Many-neighbour interaction and non-locality in traffic models
Many-neighbour interaction and non-locality in traffic models
复制标题
DOI：
10.1140/epjb/e2004-00205-y
复制
发表时间：
2004-06
期刊：
影响因子：
--
通讯作者：
R. E. Wilson;Peter Berg;S. Hooper;G. Lunt
中科院分区：
文献类型：
--
作者：
R. E. Wilson;Peter Berg;S. Hooper;G. Lunt
Bando等人提出的最佳速度模型。[1]，显示了各种最优速度函数的不切实际的加速度值[2，3]。我们讨论了如何纠正这个问题的不同方法。在将加速度和减速度降低到现实值方面，多个前瞻（多邻居相互作用）模型是最有希望的候选模型。我们专注于两个这样的模型，特别是，他们的线性稳定性，以及这些如何影响车辆动力学和波的解决方案。如前所述[4]，多个前瞻模型再现了许多真实的流动特征，我们的结果进一步支持了这种分析的必要性。然而，当它们转化为相应的连续模型时，非定域性的问题就出现了。本文讨论了在宏观模型中如何解释多近邻相互作用的三种方法。
The optimal-velocity model, as proposed by Bando et al. [1], shows unrealistic values of the acceleration for various optimal-velocity functions [2,3]. We discuss different approaches of how to correct this problem. Multiple look-ahead (many-neighbour interaction) models are the most promising candidates in reducing accelerations and decelerations to realistic values. We focus on two such models and, in particular, their linear stability and how these affect the vehicle dynamics and wave solutions. As found earlier [4], multiple look-ahead models reproduce many real flow features, and our results further support the necessity of this ansatz. However, the problem of non-locality arises when they are transformed into the corresponding continuum model. We discuss three methods of how to interpret many-neighbour interaction in macroscopic models.