基于ATIS下的OD矩阵估计  

Estimation of OD Matrix Based on Information from ATIS

作　　者：杨小宝[1] 王新伟[2] 张宁[2] 

机构地区：[1]北京交通大学交通运输学院,北京100044 [2]北京航空航天大学经济管理学院,北京100083

出　　处：《公路交通科技》2007年第9期100-103,108,共5页Journal of Highway and Transportation Research and Development

基　　金：国家自然科学基金资助项目(70521001);教育部新世纪优秀人才资助项目(NCET-04-0173);国家973计划资助项目(2006CB705503)

摘　　要：利用ATIS收集的信息,通过广义最小二乘模型,建立了一种新的OD矩阵估计方法。新方法利用ATIS能得到装有GPS车辆的路段选择比例矩阵,据此可推出网络中全部车辆的路段选择比例矩阵的期望值,克服了传统方法采集数据难、数据精度低等缺陷。新方法能获得实时可靠的数据,能对OD矩阵进行实时更新,从而增强了数据的实效性。在对新方法解的有关性质进行分析的基础上,给出了相应的求解算法,得出新方法的目标函数局部最优解是全局最优解,目标函数的解不唯一,但OD估计量唯一。数值仿真表明新算法非常有效,与传统方法相比,新方法是一种精度高、可靠性强的估计方法。Based on Advanced Transportation Information System （ATIS） and generalized least square, a new method of origin- destination matrix estimation is proposed, by which the link choice proportion matrix for vehicle equipped with GPS can be obtained by using ATIS, and the means of link choice proportion matrix for all vehicles in the network can be deducted. Moreover, the proposed method can obtain the real-time and reliable data of OD matrix in short time and update the OD matrix simultaneously due to the sub-OD matrix replacing the prior OD. In addition, the quality of solutions and algorithm in the model are discussed. The partial optimum solution of the objective function in the proposed estimation model is the global optimum solution, and the OD estimation obtained by the model is unique. Its implementation is demonstrated by a numerical example. The results show that the proposed algorithm is effective and the new method can improve precision of OD matrix estimation.

关 键 词：智能运输系统 OD矩阵 广义最小二乘法 估计 

分 类 号：U491[交通运输工程—交通运输规划与管理]