College of Information Engineering, China Jiliang University, Hangzhou, Zhejiang, China

---

Abstract—Negotiation is one of the key issues in Multi-agent system. In terms of task characteristics, on the foundation of the multi-agent negotiation model based on the acquaintance model, a task oriented extended contract-net negotiation model is proposed after analyzing the advantage and disadvantage of the classical contract-net negotiation model. Based on the model’s architecture definition, a formalized coalition definition based on task and acquaintance model and coalition formation algorithm is given, and then an extended contract-net negotiation model based on task coalition is constructed. A combination of GA and simulated annealing algorithm is used to optimize the allocation speed and quality of task coalition in the agent coalition. Experiment shows that the presented model can effectively reduce the negotiation cost and improve the efficiency of task allocation on the basis of ensuring the negotiation quality.

Index Terms—Multi-agent negotiation, task model, coalition, genetic algorithm, simulated annealing.

[PDF]

Cite: Xiang Lin, Haijun Tao, "A Task Oriented Multi-Agent Negotiation Model Based on Contract-Net," Journal of Software vol. 8, no. 9, pp. 2280-2288, 2013.