In order to tackle the shortage of existing researches in Coalition Generation Problem

a novel Complicated Coalition and a novel Virtual Agent are proposed.Then a Particle Swarms Cooperative Optimization algorithm applied to solving serial Complicated Coalition Generation Problem is put forward on the basis of the analysis of the standard discrete particle swarm optimization.The new algorithm realizes the condition that an Agent can take part in several different coalitions and a coalition can turn its hand to several different tasks and partly reduces the waste of resource and ability.The effectiveness of the algorithm is proved by experiments.