C ONJUNTO DE TRABALHO PARA EXECUÇÃO DO ALGORITMO
6 CONSIDERAÇÕES FINAIS
6.2 RECOMENDAÇÕES PARA TRABALHOS FUTUROS
Como resultado deste trabalho, nós acreditamos que as seguintes modificações merecem uma investigação mais profunda e podem vir a contribuir bastante com este trabalho e com a área:
1. Reduzir o conjunto de trabalho, principalmente utilizando alocação dinâmica. Atualmente o conjunto de trabalho é todo previamente alocado e, a depender do número de agentes e nós, pode se tornar muito grande.
2. Investigar a transferência de dados via Peer-to-Peer entre GPUs, presente na versão mais nova de CUDA. Deve ser levada em
consideração a replicação do mapa para cada GPU bem como o escalonamento do trabalho entre elas.
3. Investigar a possibilidade de abordagens multiagente, em que cada agente pode reusar o caminho previamente calculado por outro agente. Esse tipo de abordagem diminuiria o custo com o cálculo de caminhos que já tenham sido computados.
4. Investigar se existe a possibilidade de utilizar outra abordagem de paralelização. A mais utilizada atualmente está relacionada com o mapeamento de 1 agente por thread.
5. Utilizar-se de ferramentas que permitam a visualização dos caminhos calculados para cada agente, assim como a simulação da navegação em tempo real.
6. O principal trabalho futuro está no desenvolvimento de um benchmark para testes. Com isso, seria possível padronizar os testes, utilizando os mesmos parâmetros e recursos (número de agentes, tamanho do mapa, placa de vídeo utilizada, etc.), com diferentes algoritmos.
REFERÊNCIAS
AILA, T.; LAINE, S. Understanding the efficiency of ray traversal on GPUs. In: CONFERENCE ON HIGH PERFORMANCE GRAPHICS, 2009, New York.
Proceedings… New York: ACM, 2009. p. 145-149.
AMD.Multi-core processors:the next evolution in computing. AMD Multi-Core Technology Whitepaper, [S.l.], 2005.
AMD. AMD Stream Technology. Disponível em: <http://www.amd.com/stream>. Acesso em: 22 jul. 2010.
ANDRADE, G. et al. Automatic computer game balancing: a reinforcement learning approach. In:INTERNATIONAL JOINT CONFERENCE ON AUTONOMOUS AGENTS AND MULTIAGENT SYSTEMS, 4., 2005, New York. Proceedings… New York: ACM, 2005.
ATANASOV, D. General Purpose GPU programming. In:INTERNATIONAL CONFERENCE ON COMPUTER SYSTEMS AND TECHNOLOGIES,2005. [Trabalho apresentado]… [S.l.: s.n.], 2005. v.11, p. 1-6.
ATARI. Space Invaders. 1978. Disponível em:<http://www.spaceinvaders.de/>. Acesso em: 25 set. 2013.
BJÖRNSSON, Y.; HALLDÓRSSON, K. Improved heuristics for optimal pathfinding on game maps. American Association for Artificial Intelligence, [S.l.], 2006.
BLEIWEISS, A. GPU accelerated pathfinding. In:ACM SYMPOSIUM ON GRAPHICS HARDWARE, 23., 2008, Saravejo, Bosnia and Herzegovina;EUROGRAPHICS CONFERENCE ON GRAPHICS HARDWARE, 2008. Proceedings…Switzerland: Eurographics Association, 2008,p. 65-74.
______. Multi agent navigation on the GPU. [S.l.]: NVIDIA Corporation, 2009.
BOURG, D.M.; SEEMANN, G. AI for game developers. [S.l.]: O’Reilly, 2004. BROWN, E.; CAIRNS, P. A grounded investigation of immersion. In:CHI, 2004, Vienna. Proceedings… Vienna: ACM Press, 2004. p. 1297-1300.
BULUÇ, A.; GILBERT, J.R.; BUDAK, C. Solving path problems on the GPU. Parallel Computing, The Netherlands, v. 36, n.5-6, p.241-253, 2010.
BUSTOS, B. et al. A graphics hardware accelerated algorithm for nearest neighbor search. [S.l.: s.n.], 2006.
CHAMPANDARD, A.J.AI game development: synthetic creatures with learning and reactive behaviors. [S.l.: s.n.],2003.
CORMEN, T. H. et al. Introduction to algorithms. Cambridge: MIT Press, 2001.
