TY - GEN
T1 - Flexible path planning using corridor maps
AU - Overmars, Mark
AU - Karamouzas, Ioannis
AU - Geraerts, Roland
N1 - Copyright:
Copyright 2021 Elsevier B.V., All rights reserved.
PY - 2008
Y1 - 2008
N2 - Path planning is a central problem in virtual environments and games. When computer-controlled characters move around in virtual worlds they have to plan their paths to desired locations. These paths must avoid collisions with the environment and with other moving characters. Also a chosen path must be natural, meaning that it is the kind of path a real human being could take. The algorithms for planning such paths must be able to handle hundreds of characters in real-time and must be flexible. The Corridor Map Method (cmm) was recently introduced as a flexible path planning method in interactive virtual environments and games. The method is fast and flexible and the resulting paths are reasonable. However, the paths tend to take unnatural turns when characters get close to other characters or small obstacles. In this paper we will improve on the cmm by decoupling collision avoidance with the environment and local steering behavior. The result is a method that keeps the advantages of the cmm but has much more natural steering. Also the method allows for more flexibility in the desired routes of the characters.
AB - Path planning is a central problem in virtual environments and games. When computer-controlled characters move around in virtual worlds they have to plan their paths to desired locations. These paths must avoid collisions with the environment and with other moving characters. Also a chosen path must be natural, meaning that it is the kind of path a real human being could take. The algorithms for planning such paths must be able to handle hundreds of characters in real-time and must be flexible. The Corridor Map Method (cmm) was recently introduced as a flexible path planning method in interactive virtual environments and games. The method is fast and flexible and the resulting paths are reasonable. However, the paths tend to take unnatural turns when characters get close to other characters or small obstacles. In this paper we will improve on the cmm by decoupling collision avoidance with the environment and local steering behavior. The result is a method that keeps the advantages of the cmm but has much more natural steering. Also the method allows for more flexibility in the desired routes of the characters.
UR - http://www.scopus.com/inward/record.url?scp=57749195051&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=57749195051&partnerID=8YFLogxK
U2 - 10.1007/978-3-540-87744-8_1
DO - 10.1007/978-3-540-87744-8_1
M3 - Conference contribution
AN - SCOPUS:57749195051
SN - 3540877436
SN - 9783540877431
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 1
EP - 12
BT - Algorithms - ESA 2008 - 16th Annual European Symposium, Proceedings
PB - Springer Verlag
T2 - 16th Annual European Symposium on Algorithms, ESA 2008
Y2 - 15 September 2008 through 17 September 2008
ER -