Heuristic online goal recognition in continuous domains

Mor Vered, Gal A. Kaminka

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

8 Citations (Scopus)

Abstract

Goal recognition is the problem of inferring the goal of an agent, based on its observed actions. An inspiring approach—plan recognition by planning (PRP)—uses off-the-shelf planners to dynamically generate plans for given goals, eliminating the need for the traditional plan library. However, existing PRP formulation is inherently inefficient in online recognition, and cannot be used with motion planners for continuous spaces. In this paper, we utilize a different PRP formulation which allows for online goal recognition, and for application in continuous spaces. We present an online recognition algorithm, where two heuristic decision points may be used to improve run-time significantly over existing work. We specify heuristics for continuous domains, prove guarantees on their use, and empirically evaluate the algorithm over n hundreds of experiments in both a 3D navigational environment and a cooperative robotic team task.
Original languageEnglish
Title of host publicationProceedings of the 26th International Joint Conference on Artificial Intelligence
EditorsCarles Sierra
Place of PublicationMarina del Rey CA USA
PublisherInternational Joint Conferences on Artificial Intelligence
Pages4447-4454
Number of pages8
ISBN (Electronic)9780999241103
ISBN (Print)9780999241110
DOIs
Publication statusPublished - 2017
Externally publishedYes
EventInternational Joint Conference on Artificial Intelligence 2017 - Melbourne, Australia
Duration: 19 Aug 201725 Aug 2017
Conference number: 26th
https://ijcai-17.org/

Conference

ConferenceInternational Joint Conference on Artificial Intelligence 2017
Abbreviated titleIJCAI 2017
CountryAustralia
CityMelbourne
Period19/08/1725/08/17
Internet address

Keywords

  • Plan Recognition
  • Mirroring
  • Online
  • Continuous Domains

Cite this

Vered, M., & Kaminka, G. A. (2017). Heuristic online goal recognition in continuous domains. In C. Sierra (Ed.), Proceedings of the 26th International Joint Conference on Artificial Intelligence (pp. 4447-4454). Marina del Rey CA USA: International Joint Conferences on Artificial Intelligence. https://doi.org/10.24963/ijcai.2017/621
Vered, Mor ; Kaminka, Gal A. / Heuristic online goal recognition in continuous domains. Proceedings of the 26th International Joint Conference on Artificial Intelligence. editor / Carles Sierra. Marina del Rey CA USA : International Joint Conferences on Artificial Intelligence, 2017. pp. 4447-4454
@inproceedings{0608f296d06d4dceb901f5b6548cc73a,
title = "Heuristic online goal recognition in continuous domains",
abstract = "Goal recognition is the problem of inferring the goal of an agent, based on its observed actions. An inspiring approach—plan recognition by planning (PRP)—uses off-the-shelf planners to dynamically generate plans for given goals, eliminating the need for the traditional plan library. However, existing PRP formulation is inherently inefficient in online recognition, and cannot be used with motion planners for continuous spaces. In this paper, we utilize a different PRP formulation which allows for online goal recognition, and for application in continuous spaces. We present an online recognition algorithm, where two heuristic decision points may be used to improve run-time significantly over existing work. We specify heuristics for continuous domains, prove guarantees on their use, and empirically evaluate the algorithm over n hundreds of experiments in both a 3D navigational environment and a cooperative robotic team task.",
keywords = "Plan Recognition, Mirroring, Online, Continuous Domains",
author = "Mor Vered and Kaminka, {Gal A.}",
year = "2017",
doi = "10.24963/ijcai.2017/621",
language = "English",
isbn = "9780999241110",
pages = "4447--4454",
editor = "Carles Sierra",
booktitle = "Proceedings of the 26th International Joint Conference on Artificial Intelligence",
publisher = "International Joint Conferences on Artificial Intelligence",

}

Vered, M & Kaminka, GA 2017, Heuristic online goal recognition in continuous domains. in C Sierra (ed.), Proceedings of the 26th International Joint Conference on Artificial Intelligence. International Joint Conferences on Artificial Intelligence, Marina del Rey CA USA, pp. 4447-4454, International Joint Conference on Artificial Intelligence 2017, Melbourne, Australia, 19/08/17. https://doi.org/10.24963/ijcai.2017/621

Heuristic online goal recognition in continuous domains. / Vered, Mor; Kaminka, Gal A.

Proceedings of the 26th International Joint Conference on Artificial Intelligence. ed. / Carles Sierra. Marina del Rey CA USA : International Joint Conferences on Artificial Intelligence, 2017. p. 4447-4454.

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

TY - GEN

T1 - Heuristic online goal recognition in continuous domains

AU - Vered, Mor

AU - Kaminka, Gal A.

PY - 2017

Y1 - 2017

N2 - Goal recognition is the problem of inferring the goal of an agent, based on its observed actions. An inspiring approach—plan recognition by planning (PRP)—uses off-the-shelf planners to dynamically generate plans for given goals, eliminating the need for the traditional plan library. However, existing PRP formulation is inherently inefficient in online recognition, and cannot be used with motion planners for continuous spaces. In this paper, we utilize a different PRP formulation which allows for online goal recognition, and for application in continuous spaces. We present an online recognition algorithm, where two heuristic decision points may be used to improve run-time significantly over existing work. We specify heuristics for continuous domains, prove guarantees on their use, and empirically evaluate the algorithm over n hundreds of experiments in both a 3D navigational environment and a cooperative robotic team task.

AB - Goal recognition is the problem of inferring the goal of an agent, based on its observed actions. An inspiring approach—plan recognition by planning (PRP)—uses off-the-shelf planners to dynamically generate plans for given goals, eliminating the need for the traditional plan library. However, existing PRP formulation is inherently inefficient in online recognition, and cannot be used with motion planners for continuous spaces. In this paper, we utilize a different PRP formulation which allows for online goal recognition, and for application in continuous spaces. We present an online recognition algorithm, where two heuristic decision points may be used to improve run-time significantly over existing work. We specify heuristics for continuous domains, prove guarantees on their use, and empirically evaluate the algorithm over n hundreds of experiments in both a 3D navigational environment and a cooperative robotic team task.

KW - Plan Recognition

KW - Mirroring

KW - Online

KW - Continuous Domains

U2 - 10.24963/ijcai.2017/621

DO - 10.24963/ijcai.2017/621

M3 - Conference Paper

SN - 9780999241110

SP - 4447

EP - 4454

BT - Proceedings of the 26th International Joint Conference on Artificial Intelligence

A2 - Sierra, Carles

PB - International Joint Conferences on Artificial Intelligence

CY - Marina del Rey CA USA

ER -

Vered M, Kaminka GA. Heuristic online goal recognition in continuous domains. In Sierra C, editor, Proceedings of the 26th International Joint Conference on Artificial Intelligence. Marina del Rey CA USA: International Joint Conferences on Artificial Intelligence. 2017. p. 4447-4454 https://doi.org/10.24963/ijcai.2017/621