Fighting fire with agents: an agent coordination model for simulated firefighting

ABSTRACT

In this paper we propose a model for coordinating teams of computational agents. This model is especially aimed for coordinating agents performing in a simulated environment of forest firefighting, although it may be used in other domains. We will start by introducing the Pyrosim platform where we are carrying out our experiments. Pyrosim is a tool developed in our laboratory that simulates a forest fire environment where software agents act under the role of firefighters that have to cooperate in order to control the fire. We will proceed by presenting a model for team coordination. With this model it is possible to define firefighting tactics that originate different team approaches to the fire. These tactics are conducted by a single agent (the Leader) that communicates high level tasks to the other agents. Agents have local autonomy and are able of cooperating locally for carrying out their tasks without using communication. Finally, we will present some results of our experiments using the proposed coordination model in two different scenarios. We will use these results to address the problem of automatic tactic selection where we are currently working on.

REFERENCES

• 1: Divisão da Defesa da Floresta Contra Incêndios. Incêndios florestais -- relatório de 2005. Technical report, Ministério da Agricultura do Desenvolvimento Rural e das Pescas, Portugal, January 2006. http://www.dgrf.minagricultura.pt/v4/dgf/pub.php?ndx=2271.
• 2: Marco Wiering, Fillipo Mignogna, and Bernard Maassen. Evolving neural networks for forest fire control. In M. van Otterlo, M. Poel, and A. Nijholt, editors, Benelearn '05: Proceedings of the 14th Belgian-Dutch Conference on Machine Learning, pages 113--120, 2005.
• 3: Marco Wiering and Marco Dorigo. Learning to control forest fires. In H. Haasis and K. Ranze, editors, Proceedings of the 12th international Symposium on 'Computer Science for Environmental Protection', pages 378--388, 1998.
• 4: Luís Sarmento. An emotion-based agent architecture. Master's thesis, Faculdade de Ciências da Universidade do Porto, 2004.
• 5: Marc Nicolas and Grant Beebe. The training of forest firefighters in indonesia. Technical report, German Agency for Technical Cooperation, European Union, and Government of Indonesia, 1999.
• 6: Peter Stone , Manuela Veloso, Task decomposition, dynamic role assignment, and low-bandwidth communication for real-time strategic teamwork, Artificial Intelligence, v.110 n.2, p.241-273, June 1999 [doi>10.1016/S0004-3702(99)00025-9]
• 7: Peter Herald Stone , Manuela M. Veloso, Layered learning in multiagent systems, Carnegie Mellon University, Pittsburgh, PA, 1998
• 8: Luís Paulo Reis. Coordenação em Sistemas Multi-Agente: Aplicações na Gestão Universitária e Futebol Robótico. PhD thesis, Faculdade de Engenharia da Universidade do Porto, Porto, Portugal, June 2003.
• 9: Luís Paulo Reis , Nuno Lau , Eugenio Oliveira, Situation Based Strategic Positioning for Coordinating a Team of Homogeneous Agents, Balancing Reactivity and Social Deliberation in Multi-Agent Systems, From RoboCup to Real-World Applications (selected papers from the ECAI 2000 Workshop and additional contributions), p.175-197, January 2001
• 10: Luís Paulo Reis , Nuno Lau, FC Portugal Team Description: RoboCup 2000 Simulation League Champion, RoboCup 2000: Robot Soccer World Cup IV, p.29-40, January 2001
• 11: Daniel Moura. Coordinating a team of agents in the forest firefighting domain. Master's thesis, Faculdade de Engenharia da Universidade do Porto, 2006. http://dcm.web.simplesnet.pt/publications/MSc_dcm_06.pdf.

INDEX TERMS

Keywords:
coordination, firefighting, multi-agent systems, teamwork