CSIKSZENTMIHALYI, M. Flow: the psychology of optimal experience. New York: Harper & Row, 1990.
CUDA. Disponível em:<http://www.nvidia.com/object/cuda_home_new.html>. Acesso em: 25 set. 2013.
DARKEN, C. J.; PAULL, G. H. Finding cover in dynamic environments. In: AI game programming wisdom 3. Boston:Charles River Media, 2006. p. 405-415.
EA SPORTS. FIFA Soccer 2010. Disponível em:<http://www.ea.com/games/fifa- soccer-10>. Acesso em: 25 set. 2013.
ERRA, U. et al. Massive simulation using GPU of a distributed behavioral model of a flock with obstacle avoidance. In: GIROD, B.; MAGNOR, M. A.; SEIDEL, H.-P. (Ed.).VMV. [S.l.]: Aka GmbH, 2004. p. 233-240.
EVEN, S. Graph algorithms. 2nd. ed. New York: Cambridge University Press, 2011.
FISHER, L. G.; SILVEIRA, R.; NEDEL, L. GPU accelerated path-planning for multi- agents in virtual environments. SBGames, [S.l.], 2009.
GALSTYAN, A.; KOLAR, S.; LERMAN, K. Resource allocation games with changing resource capacities. In:INTERNATIONAL CONFERENCE ON AUTONOMOUS AGENTS AND MULTI-AGENT SYSTEMS, 2003, Melbourne.
GABB, H.;LAKE, A. Threading 3D game engine basics. 2005. Disponível em:<http://www.gamasutra.com/view/feature/130873/threading_3d_game_engine_ basics.php>. Acesso em: 25 set. 2013.
GRAHAM, R.; MCCABE, H.; SHERIDAN, S. Pathfinding in computer games. ITB Journal, [S.l.], n. 8, 2003.
HARISH, P.; NARAYANAN, P. J. Accelerating large graphs algorithms on the GPU using CUDA. In: ALURU, S. et al (Ed.).HiPC, LNCS. Heidelberg: Springer, 2007. v. 4873, p. 197-208.
HART, P.; NILSSON, N. J.; RAPHAEL, B. A formal basis for the heuristic determination of minimum cost paths. IEEE Transaction on System Sciences and Cybernetics, [S.l.], v. 4, p. 100-107,1968.
HOLTE, R. C. et al. Hierarchical A*: searching abstraction hierarchies efficiently. In: NATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE, 13., 1996. Proceedings... [S.l.: s.n.], 1996. p. 530-535.
HUMPHRIES, J.Avatar vs Modern Warfare 2: the billion dollar behemoths. Business Management, 2010. Disponível
em:<http://www.bme.eu.com/news/avatar-vs-modern-warfare-2>. Acesso em: 22fev. 2010.
JANSEN, R.; STURTEVANT, N.A new approach to cooperative pathfinding. In:INTERNATIONAL JOINT CONFERENCE ON AUTONOMOUS AGENTS AND MULTIAGENTSYSTEMS,7., 2008, Richland, SC; Estoril, Portugal. Proceedings… [S.l.: s.n.], 2008. v. 3, p. 1401-1404.
KATZ, G.J.; KIDER Jr, J.T. All-Pairs shortest-paths for large graphs on the GPU. In: LUEBKE, David; OWENS, John D. (Ed.). Graphics hardware. Pennsylvania: [s.n.], 2008. p. 47-55.
KIDER, J. et al. High-dimensionalplanning on the GPU. In: IEEE INTERNATIONAL CONFERENCE ONROBOTICS AND AUTOMATION, 2010. [Trabalho
apresentado]…[S.l.: s.n.], 2010. p. 2515-2522.
LARSEN, E. S.; MCALLISTER, D. Fast matrix multiplies using graphics hardware. Supercomputing, [S.l.], 2001.
LAVALLE,S.M. Planning algorithms. Cambridge University Press,2006. 842 p. Disponível em: <http://msl.cs.uiuc.edu/planning/>. Acesso em: 25 set. 2013.
MATEAS, M. Expressive AI: games and artificial intelligence. [S.l.: s.n.],2003.
MICIKEVICIUS, P. General parallel computation on commodity graphics hardware: case study with the all-pairs shortest paths problem. In: ARABNIA, H. R.(Ed.). PDPTA. [S.l.]:CSREA Press, 2004. p. 1359-1365.
MONTRYM, J.;MORETON, H. The GeForce 6800. IEEE Micro, [S.l.], v. 25, n.2, p. 41-51, 2005.
