TY - GEN
T1 - Exponential deepening A∗ for real-time agent-centered search
AU - Sharon, Guni
AU - Felner, Ariel
AU - Sturtevant, Nathan R.
N1 - Publisher Copyright:
Copyright © 2014, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.
PY - 2014/1/1
Y1 - 2014/1/1
N2 - In the Real-Time Agent-Centered Search (RTACS) problem, an agent has to arrive at a goal location while acting and reasoning in the physical world. Traditionally, RTACS problems are solved by propagating and updating heuristic values of states visited by the agent. In existing RTACS algorithms the agent may revisit each state many times causing the entire procedure to be quadratic in the state space. We study the Iterative Deepening (ID) approach for solving RTACS and introduce Exponential Deepening A∗ (EDA∗), an RTACS algorithm where the threshold between successive Depth-First calls is increased exponentially. EDA∗ is proven to hold a worst case bound that is linear in the state space. Experimental results supporting this bound are presented and demonstrate up to 10x reduction over existing RTACS solvers wrt distance traveled, states expanded and CPU runtime.
AB - In the Real-Time Agent-Centered Search (RTACS) problem, an agent has to arrive at a goal location while acting and reasoning in the physical world. Traditionally, RTACS problems are solved by propagating and updating heuristic values of states visited by the agent. In existing RTACS algorithms the agent may revisit each state many times causing the entire procedure to be quadratic in the state space. We study the Iterative Deepening (ID) approach for solving RTACS and introduce Exponential Deepening A∗ (EDA∗), an RTACS algorithm where the threshold between successive Depth-First calls is increased exponentially. EDA∗ is proven to hold a worst case bound that is linear in the state space. Experimental results supporting this bound are presented and demonstrate up to 10x reduction over existing RTACS solvers wrt distance traveled, states expanded and CPU runtime.
UR - http://www.scopus.com/inward/record.url?scp=84908167974&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:84908167974
T3 - Proceedings of the National Conference on Artificial Intelligence
SP - 871
EP - 877
BT - Proceedings of the National Conference on Artificial Intelligence
PB - AI Access Foundation
T2 - 28th AAAI Conference on Artificial Intelligence, AAAI 2014, 26th Innovative Applications of Artificial Intelligence Conference, IAAI 2014 and the 5th Symposium on Educational Advances in Artificial Intelligence, EAAI 2014
Y2 - 27 July 2014 through 31 July 2014
ER -