Studying Online Multi-Agent Path Finding

Jonathan Morag; Roni Stern; Ariel Felner; Dor Atzmon; Eli Boyarski

Studying Online Multi-Agent Path Finding

Jonathan Morag
, Roni Stern
, Ariel Felner
, Dor Atzmon
, Eli Boyarski

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Multi-agent path finding (MAPF) is the problem of planning a set of non-conflicting plans on a graph, for a set of agents. Online MAPF extends MAPF by considering a more realistic problem in which new agents may appear over time. While planning, an online solver does not know whether and which agents will join in the future. Therefore, in online problems the notion of snapshot-optimal was defined, where only current knowledge is considered. The quality of such a solution may be weaker than the quality of a solution to an equivalent offline MAPF problem (offline-optimality), where the solver is preinformed of all the agents that will appear in the future. In this paper we explore, theoretically and empirically, the quality of snapshot-optimal solutions compared to offline-optimal solutions.

Original language	English
Title of host publication	14th International Symposium on Combinatorial Search, SoCS 2021
Editors	Hang Ma, Ivan Serina
Publisher	Association for the Advancement of Artificial Intelligence
Pages	228-230
Number of pages	3
ISBN (Electronic)	9781713834557
State	Published - 1 Jan 2021
Event	14th International Symposium on Combinatorial Search, SoCS 2021 - Guangzhou, Virtual, China Duration: 26 Jul 2021 → 30 Jul 2021

Publication series

Name	14th International Symposium on Combinatorial Search, SoCS 2021

Conference

Conference	14th International Symposium on Combinatorial Search, SoCS 2021
Country/Territory	China
City	Guangzhou, Virtual
Period	26/07/21 → 30/07/21

ASJC Scopus subject areas

Computer Networks and Communications

Cite this

@inproceedings{a487a128d10641ea80edda17738e052f,

title = "Studying Online Multi-Agent Path Finding",

abstract = "Multi-agent path finding (MAPF) is the problem of planning a set of non-conflicting plans on a graph, for a set of agents. Online MAPF extends MAPF by considering a more realistic problem in which new agents may appear over time. While planning, an online solver does not know whether and which agents will join in the future. Therefore, in online problems the notion of snapshot-optimal was defined, where only current knowledge is considered. The quality of such a solution may be weaker than the quality of a solution to an equivalent offline MAPF problem (offline-optimality), where the solver is preinformed of all the agents that will appear in the future. In this paper we explore, theoretically and empirically, the quality of snapshot-optimal solutions compared to offline-optimal solutions.",

author = "Jonathan Morag and Roni Stern and Ariel Felner and Dor Atzmon and Eli Boyarski",

note = "Publisher Copyright: Copyright {\textcopyright} 2021, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.; 14th International Symposium on Combinatorial Search, SoCS 2021 ; Conference date: 26-07-2021 Through 30-07-2021",

year = "2021",

month = jan,

day = "1",

language = "English",

series = "14th International Symposium on Combinatorial Search, SoCS 2021",

publisher = "Association for the Advancement of Artificial Intelligence",

pages = "228--230",

editor = "Hang Ma and Ivan Serina",

booktitle = "14th International Symposium on Combinatorial Search, SoCS 2021",

}

Morag, J, Stern, R , Felner, A, Atzmon, D & Boyarski, E 2021, Studying Online Multi-Agent Path Finding. in H Ma & I Serina (eds), 14th International Symposium on Combinatorial Search, SoCS 2021. 14th International Symposium on Combinatorial Search, SoCS 2021, Association for the Advancement of Artificial Intelligence, pp. 228-230, 14th International Symposium on Combinatorial Search, SoCS 2021, Guangzhou, Virtual, China, 26/07/21.

Studying Online Multi-Agent Path Finding. / Morag, Jonathan; Stern, Roni ; Felner, Ariel et al.
14th International Symposium on Combinatorial Search, SoCS 2021. ed. / Hang Ma; Ivan Serina. Association for the Advancement of Artificial Intelligence, 2021. p. 228-230 (14th International Symposium on Combinatorial Search, SoCS 2021).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Studying Online Multi-Agent Path Finding

AU - Morag, Jonathan

AU - Stern, Roni

AU - Felner, Ariel

AU - Atzmon, Dor

AU - Boyarski, Eli

PY - 2021/1/1

Y1 - 2021/1/1

N2 - Multi-agent path finding (MAPF) is the problem of planning a set of non-conflicting plans on a graph, for a set of agents. Online MAPF extends MAPF by considering a more realistic problem in which new agents may appear over time. While planning, an online solver does not know whether and which agents will join in the future. Therefore, in online problems the notion of snapshot-optimal was defined, where only current knowledge is considered. The quality of such a solution may be weaker than the quality of a solution to an equivalent offline MAPF problem (offline-optimality), where the solver is preinformed of all the agents that will appear in the future. In this paper we explore, theoretically and empirically, the quality of snapshot-optimal solutions compared to offline-optimal solutions.

AB - Multi-agent path finding (MAPF) is the problem of planning a set of non-conflicting plans on a graph, for a set of agents. Online MAPF extends MAPF by considering a more realistic problem in which new agents may appear over time. While planning, an online solver does not know whether and which agents will join in the future. Therefore, in online problems the notion of snapshot-optimal was defined, where only current knowledge is considered. The quality of such a solution may be weaker than the quality of a solution to an equivalent offline MAPF problem (offline-optimality), where the solver is preinformed of all the agents that will appear in the future. In this paper we explore, theoretically and empirically, the quality of snapshot-optimal solutions compared to offline-optimal solutions.

UR - https://www.scopus.com/pages/publications/85124609155

M3 - Conference contribution

AN - SCOPUS:85124609155

T3 - 14th International Symposium on Combinatorial Search, SoCS 2021

SP - 228

EP - 230

BT - 14th International Symposium on Combinatorial Search, SoCS 2021

A2 - Ma, Hang

A2 - Serina, Ivan

PB - Association for the Advancement of Artificial Intelligence

T2 - 14th International Symposium on Combinatorial Search, SoCS 2021

Y2 - 26 July 2021 through 30 July 2021

ER -

Studying Online Multi-Agent Path Finding

Abstract

Publication series

Conference

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this