Distributed cooperative control of multi-UAV system based on consensus theory has been used widely in UAV operations in which the consistent state and collaborative demand are achieved through consensus control algorithm.The mathematical description of gathering problem was established the problem was solved under decomposition strategy based on the method of coordination variables and coordination function.To realize optimization of cooperative control average consensus control algorithm was improved the expression of basic optimized consensus control algorithm was given by Hamilton-Jacobi-Bellman equation.The method got better in terms of dynamic response and optimal cost by introducing outdated state difference to control algorithm.The weighted matrix of the cost function was optimized by genetic algorithm the dynamic response and optimal cost was improved further.In the meanwhile the task execution time was shortened.Theoretical analysis and simulation results verify the feasibility and effectiveness of the method.