基于多智能体的CGF建模研究

Research on the CGF Modeling Based on Multi-Agent

为了解决某分布式虚拟训练系统中计算机生成兵力建模问题,对计算机生成兵力建模需求进行了全面分析。在此基础上,研究了基于多智能体的CGF模型。设计了由任务规划层和路径规划层组成的混合型智能体。根据智能体任务规划特点,提出了基于ANFIS的智能体任务务规划算法。通过仿真实验对智能体模型和算法的正确性进行验证,结果表明提出的智能体模型满足了建模需求,任务规划算法具有较好的精度。

The CGF model requirement is analysized comprehensively to solve the issue of CGF modeling in the development of some distributed virtual training system. On the basis of the requirement analysis, the CGF model based on multi-agent is studied. The agent is designed, which is composed of task planning layer and route planning layer. According to the characteristic of the task planning of the agent, the task planning algorithm based on ANFIS is proposed. Simulation experiments are conducted to validate the correctness of the model and the algorithm, and the resluts show that the proposed agent model meets the model requirements and the task planning algorithm is accurate.