TY - GEN
T1 - Improving HLRTA
AU - Hernández, Carlos
AU - Meseguer, Pedro
PY - 2007
Y1 - 2007
N2 - Real-time search methods allow an agent to move in unknown environments. We provide two enhancements to the real-time search algorithm HLRTA*(k). First, we give a better way to perform bounded propagation, generating the HLRTA*LS(k) algorithm. Second, we consider the option of doing more than one action per planning step, by analyzing the quality of the heuristic found during lookahead, producing the HLRTA*(k, d) algorithm. We provide experimental evidence of the benefits of both algorithms, with respect to other real-time algorithms on existing benchmarks.
AB - Real-time search methods allow an agent to move in unknown environments. We provide two enhancements to the real-time search algorithm HLRTA*(k). First, we give a better way to perform bounded propagation, generating the HLRTA*LS(k) algorithm. Second, we consider the option of doing more than one action per planning step, by analyzing the quality of the heuristic found during lookahead, producing the HLRTA*(k, d) algorithm. We provide experimental evidence of the benefits of both algorithms, with respect to other real-time algorithms on existing benchmarks.
UR - http://www.scopus.com/inward/record.url?scp=38349043918&partnerID=8YFLogxK
U2 - 10.1007/978-3-540-75271-4_12
DO - 10.1007/978-3-540-75271-4_12
M3 - Conference contribution
AN - SCOPUS:38349043918
SN - 9783540752707
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 110
EP - 119
BT - Current Topics in Artificial Intelligence - 12th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2007
PB - Springer Verlag
T2 - 12th Conference of the Spanish Association for Artificial Intelligence and its associated Conference on Technology Transfer on Artificial Intelligence, CAEPIA/TTIA 2007
Y2 - 12 November 2007 through 16 November 2007
ER -