针对通信时延与梯度时延共存下的联盟博弈，提出基于对偶平均技术与时延梯度的分布式对偶平均算法来求解纳什均衡。采用增广图方法表征通信时延与利用布雷格曼散度度量时延梯度与当前梯度之间的误差，理论分析表明提出的分布式对偶平均算法以次线性收敛率收敛至纳什均衡。同时研究结果阐明通信时延与梯度时延对算法收敛误差的影响。最后，将所提出的分布式对偶平均算法应用到无人机集群的编队控制中验证算法的有效性。

To address coalition games under communication and gradient delays, it proposes a distributed algorithm based on the dual average and delayed gradient to seek the Nash equilibrium. With the help of augmented graphs and Bregman divergence, it demonstrates that the proposed algorithm converges to the Nash equilibrium at a sub-linear rate, and the effect of communication and gradient delays on the convergence error is also clarified. Simulations in formation problems of unmanned aerial vehicle swarms are given to verify the effectiveness of the proposed algorithm.