Space mean speed is an important index to identify traffic state on urban roads,but it is very difficult to be measured in real traffic conditions,and it is often calculated from fixed detector data.A methodology was proposed to estimate space mean speed for non uniform traffic flow.The method forecasts section based on the Calmans' filter,and a regression equation is given between speed transfer coefficient "α" and the standard deviation of velocity based on statistical analysis on inflow and outflow characteristics under different traffic state.Using the travel time collected by probe vehicles to calculate space mean speed,as the measured data,a comparison is carried out between dynamic value method and harmonic average estimation method.Results show that,dynamic value method is in good agreement with the actual traffic behavior of urban road,and has strong operability and good adaptability.%空间平均速度是反映交通状态的一个重要指标;但实际路况下的空间平均速度测量难度很大,其值一般通过定点检测交通数据估算得出.为此,提出了一种新估算方法来获取非均匀交通流的空间平均速度.该方法基于卡尔曼滤波预测路段交通流量,分析不同交通流状态下的路段车辆流入、流出特性,提出了空间速度“转化系数”α,建立了基于时间平均速度的空间平均速度估计模型.采用浮动车采集的行程时间计算路段区间平均速度,作为实测数据对动态α值法和调和平均值法的估计精度进行对比分析,结果说明,动态α值法可更好地吻合城市道路的实际交通状况,可操作性强,适应性好.
展开▼
