引用本文: | 王昱,章卫国,傅莉,黄得刚,李勇.基于精英改选机制的粒子群算法的空战纳什均衡策略逼近[J].控制理论与应用,2015,32(7):857~865.[点击复制] |
WANG Yu,ZHANG Wei-guo,FU Li,HUANG De-gang,LI Yong.Nash equilibrium strategies approach for aerial combat based on elite re-election particle swarm optimization[J].Control Theory and Technology,2015,32(7):857~865.[点击复制] |
|
基于精英改选机制的粒子群算法的空战纳什均衡策略逼近 |
Nash equilibrium strategies approach for aerial combat based on elite re-election particle swarm optimization |
摘要点击 3772 全文点击 1568 投稿时间:2015-01-05 修订日期:2015-03-26 |
查看全文 查看/发表评论 下载PDF阅读器 |
DOI编号 10.7641/CTA.2015.50008 |
2015,32(7):857-865 |
中文关键词 纳什均衡 动态目标分配 优化 精英改选粒子群 多样性 混合策略 |
英文关键词 Nash equilibrium dynamic target assignment optimization ERPSO diversity mixed strategies |
基金项目 国家自然科学基金项目(61374032), 辽宁省 教育庁科学研究一般项目(L2015412), 沈阳市科技创新团队项目(src201204)资助. |
|
中文摘要 |
针对无人机协同攻击的动态多策略性, 应用纳什均衡概念, 考虑联合生存概率 和武器消耗等因素, 融合双方价值函数计算和双矩阵对策纳什均衡点的求解方法, 建立一种多战斗步空战动态目标分配优化模型. 提出基于精英改选机制的粒子群(elite re-election particle swarm optimization, ERPSO)算法, 在群体极值引导能力不足时, 通过对其克隆、变异和重新初始化等操作增加个体的多样性, 保留传统粒子群算法结构简单、快速收敛等优点, 改善算法易于陷入局部 极小的问题. 将ERPSO算法应用于目标分配模型求解纳什均衡点, 获取更为精确的双方 混合策略, 确保实时性和准确性, 验证了模型和方法的有效性. |
英文摘要 |
A multi-combat step dynamic target assignment optimization model is built based on Nash equilibrium concept by considering the joint survival probability and weapons consumption factors for dynamic multi-strategy UAV cooperative attack. In building the model the value function calculation for the two belligerent parties and the solving method of bimatrix game Nash equilibrium point are applied. Then, an elite re-election particle swarm optimization (ERPSO) is proposed based on the elite reelection mechanism; therefore, the diversity of individuals can be increased when leading capacity of group extreme value is deficient through cloning, mutation and re-initialization operation. The advantages of traditional particle swarm optimization (PSO) algorithm such as simple structure, fast convergence are retained, but the shortcoming of falling into local minimum is corrected. Finally, the ERPSO algorithm is applied to the dynamic target assignment model to solve the Nash equilibrium point for obtaining accurate hybrid strategies for both sides. It is confirmed that the real-time demands and accuracy are satisfied, and the validity of the proposed model and method is established. |
|
|
|
|
|