Gal A. Kaminka's Publications

Sorted by DateClassified by Publication TypeClassified by TopicGrouped by Student (current)Grouped by Former Students

Online Goal Recognition Combining Landmarks and Planning

Mor Vered, Ramon Fraga Pereira, Maur\'icio Cec\'ilio Magnaguagno, Felipe Meneguzzi, and Gal A. Kaminka. Online Goal Recognition Combining Landmarks and Planning. In IJCAI Workshop on Goal Reasoning, 2017.

Download

(unavailable)

Abstract

Goal recognition is the problem of recognizing the goal of an agent based on an incomplete sequence of observations. Recognizing goals with minimal domain knowledge as an agent executes its plan requires efficient algorithms to sift through a large space of hypotheses. In this paper, we develop an online approach to recognize goals in both continuous and discrete domains using a combination of goal mirroring and a generalized notion of landmarks adapted from the planning literature. Extensive experiments demonstrate the approach is more efficient than state-of-the-art online recognition, and substantially more accurate.

Additional Information

BibTeX

@inproceedings{goalreason17ws-felipe,
	author = {Mor Vered and Ramon Fraga Pereira and Maur\'{i}cio Cec\'{i}lio Magnaguagno and Felipe Meneguzzi and Gal A. Kaminka},
	title = {Online Goal Recognition Combining Landmarks and Planning},
	booktitle = {{IJCAI} Workshop on Goal Reasoning},
	OPTcrossref = {crossref},
	OPTkey = {key},
	OPTpages = {pages},
	year = {2017},
	OPTeditor = {editor},
	OPTvolume = {volume},
	OPTnumber = {number},
	OPTseries = {series},
	OPTaddress = {address},
	OPTmonth = {month},
	OPTorganization = {organization},
	OPTpublisher = {publisher},
	OPTnote = {},
	OPTannote = {annote},
	wwwnote = {}, 
	OPTkeywords = {},
	abstract = {
	Goal recognition is the problem of recognizing the 
	goal of an agent based on an incomplete sequence 
	of observations. Recognizing goals with minimal 
	domain knowledge as an agent executes its plan requires 
	efficient algorithms to sift through a large 
	space of hypotheses. In this paper, we develop an 
	online approach to recognize goals in both continuous 
	and discrete domains using a combination of 
	goal mirroring and a generalized notion of landmarks 
	adapted from the planning literature. Extensive 
	experiments demonstrate the approach is more efficient 
	than state-of-the-art online recognition, and 
	substantially more accurate.
	},	
}

Generated by bib2html.pl (written by Patrick Riley ) on Sat Feb 24, 2018 00:31:02