城市交通中车队状态观测器的一种新算法

本文阐述了离散时间点过程理论,时变马尔科夫链及鞅差分序列在城市交通车队状态观测器中的应用。并在此基础上,改进了[5]中的估计算法。用本文提出的非线性最小方差估计算法,对提供的交通状态进行估计,所得结果比[5]中算法精度有明显提高。在大连市某交通干线计算机控制系统中初步应用,取得了令人满意的结果。

This paper presents some applications of discrete - time point process theory, time - varying Markov chain and Martingle difference in urban traffic computer-based control systems. On this basis We established a new queue model. The new algorithm we proposed to predicte the state of queue in the model was tested in comparison with the previous ones in [5] by simulation. Results of the test are shown indicating that an apparant better performance then those of previous ones in [5] is achived. The algorithm was applied in the computer-based urban traffic control system for a main street in Dalian City, and a satisfactory performance was obtained.