多智能体交通拥堵自组织控制策略研究
发布时间：2018-07-21 14:29
【摘要】：随着科技的飞速发展,社会普遍生活水平的逐渐提高,家家户户都拥有小汽车,给人们带来了很大的方便。然而在大城市里,因拥有的车辆数过多,道路供不应求,导致交通拥挤频频发生,给社会带来重大的损失。引起人们对交通拥挤的广大关注,为了更好地缓解城市的交通拥挤,就需要从根源开始探讨了。交通拥堵主要是由确定事件和随机事件引起的,对于确定事件而言如上下班高峰期,通过合适的数学建模是可以最大化的利用现有道路来缓解拥堵情况,主要受限于道路本身的负载量,只能通过提高硬件设施来优化,在本论文中不做太多讨论；对于随机事件如突发的交通事故,由于发生事故的时间、地点,都是属于未知的,从而无法按常规的方法来解决即便道路的负载量远大于通行车辆,也是不可避免的,而这种随机事件给道路交通带来严重的威胁,目前都是由交警亲自上线,指挥疏散来缓解交通拥挤的,因此一起交通事故可能导致4到8个小时的交通拥堵,甚至交通瘫痪。本论文针对城市这种突发性交通拥堵,探讨一种新型的多智能体交通系统,该系统为了减少城市的这种突发性交通性拥堵使整个城市路网的通行力滞后,实现实时监控各个路口的通行情况,在第一时间内找到发生突发性的交通拥堵的地点,也就是路口。并利用GA算法构建的智能体在线指挥信号灯打破常规的相位和周期信号控制,并协调多个智能体相互配合快速建立动态的分流路线,使上游车流量得到及时的指挥分配到合适的道路上去,避免更多的车辆融入到交通拥挤路段导致堵上加堵,也能及时阻止拥堵快速传播到整个交通网中。本论文通过利用Visual Baisc 6.0和 VISSIM4.3软件,搭建了虚拟的智能交通系统仿真平台,验证了该系统是可以有效的提高道路通行率的。
[Abstract]:With the rapid development of science and technology and the gradual improvement of the standard of living in the society, every household owns cars and brings great convenience to people. However, in the big cities, the number of vehicles and the shortage of roads lead to the heavy traffic congestion and the great loss to the society. In order to better alleviate urban traffic congestion, it is necessary to start from the origin of the traffic congestion. Traffic congestion is mainly caused by the determination of events and random events. For the determination of events, such as rush hour at work, through appropriate mathematical modeling, it is possible to maximize the use of existing roads to alleviate congestion, mainly limited to The load of the road itself can only be optimized by improving the hardware facilities, not too much discussion in this paper. For random events such as sudden traffic accidents, the time and location of the accident are unknown, which can not be solved according to the conventional way that the load of the road road is far greater than the passing vehicle. This random event poses a serious threat to road traffic. At present, the traffic traffic traffic traffic congestion can be caused by traffic traffic congestion and even traffic paralysis. This paper deals with a new type of traffic congestion in the city. In order to reduce the traffic congestion caused by the sudden traffic congestion in the city, the system can reduce the traffic capacity of the whole urban road network, and realize the real time monitoring of the traffic situation of each intersection. In the first time, it finds the place where the sudden traffic congestion occurs, that is, the intersection. And the GA algorithm is used to construct the online command letter of the agent. The lamp breaks the conventional phase and periodic signal control, and coordinates multiple agents to cooperate with each other quickly to establish a dynamic distributary route, so that the flow of the upstream vehicle can be assigned to the appropriate road in a timely manner, avoiding more vehicles being integrated into the traffic congestion section and blocking the traffic jam, and can also prevent the congestion quickly spread to the whole. In this paper, a virtual intelligent transportation system simulation platform is built by using Visual Baisc 6 and VISSIM4.3 software, which proves that the system can effectively improve the road traffic rate.
【学位授予单位】：长沙理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U495
本文编号：2135826
[Abstract]:With the rapid development of science and technology and the gradual improvement of the standard of living in the society, every household owns cars and brings great convenience to people. However, in the big cities, the number of vehicles and the shortage of roads lead to the heavy traffic congestion and the great loss to the society. In order to better alleviate urban traffic congestion, it is necessary to start from the origin of the traffic congestion. Traffic congestion is mainly caused by the determination of events and random events. For the determination of events, such as rush hour at work, through appropriate mathematical modeling, it is possible to maximize the use of existing roads to alleviate congestion, mainly limited to The load of the road itself can only be optimized by improving the hardware facilities, not too much discussion in this paper. For random events such as sudden traffic accidents, the time and location of the accident are unknown, which can not be solved according to the conventional way that the load of the road road is far greater than the passing vehicle. This random event poses a serious threat to road traffic. At present, the traffic traffic traffic traffic congestion can be caused by traffic traffic congestion and even traffic paralysis. This paper deals with a new type of traffic congestion in the city. In order to reduce the traffic congestion caused by the sudden traffic congestion in the city, the system can reduce the traffic capacity of the whole urban road network, and realize the real time monitoring of the traffic situation of each intersection. In the first time, it finds the place where the sudden traffic congestion occurs, that is, the intersection. And the GA algorithm is used to construct the online command letter of the agent. The lamp breaks the conventional phase and periodic signal control, and coordinates multiple agents to cooperate with each other quickly to establish a dynamic distributary route, so that the flow of the upstream vehicle can be assigned to the appropriate road in a timely manner, avoiding more vehicles being integrated into the traffic congestion section and blocking the traffic jam, and can also prevent the congestion quickly spread to the whole. In this paper, a virtual intelligent transportation system simulation platform is built by using Visual Baisc 6 and VISSIM4.3 software, which proves that the system can effectively improve the road traffic rate.
【学位授予单位】：长沙理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U495
【参考文献】
相关期刊论文 前2条
1 赵建有,赵丽平;基于多智能体的城市交通流控制原型系统[J];交通运输工程学报;2003年03期
2 杨帆;云美萍;杨晓光;;车路协同系统下多智能体微观交通流模型[J];同济大学学报(自然科学版);2012年08期
，本文编号：2135826
本文链接：https://www.wllwen.com/kejilunwen/jiaotonggongchenglunwen/2135826.html
