneous robots accomplishing a global task. First, we formulate the autonomous participation problem for multi-robot systems. Using Linear Temporal Logic (LTL), robots autonomously distribute new sub-tasks while still ensuring the satisfaction of their current tasks. Each robot evaluates its ability to satisfy both its current task and the new sub-tasks, then resynthesizes its behavior accordingly. We then present a novel task grammar that extends LTL to increase its expressivity for formulating collaborative tasks in a multi-robot context. Current approaches often require users to specify the numbers and types of robots for the tasks; in contrast, our task grammar focuses on the actions required and how those relate to the robot executing them (e.g. "the same robot that picked up the package must drop it off"). We also provide a synthesis framework and synchronization policies for the robots to collaborate with each other when required. The work in this dissertation provides approaches for both discrete and continuous actions. To increase robustness, this dissertation also includes a method for robots to replan and resynthesize their behavior in response to modifications in individual robot capabilities during execution. The replanning approach maintains task satisfaction while minimizing changes at both the global team assignment and local behavior levels. Finally, the dissertation presents a decentralized, context-based, on-board planning algorithm for Earth-Observation (EO) satellite systems. Each satellite first decides whether it can participate, then if it should participate, and finally either formally verifies a potential team, or synthesizes an optimal team for the mission.

Degree

thesis:*
Name thesis:degree_name
Ph. D., Mechanical Engineering
Level thesis:degree_level
Doctor of Philosophy
Discipline thesis:degree_discipline
Mechanical Engineering
Grantor
Cornell University
Year dc:date.issued
2024

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Fang, Amy
Chair dc:contributor.chair
  • Kress Gazit, Hadas
Committee members dc:contributor.committeemember
  • Campbell, Mark
  • Petersen, Kirstin

Subjects

dc:subject × 4

Rights

Language dc:language.iso
en

Identifiers

dc:identifier.*
Dc Identifier Bibid
16611839
Dc Identifier Other
ProQuest Submission ID: 14604
ProQuest Publication ID: 31489098

Chain of custody

source
Harvested from
Cornell University
Base URL
ecommons.cornell.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Fang, Amy. High-Level Collaborative Task Planning for Heterogeneous Multi-Robot Systems. Doctor of Philosophy thesis, Cornell University, 2024. https://hdl.handle.net/1813/116445