Landmark-Based Heuristics for Goal Recognition

Ramon Fraga Pereira, Nir Oren, Felipe Meneguzzi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

14 Citations (Scopus)
9 Downloads (Pure)

Abstract

Automated planning can be used to efficiently recognize goals and plans from partial or full observed action sequences. In this paper, we propose goal recognition heuristics that rely on information from planning landmarks — facts or actions that must occur if a plan is to achieve a goal when starting from some initial state. We develop two such heuristics: the first estimates goal completion by considering the ratio between achieved and extracted landmarks of a candidate goal, while the second takes into account how unique each landmark is among landmarks for all candidate goals. We empirically evaluate these heuristics over both standard goal/plan recognition problems, and a set of very large problems. We show that our heuristics can recognize goals more accurately, and run orders of magnitude faster, than the current state-of-the-art.
Original languageEnglish
Title of host publicationThirty-First AAAI Conference on Artificial Intelligence (AAAI-17)
PublisherAAAI Press
Pages3622-3628
Number of pages7
Publication statusPublished - Aug 2017
EventThirty-First AAAI Conference on Artificial Intelligence (AAAI-17) - San Francisco, United States
Duration: 4 Feb 20179 Feb 2017

Publication series

NameAAAI Conference on Artificial Intelligence (AAAI)
PublisherAAAI
ISSN (Print)2159-5399
ISSN (Electronic)2374-3468

Conference

ConferenceThirty-First AAAI Conference on Artificial Intelligence (AAAI-17)
CountryUnited States
CitySan Francisco
Period4/02/179/02/17

Fingerprint

Planning

Cite this

Pereira, R. F., Oren, N., & Meneguzzi, F. (2017). Landmark-Based Heuristics for Goal Recognition. In Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17) (pp. 3622-3628). (AAAI Conference on Artificial Intelligence (AAAI)). AAAI Press.

Landmark-Based Heuristics for Goal Recognition. / Pereira, Ramon Fraga ; Oren, Nir; Meneguzzi, Felipe.

Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17). AAAI Press, 2017. p. 3622-3628 (AAAI Conference on Artificial Intelligence (AAAI)).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Pereira, RF, Oren, N & Meneguzzi, F 2017, Landmark-Based Heuristics for Goal Recognition. in Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17). AAAI Conference on Artificial Intelligence (AAAI), AAAI Press, pp. 3622-3628, Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17), San Francisco, United States, 4/02/17.
Pereira RF, Oren N, Meneguzzi F. Landmark-Based Heuristics for Goal Recognition. In Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17). AAAI Press. 2017. p. 3622-3628. (AAAI Conference on Artificial Intelligence (AAAI)).
Pereira, Ramon Fraga ; Oren, Nir ; Meneguzzi, Felipe. / Landmark-Based Heuristics for Goal Recognition. Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17). AAAI Press, 2017. pp. 3622-3628 (AAAI Conference on Artificial Intelligence (AAAI)).
@inproceedings{9d2d0df66dcc430cb9c1d53d9ceeac1a,
title = "Landmark-Based Heuristics for Goal Recognition",
abstract = "Automated planning can be used to efficiently recognize goals and plans from partial or full observed action sequences. In this paper, we propose goal recognition heuristics that rely on information from planning landmarks — facts or actions that must occur if a plan is to achieve a goal when starting from some initial state. We develop two such heuristics: the first estimates goal completion by considering the ratio between achieved and extracted landmarks of a candidate goal, while the second takes into account how unique each landmark is among landmarks for all candidate goals. We empirically evaluate these heuristics over both standard goal/plan recognition problems, and a set of very large problems. We show that our heuristics can recognize goals more accurately, and run orders of magnitude faster, than the current state-of-the-art.",
author = "Pereira, {Ramon Fraga} and Nir Oren and Felipe Meneguzzi",
note = "6 volumes http://www.aaai.org/Press/Proceedings/aaai17.php",
year = "2017",
month = "8",
language = "English",
series = "AAAI Conference on Artificial Intelligence (AAAI)",
publisher = "AAAI Press",
pages = "3622--3628",
booktitle = "Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17)",

}

TY - GEN

T1 - Landmark-Based Heuristics for Goal Recognition

AU - Pereira, Ramon Fraga

AU - Oren, Nir

AU - Meneguzzi, Felipe

N1 - 6 volumes http://www.aaai.org/Press/Proceedings/aaai17.php

PY - 2017/8

Y1 - 2017/8

N2 - Automated planning can be used to efficiently recognize goals and plans from partial or full observed action sequences. In this paper, we propose goal recognition heuristics that rely on information from planning landmarks — facts or actions that must occur if a plan is to achieve a goal when starting from some initial state. We develop two such heuristics: the first estimates goal completion by considering the ratio between achieved and extracted landmarks of a candidate goal, while the second takes into account how unique each landmark is among landmarks for all candidate goals. We empirically evaluate these heuristics over both standard goal/plan recognition problems, and a set of very large problems. We show that our heuristics can recognize goals more accurately, and run orders of magnitude faster, than the current state-of-the-art.

AB - Automated planning can be used to efficiently recognize goals and plans from partial or full observed action sequences. In this paper, we propose goal recognition heuristics that rely on information from planning landmarks — facts or actions that must occur if a plan is to achieve a goal when starting from some initial state. We develop two such heuristics: the first estimates goal completion by considering the ratio between achieved and extracted landmarks of a candidate goal, while the second takes into account how unique each landmark is among landmarks for all candidate goals. We empirically evaluate these heuristics over both standard goal/plan recognition problems, and a set of very large problems. We show that our heuristics can recognize goals more accurately, and run orders of magnitude faster, than the current state-of-the-art.

UR - http://www.aaai.org/Press/Proceedings/aaai17.php

M3 - Conference contribution

T3 - AAAI Conference on Artificial Intelligence (AAAI)

SP - 3622

EP - 3628

BT - Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17)

PB - AAAI Press

ER -