基于智能体技术的CGF行为模拟研究  

Study on Agent-based CGF Behaviors Simulation

作　　者：孙少斌[1] 陈璐[1] 张仁友[1] 

机构地区：[1]蚌埠坦克学院,安徽蚌埠233050

出　　处：《火力与指挥控制》2009年第3期79-83,共5页Fire Control & Command Control

基　　金：获军队科技进步二等奖

摘　　要：为了有效模拟CGF物理行为和智能行为的不同特性,提出了一个复合CGF智能体模型,将CGF的行为模拟分为认知、内部环境、规划决策和行动四个模块,该模型综合了反应式行为模式和基于规划模式两方面优点。利用行为模式采用层次化方法建立CGF物理行为模型,使CGF智能体能对动态战场环境做出灵活的反应。结合CGF不同类型任务规划决策方法的分析介绍了基于规划模式的CGF智能行为模拟方法,使CGF智能体能按照战术原则进行作战任务规划。采用基于军事命令控制结构对CGF进行组织和控制,为CGF间的指挥协同提供了与实际作战环境相似的框架。In order to effectively simulate the physical behaviors and intelligent behaviors of CGF agents, a composite model of CGF agents is introduced, which divides CGF behavior simulation into four modules： perception, inner environment, mission planning, and action. The model combines the use of patterns of reactive behavior with plan-based patterns. A hierarchical modeling method is adopted to model CGF physical behavior by using behavior pattern. The patterns of reactive behavior allow CGF agents to react flexibly to changes in its battlefield environment. Based on the introduction of different missions planning, CGF intelligent behaviors simulation by using plan-based patterns is discussed. The plan-based patterns provide CGF with the ability of devising plans deliberately according to its tactical tenets. The hierarchical framework of CGF organizing and controlling based on military command structures provides CGF with a mechanism to command and coordinate in a manner similar to that found in real battlefield.

关 键 词：CGF 物理行为 智能行为 行为模式 任务规划 

分 类 号：TP31[自动化与计算机技术—计算机软件与理论]