基于Multi-Agent的空战作战行为协同方法  

Coordination method of air combat behavior based on Multi-Agent

作　　者：郑俊[1] 张安[1] 王云辉[1] 

机构地区：[1]西北工业大学电子信息学院,陕西西安710072

出　　处：《系统工程与电子技术》2009年第11期2663-2667,共5页Systems Engineering and Electronics

基　　金：陕西省自然科学基金(2006F45);高等学校博士学科点专项科研基金(00060699026)资助课题

摘　　要：针对空战中多种作战行为的最优协同使用,构建了Multi-Agent模型,该模型由网络接口层、中央协调层和作战行动层三层组成。在中央协调层中,提出了中央协调算法以实现作战行为的协调使用。该算法通过引入影响因子建立综合作战效能模型,反映了多种软硬杀伤武器、飞行控制系统和传感器系统协同使用时各作战行为的交互影响作用。综合上层全局作战意图和底层行动单元获取的信息,得到所有协同方案的预估综合作战效能。通过以综合作战效能预估值最大为准则,动态寻优所有组合空间的方法,可获得最佳协同方案。通过仿真得到在动态战场态势下的作战行为组合为最佳协同方案,表明了该协同方法的正确性。In order to achieve optimal coordinated use of multi combat behavior in air combat, a Multi- Agent model is constructed, which includes network interface layer, central cooperation layer and combat action layer. In the central cooperation layer, a central cooperation algorithm is put forward to realize the coordinated use of combat behavior. Firstly, by introducing the concept of an impact factor in such an algorithm,an integra tive combat effectiveness model is established while taking into account the mutual influence of combat behavior, which are implemented by various kinds of hard kill and soft kill weapons, flight control system or sensor sys tem. Then the global combat intention from upper layer and the information from bottom action layer are inte- grated to get the integrative combat effectiveness. Finally, the algorithm dynamically optimizes the search space of all coordination schemes to find out the optimal one, using maximal estimated integrative effectiveness as the cooperation criteria. The simulation results show that the combat behavior schemes, which are obtained in dy- namic battlefield situations based on the proposed model, are the optimal coordination schemes, and the feasibil ity of the model is validated consequently.

关 键 词：系统工程 协同优化 MULTI-AGENT 作战行为 效能 决策 

分 类 号：N945.25[自然科学总论—系统科学]