Um Algoritmo para Extração de um Plano BDI que Obedece uma Política MDP Ótima
Resumo
Neste artigo, apresenta-se uma análise da proposta de trabalhar com uma abordagem híbrida BDI-MDP encontrada na literatura, e introduz-se um algoritmo que obtém planos para agentes BDI que obedecem uma política ótima, apresentando um exemplo ilustrativo de sua aplicação. Políticas são mapeadas para planos de agentes BDI, de acordo com uma dada intenção, considerando que os planos derivados de uma política ótima são aqueles adotados pelo agente BDI que seleciona o plano com a maior utilidade, e uma reconsideração de estratégia ótima. Apresenta-se também uma versão computacional deste mapeamento, na forma de um algoritmo de busca que segue uma dada política através do espaço de estados.
Referências
Boutilier, C., Dean, T., and Hanks, S. (1999). Decision theoretic planning: Structural assumptions and computational leverage. Journal Artificial Intelligence Res., 10.
Bratman, M. E., Israel, D. J., and Pollack, M. E. (1988). Plans and resource-bounded practical reasoning. Computational Intelligence, 4(4):349–355.
Gupta, T., Varakantham, P., Rauenbusch, T. W., and Tambe, M. (2007). Demonstration of teamwork in uncertain domains using hybrid bdi-pomdp systems. In 6th Intl. Conf. Autonomous Agents and Multi-Agent Systems, Demo Track.
Lovejoy, W. S. (1991). A survey of algorithmic methods for partially observed Markov decision processes. Annals of Operations Research, 28(1–4):47–66.
Meuleau, N., Hauskrecht, M., Kim, K.-E., Peshkin, L., Kaelbling, L. P., Dean, T., and Boutilier, C. (1998). Solving very large weakly coupled markov decision processes. In AAAI/IAAI, pages 165–172.
Nair, R. and Tambe, M. (2005). Hybrid BDI-POMDP framework for multiagent teaming. Journal of Artificial Intelligence Research, 23:367–420.
Parr, R. (1998). Flexible decomposition algorithms for weakly coupled markov decision problems. In Cooper, G. F. and Moral, S., editors, Proc. 14th Conf. Uncertainty in Artificial Intelligence, Madison, pages 422–430. Morgan Kaufmann.
Paruchuri, P., Bowring, E., Nair, R., Pearce, J., Schurr, N., Tambe, M., and Varakantham, P. (2006). Multiagent teamwork: Hybrid approaches. In Computer society of India Communications. CSI. (Invited Talk, avaliable at [link]).
Rao, A. S. and Georgeff, M. P. (1992). An abstract architecture for rational agents. In KR, pages 439–449.
Russell, S. J. and Norvig, P. (1995). Artificial inteligence : A modern approach.
Simari, G. I. and Parsons, S. (2004). On approximating the best decision for and autonomous agent. In Proc. 6th Work. Game Theoretic Decision Agents, pages 91–100.
Simari, G. I. and Parsons, S. (2006). On the relationship between MDPs and the BDI architecture. In Nakashima, H., Wellman, M. P., Weiss, G., and Stone, P., editors, 5th International Joint Conference on Autonomous Agents and Multiagent Systems (AAMAS 2006), Hakodate, Japan, May 8-12, 2006, pages 1041–1048. ACM.
Bratman, M. E., Israel, D. J., and Pollack, M. E. (1988). Plans and resource-bounded practical reasoning. Computational Intelligence, 4(4):349–355.
Gupta, T., Varakantham, P., Rauenbusch, T. W., and Tambe, M. (2007). Demonstration of teamwork in uncertain domains using hybrid bdi-pomdp systems. In 6th Intl. Conf. Autonomous Agents and Multi-Agent Systems, Demo Track.
Lovejoy, W. S. (1991). A survey of algorithmic methods for partially observed Markov decision processes. Annals of Operations Research, 28(1–4):47–66.
Meuleau, N., Hauskrecht, M., Kim, K.-E., Peshkin, L., Kaelbling, L. P., Dean, T., and Boutilier, C. (1998). Solving very large weakly coupled markov decision processes. In AAAI/IAAI, pages 165–172.
Nair, R. and Tambe, M. (2005). Hybrid BDI-POMDP framework for multiagent teaming. Journal of Artificial Intelligence Research, 23:367–420.
Parr, R. (1998). Flexible decomposition algorithms for weakly coupled markov decision problems. In Cooper, G. F. and Moral, S., editors, Proc. 14th Conf. Uncertainty in Artificial Intelligence, Madison, pages 422–430. Morgan Kaufmann.
Paruchuri, P., Bowring, E., Nair, R., Pearce, J., Schurr, N., Tambe, M., and Varakantham, P. (2006). Multiagent teamwork: Hybrid approaches. In Computer society of India Communications. CSI. (Invited Talk, avaliable at [link]).
Rao, A. S. and Georgeff, M. P. (1992). An abstract architecture for rational agents. In KR, pages 439–449.
Russell, S. J. and Norvig, P. (1995). Artificial inteligence : A modern approach.
Simari, G. I. and Parsons, S. (2004). On approximating the best decision for and autonomous agent. In Proc. 6th Work. Game Theoretic Decision Agents, pages 91–100.
Simari, G. I. and Parsons, S. (2006). On the relationship between MDPs and the BDI architecture. In Nakashima, H., Wellman, M. P., Weiss, G., and Stone, P., editors, 5th International Joint Conference on Autonomous Agents and Multiagent Systems (AAMAS 2006), Hakodate, Japan, May 8-12, 2006, pages 1041–1048. ACM.
Publicado
16/04/2007
Como Citar
PEREIRA, Diego R.; DIMURO, Graçaliz P..
Um Algoritmo para Extração de um Plano BDI que Obedece uma Política MDP Ótima. In: WORKSHOP-ESCOLA DE SISTEMAS DE AGENTES, SEUS AMBIENTES E APLICAÇÕES (WESAAC), 1. , 2007, Pelotas/RS.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2007
.
p. 142-154.
ISSN 2326-5434.
DOI: https://doi.org/10.5753/wesaac.2007.33041.
