复杂适应性系统的仿真与研究--基于CAS理论的交通模拟 |
| |
引用本文: | 顾珊珊,陈禹.复杂适应性系统的仿真与研究--基于CAS理论的交通模拟[J].复杂系统与复杂性科学,2004,1(1):82-88. |
| |
作者姓名: | 顾珊珊 陈禹 |
| |
作者单位: | 中国人民大学经济科学实验室,北京,100872 |
| |
摘 要: | 为更好地研究交通系统,以CAS理论为指导思想,利用SWARM平台建立了基于多主体离散的动态交通模拟系统。使用自底向上的建模方法进行模拟:将车辆及信号灯作为具有适应性的主体.利用元胞自动机模拟动态交通流;同时将激励学习方法与遗传算法相结合,对信号灯周期进行自适应优化。通过主体不断的“学习”,交通系统在宏观方面涌现出一定的动态特征和规律。进而,我们比较了完全自组织控制模式以及加入预警机制两种情况下各宏观量的变化情况。从实验结果可知,预警机制可以给车辆宏观方面的指导,从而提高车辆通行能力,减少交通拥塞。
|
关 键 词: | CAS理论 复杂适应性系统 交通模拟系统 元胞自动机 遗传算法 激励学习 |
文章编号: | 1672-3813(2004)01-0082-07 |
修稿时间: | 2003年11月20 |
Modeling & Simulation of Complex Adaptive System--A Traffic Simulation System |
| |
Authors: | GU Shan-shan CHEN Yu |
| |
Abstract: | Traffic system is a complex adaptive system (CAS). Intuited by CAS theory, we present a decentralized multi-agent based dynamic traffic simulation system with SWARM platform. We develop the traffic systems from bottom up, where vehicles and traffic signals are defined as agents. The interaction of these individual agents are simulated on a road map, thus leading to emergent patterns of traffic dynamics. Cellular Automaton is used to simulate the traffic flow and an evolutionary design is applied to self-organized control of the traffic signals. Furthermore, we compare the control approach if we join the forecast mechanism with the pure self-organized control method. From the experimental results, it can be concluded that the performance of the former approach is much superior to the latter one in that it can reduce the traffic jam. |
| |
Keywords: | CAS theory cellular automata genetic algorithm reinforcement learning traffic flow |
本文献已被 CNKI 维普 万方数据 等数据库收录! |