Online goal recognition as reasoning over landmarks

M. Vered, R. F. Pereira, M. C. Magnaguagno, F. Meneguzzi, G. A. Kaminka

Research output: Contribution to journalArticlepeer-review


Online goal recognition is the problem of recognizing the goal of an agent based on an incomplete sequence of observations with as few observations as possible. Recognizing goals with minimal domain knowledge as an agent executes its plan requires efficient algorithms to sift through a large space of hypotheses. 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 and substantially more accurate than the state-of-the-art.
Original languageEnglish
Pages (from-to)638-645
Number of pages8
JournalThe Workshops of the Thirty-Second AAAI Conference on Artificial Intelligence
StatePublished - 2018


Dive into the research topics of 'Online goal recognition as reasoning over landmarks'. Together they form a unique fingerprint.

Cite this