Ghent University Academic Bibliography

Advanced

Hybrid path planning for massive crowd simulation on the GPU

Aljosha Demeulemeester UGent, Charles Hollemeersch UGent, Pieter Mees UGent, Bart Pieters UGent, Peter Lambert UGent and Rik Van de Walle UGent (2011) Lecture Notes in Computer Science. 7060. p.304-315
abstract
In modern day games, it is often desirable to have many agents navigating intelligently through detailed environments. However, intelligent navigation remains a computationally expensive and complicated problem. In the past, the continuum crowds algorithm demonstrated the value of using a dynamic potential field to guide many agents to a common goal location. However this algorithm is prohibitively resource intensive for real time applications using large and detailed virtual worlds. In this paper, we propose a novel hybrid system that first uses a coarse A* path finding step. This helps to eliminate unnecessary work during the potential field generation by excluding areas of the world from the potential field calculation. Additionally, we show how an optimized potential field solver can be implemented on the GPU using the concepts of persistent threads and inter-block communication. Results show that our system achieves considerable speedups compared to existing path planning systems and that up to 100,000 agents can be simulated and rendered in real time on a mainstream GPU.
Please use this url to cite or link to this publication:
author
organization
year
type
conference
publication status
published
subject
keyword
GPU Acceleration, Real-time, Crowd Simulation, Path Planning
in
Lecture Notes in Computer Science
Lect. Notes Comput. Sci.
editor
Jan M Allbeck and Petros Faloutsos
volume
7060
issue title
Motion in games
pages
304 - 315
publisher
Springer
place of publication
Berlin, Germany
conference name
4th International conference on Motion in Games (MIG 2011)
conference location
Edinburgh, Scotland, UK
conference start
2011-11-13
conference end
2011-11-15
Web of Science type
Conference Paper
Web of Science id
12510177
ISSN
0302-9743
ISBN
9783642250903
9783642250897
DOI
10.1007/978-3-642-25090-3_26
language
English
UGent publication?
yes
classification
C1
copyright statement
I have transferred the copyright for this publication to the publisher
VABB id
c:vabb:339835
VABB type
VABB-5
id
2035755
handle
http://hdl.handle.net/1854/LU-2035755
date created
2012-02-16 15:10:02
date last changed
2012-03-15 13:27:09
@inproceedings{2035755,
  abstract     = {In modern day games, it is often desirable to have many agents navigating intelligently through detailed environments. However, intelligent navigation remains a computationally expensive and complicated problem. In the past, the continuum crowds algorithm demonstrated the value of using a dynamic potential field to guide many agents to a common goal location. However this algorithm is prohibitively resource intensive for real time applications using large and detailed virtual worlds. In this paper, we propose a novel hybrid system that first uses a coarse A* path finding step. This helps to eliminate unnecessary work during the potential field generation by excluding areas of the world from the potential field calculation. Additionally, we show how an optimized potential field solver can be implemented on the GPU using the concepts of persistent threads and inter-block communication. Results show that our system achieves considerable speedups compared to existing path planning systems and that up to 100,000 agents can be simulated and rendered in real time on a mainstream GPU.},
  author       = {Demeulemeester, Aljosha and Hollemeersch, Charles and Mees, Pieter and Pieters, Bart and Lambert, Peter and Van de Walle, Rik},
  booktitle    = {Lecture Notes in Computer Science},
  editor       = {Allbeck, Jan M and Faloutsos, Petros },
  isbn         = {9783642250903},
  issn         = {0302-9743},
  keyword      = {GPU Acceleration,Real-time,Crowd Simulation,Path Planning},
  language     = {eng},
  location     = {Edinburgh, Scotland, UK},
  pages        = {304--315},
  publisher    = {Springer},
  title        = {Hybrid path planning for massive crowd simulation on the GPU},
  url          = {http://dx.doi.org/10.1007/978-3-642-25090-3\_26},
  volume       = {7060},
  year         = {2011},
}

Chicago
Demeulemeester, Aljosha, Charles Hollemeersch, Pieter Mees, Bart Pieters, Peter Lambert, and Rik Van de Walle. 2011. “Hybrid Path Planning for Massive Crowd Simulation on the GPU.” In Lecture Notes in Computer Science, ed. Jan M Allbeck and Petros Faloutsos, 7060:304–315. Berlin, Germany: Springer.
APA
Demeulemeester, A., Hollemeersch, C., Mees, P., Pieters, B., Lambert, P., & Van de Walle, R. (2011). Hybrid path planning for massive crowd simulation on the GPU. In J. M. Allbeck & P. Faloutsos (Eds.), Lecture Notes in Computer Science (Vol. 7060, pp. 304–315). Presented at the 4th International conference on Motion in Games (MIG 2011), Berlin, Germany: Springer.
Vancouver
1.
Demeulemeester A, Hollemeersch C, Mees P, Pieters B, Lambert P, Van de Walle R. Hybrid path planning for massive crowd simulation on the GPU. In: Allbeck JM, Faloutsos P, editors. Lecture Notes in Computer Science. Berlin, Germany: Springer; 2011. p. 304–15.
MLA
Demeulemeester, Aljosha, Charles Hollemeersch, Pieter Mees, et al. “Hybrid Path Planning for Massive Crowd Simulation on the GPU.” Lecture Notes in Computer Science. Ed. Jan M Allbeck & Petros Faloutsos. Vol. 7060. Berlin, Germany: Springer, 2011. 304–315. Print.