提出将多智能体技术集成到虚拟地理环境(VGE)中,以有效地虚拟表达面向“人”GIS。主要研究VGE中基于遗传算法(GA)和多智能体系统(MAS)的多用户之间的协作。分析和研究了基于GA的多用户集体操作行为学习的算法,建立多智能体与对象之间动态交互的动力学作用模型和运动策略;最后以进化的3个智能体如何协作完成群体推立方体箱任务为例,基于V-Realm Builder 2.0设计了VGE中3用户集体推箱的虚拟世界原型,并用Simulink 6.0进行建模与虚拟演示。
In order to express GIS oriented human virtually and effectively, the Multi-Agent technology was pat forward to integrate into virtual geographic environment (VGE). The cooperation among multi-user in VGE was researched based on Genetic Algorithm (GA) and Multi-Agent System (MAS). The arithmetic of collectivity operating behavior learning of Multi-Agent was analyzed and researched based on Genetic Algorithm (GA), and dynamics action model was established which Multi-Agents and object interacted dynamically and colony movement strategy. An example showed how 3 evolutional Agents cooperated to complete the task of colony pushing cubical box, and a virtual world prototype of virtual three users pushing box collectively was designed based on V-Realm Builder 2.0 modeling and virtual demo with Simulink 6.0 were made.