Coalition formation is a process to form a group and solve a problem via cooperation. Because of the rising of network, each computing device can communicate through network. We can integrate those resources and use them by coalition formation. In this paper, we propose a new architecture which integrates case-based reasoning, negotiation, and reinforcement learning to improve the coalition formation process. Coalitions of our architecture suit for executing longterm task or for accomplishing a task in high mobility networks. Our experiments also show the advantage of our architecture in these two different environments.