考虑多车效应的改进耦合映射跟驰模型研究

发布时间：2018-02-01 06:25

本文关键词： 耦合映射跟驰模型优化速度多车效应稳定性拥堵抑制　出处：《重庆大学》2014年硕士论文　论文类型：学位论文

【摘要】：车辆跟驰模型(Car-Following Model)描述了单车道无超车行驶车队中后车跟随前车的动力学过程，从微观层面刻画了单车道上交通流的演化特性，对现代交通的模拟、评价及管理控制有着重要的理论价值和实际意义。耦合映射(CoupledMap)跟驰模型是车辆跟驰模型的一种离散版本，适于描述复杂的交通控制问题，且算法简洁，便于计算机仿真实现，但是传统CM模型考虑的驾驶因素较少，对实际交通的刻画比较简单。随着交通系统内车辆密度的增加，车与车之间的关系越来越紧密，当前车的行驶状态不仅受其直接前导车的影响，而且还受前方多辆车和后方车辆状态的影响，而传统的耦合映射模型没有考虑这种多车效应，为此，本文从前后多车效应出发，对CM模型进行改进，并研究了基于改进模型的交通拥堵抑制方法，为如何提供信息引导驾驶行为的改变以抑制交通拥堵提供了一定参考。论文的主要工作如下：首先，考虑多前车车头间距信息，对CM模型中的优化速度(OV)函数进行改进，提出了基于多前车效应的耦合映射(MPVCM)跟驰模型，利用线性系统理论导出了其稳定性条件，在开放边界条件下对模型进行数值模拟，，结果表明：对比传统CM模型，MPVCM模型明显改善了小扰动引起的车流失稳现象，提高了车流稳定性，并且得到考虑前方车辆的最佳数目为3辆。然后，由于实际交通中驾驶员会通过后视镜或借助ITS观察后车状态变化，为此在MPVCM模型上进一步考虑后车效应，设计相应的后视优化速度函数，提出了考虑前后多车综合效应的CECM模型，对其进行了稳定性分析与数值模拟，结果显示：相比MPVCM模型，CECM模型中的稳态车头间距增大，车速波动幅度更小，更加有效地改善了小扰动引起的车流失稳现象，进一步提高车流稳定性。最后，考虑到邻近前车与当前车速度差和安全间距是驾驶员在跟随行驶中最直接关注的信息因素，提出了基于本文CECM模型的速度差-安全间距(VDSH)综合效应拥堵抑制方法，理论推导了不产生交通拥堵的条件，给出了反馈增益取值的一般方法。理论分析和模拟结果表明：VDSH-CECM模型对比VDSH-CM模型的拥堵抑制效果更好，反馈增益取值范围更大。同时，统计不同车辆参数对应反馈增益范围发现其取值区间与驾驶员敏感系数和车辆最大速度之间存在线性相关性，且这种相关性与实际交通情况吻合。
[Abstract]:Car-following Model describes the dynamic process of the rear car following the front car in a single-lane no-overtaking vehicle fleet. The evolutionary characteristics of traffic flow in a single lane are described from the microscopic level, and the simulation of modern traffic is presented. Evaluation and management control have important theoretical and practical significance. Coupled map coupled Map) car-following model is a discrete version of car-following model. It is suitable to describe complex traffic control problems, and the algorithm is simple and convenient for computer simulation. However, the traditional CM model takes less driving factors into account, and the description of actual traffic is relatively simple. With the increase of vehicle density in traffic system, the relationship between vehicle and vehicle is more and more close. The current driving state of vehicle is not only affected by its direct leading vehicle. But the traditional coupling mapping model does not take this multi-vehicle effect into account. Therefore, this paper improves the CM model based on the multi-vehicle effect. And the traffic congestion suppression method based on the improved model is studied, which provides a certain reference for how to provide information to guide driving behavior change to curb traffic congestion. The main work of this paper is as follows: Firstly, considering the headspace information of multi-front vehicle, the optimized speed and OV) function in CM model is improved, and a coupled mapping (MPVCM) car-following model based on multi-front vehicle effect is proposed. The stability conditions are derived by using the linear system theory and the numerical simulation of the model is carried out under the open boundary condition. The results show that the pair is better than the traditional CM model. The MPVCM model obviously improves the instability of vehicle flow caused by small disturbance and improves the stability of vehicle flow. The optimal number of vehicles considered in front is 3. Then, because the driver will observe the change of the vehicle state through the rearview mirror or with the help of ITS in the actual traffic, this paper further considers the rear vehicle effect in the MPVCM model, and designs the corresponding optimized speed function of the rear view. The stability analysis and numerical simulation of the CECM model considering the comprehensive effect of front and rear vehicles are presented. The results show that the steady headway spacing in the MPVCM model is larger than that in the MPVCM model. The fluctuation of vehicle speed is smaller, which can effectively improve the instability of vehicle flow caused by small disturbance and further improve the stability of vehicle flow. Finally, considering that the speed difference and safety distance between the adjacent front car and the current vehicle are the most direct information factors that the driver pays close attention to in following the driving. Based on the CECM model of this paper, a new method of congestion suppression based on the CECM model is proposed, and the condition of no traffic congestion is derived theoretically. The theoretical analysis and simulation results show that the VDSH-CECM model is more effective than the VDSH-CM model in congestion suppression. The feedback gain range is larger. At the same time, according to the feedback gain range of different vehicle parameters, it is found that there is a linear correlation between the range of feedback gain and the driver sensitivity coefficient and the maximum vehicle speed. And this correlation is consistent with the actual traffic situation.
【学位授予单位】：重庆大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U491

【参考文献】