NAREYEK, A. AI in computer games. Guest Researcher, Queue, v.1, n.10, 2004.
OWENS, J. et al. A survey of general-purpose computation on graphics hardware. Start of The Art Report, Eurographics, [S.l.], p. 21-51, 2005.
PATEL,A.Amit’s A* pages. 2008.Disponível em:
<http://theory.stanford.edu/~amitp/GameProgramming/>. Acesso em: 25set. 2013.
PHAM, D. et al. The design and implementation of a first-generation CELL processor. In:IEEE INTERNATIONALSOLID-STATE CIRCUIT CONFERENCE, 2005. [Trabalho apresentado]… [S.l.: s.n.], 2005.v. 1, p. 184-592.
PITTMAN, J. The Pac-man dossier. 2009. Disponível em:
<http://www.gamasutra.com/view/feature/132330/the_pacman_dossier.php>.
Acesso em: 25 set. 2013.
RABIN, S. A* aesthetic optimizations. Game Programming Gems,[S.l.], p. 264- 271,2000.
______. A* speed optimizations. Game Programming Gems,[S.l.], p. 272-287, 2000.
REYNOLDS,C. Big fast crowds on PS3, in sandbox’06. In: ACM SIGGRAPH SYMPOSIUM ON VIDEOGAMES,2006. Proceedings… New York: ACM Press, 2006. p.113-121.
ROCKSTAR GAMES. GTA: Grand Theft Auto V. Disponível em: <http://www.rockstargames.com/V/>. Acesso em: 25 set. 2013.
ROUSE 3rd R. What’s your perspective? Gaming and Graphics, New York, v. 33, n. 3,p. 9-12, Aug. 1999.
ŠERÝ, O. et al. Level-of-detail in behaviour of virtual humans. In: WIEDERMANN, J. et al.(Eds.) SOFSEM 2006. LNCS. Heidelberg: Springer, 2006.v. 3831, p. 565– 574.
SHOPF, J. et al. March of the Froblins: simulation and rendering of massive crowds of intelligence and detailed creatures on a GPU. [S.l.: s.n.], 2008.
SILVER, D. Cooperative pathfinding. American Association for Artificial Intelligence, [S.l.], 2005.
STOUT, B. The basics for A* path planning. Game Programming Gems, [S.l.], p. 254-263, 2000.
STENTZ, A. Optimal and efficient path planning for partially-known environments. In:IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION,1994. Proceedings… [S.l.: s.n.], 1994. p. 3310- 3317.
STENTZ, A.; LIKHACHEV, M. R* search.In:AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2008. Proceedings… [S.l.]: AAAI Press, 2008. v. 1, p. 344-350.
TATARCHUK, N. Richer worlds for next Gen Games: data amplification techniques Survey. GDC Europe, [S.l.], 2005.
TOZOUR, P. Search space representations. In: AI game programming wisdom 2. Hingham: Charles River Media,2003. p. 405-415.
TRAN, Q.N. Designing efficient many-core parallel algorithms for all-pairs shortest- paths using CUDA.In: INTERNATIONAL CONFERENCE ON INFORMATION TECHNOLOGYNEW GENERATIONS, 7., 2010, Washington. Proceedings… Washington: IEEE Computer Society, 2010. p. 7-12.
TREUILLE, A.;COOPER, S.;POPOVIC, Z.Continuum crowds.ACM Transaction onGraphics (TOG).In: SIGGRAPH, 2006. Proceedings... [S.l.: s.n.], 2006. v. 25, n.3, p.1160-1168.
TRINDADE, E. et al. Algoritmos de busca em tempo real aplicados a jogos digitais. In:SBGAMES, 2008, Belo Horizonte. Proceedings… Belo Horizonte: [s.n.], 2008. p.141-150.
VALVE. Left 4 Dead 2. 2009.Disponível em:<http://www.l4d.com/blog>. Acesso em: 25 set. 2013.
WANG, K.C.; BOTEA, A. Fast and memory-efficient multi-agent pathfinding. In: INTERNATIONAL CONFERENCE ON AUTOMATED PLANNING AND SCHEDULING, 8., 2008. Proceedings… [S.l.: s.n.], 2008.
WATT, A. 3D computer graphics.New York: Pearson, 2000.
YANG, R.; WELCH, G. Fast image segmentation and smoothing using commodity graphics hardware. Journal of Graphics Tools, [S.l.], p.91-100.2003.Special issue on Hardware-Accelerated Rendering Techniques.