基于数据弥补的自稳控制交通流建模与仿真

Modeling and Simulation of Self-Stabilizing Control Traffic Flow Based on Data Compensation

摘要: 传统车辆交通流模型根据车头距、当前速度及前车速度等数据建立，近年来，有些学者在传统交通流模型上加入了当前速度和历史速度的差，以提升交通流对小扰动传播的稳定性，并称之为自稳控制。但是这种自稳控制所需的信息容易因为一些因素丢失，导致交通流稳定性下降。本文试图在自稳数据丢失的情况下做数据的弥补，通过使用最近邻前车的相同类型的数据对丢失的自稳控制所需数据进行替换，并提出基于数据弥补的自稳控制交通流模型。为了验证对丢失数据的补偿是否能使自稳车辆稳定，对补偿方案进行了建模以及线性稳定性分析。理论分析结果表明，本文提出的补偿策略使交通流更稳定，仿真结果验证了理论分析结果的正确性。

Abstract: Traditional traffic flow models for vehicles are built from data on the headway, current speed, and speed of the preceding vehicle. In recent years, some scholars have added the difference between the current and historical velocity to traditional traffic flow models to improve the stability of traffic flow to small perturbation propagation, which is called self-stabilizing control. However, the information needed for self-stabilizing control is easily lost due to various factors, which leads to a decrease in traffic flow stability. This paper attempts to compensate the data in the case of the loss of the stability data, by using the same categorical data of the nearest preceding vehicle to replace the lost data needed for the stability control, and propose a traffic flow model based on data compensation. To verify whether compensation for missing data can stabilize a self-stabilizing vehicle, the compensation scheme is modeled and a linear stability analysis is performed. The theoretical analysis results show that the proposed compensation strategy makes traffic flow more stable, and the simulation results verify the correctness of the theoretical analysis results.