A Learning Behavioral Model of CGF

摘要

The Computer Generated Force (CGF) has decision ability by self-learning mechanism, which is an important research field in applying machine learning technology to military simulation. On the basis of modeling architecture of Agent and Learning Classifier Systems (LCSs) technologies, a learning behavioral model framework based on Genetic Algorithms (GA) of CGF is proposed. It discusses elaborately the learning process of this model, in which memory function is first introduced to accelerate. Also, a visible validation system is designed. The simulated results indicate that the learning model is available and feasible.