基于无线探针数据的区间车速估计方法与分析

Interval Vehicle Speed Estimation Method and Analysis Based on Wireless Probe Data

针对无线探针技术在交通数据检测中的应用问题,基于无线探针中Wi-Fi和蓝牙各自的技术特点与数据特征,制定基于集合Kalman滤波的区间车速估计方法.以卡口数据为基准,评估无线探针数据在区间车速估计中的精度,分析估计误差与实际速度的相关性及Wi-Fi与蓝牙数据在区间车速估计中的异同.研究发现低速时无线探针估计车速偏高,高速时则相反.定义误差为零的速度为临界速度,Wi-Fi和蓝牙的临界速度分别为38和51 km/h.采用本文提出的区间车速估计方法,可实现63.2%的时间窗误差下降,总下降幅度为18.1%.

Aiming at the application problem of wireless probe technology in traffic data detection,based on the respective technical characteristics and data characteristics of Wi-Fi and Bluetooth in wireless probe,an interval vehicle speed estimation method based on ensemble Kalman filter was developed.Based on bayonet data,the accuracy of wireless probe data in interval speed estimation was evaluated,and the correlation between estimation error and actual speed and the similarities and differences between Wi-Fi and Bluetooth data in interval speed estimation were analyzed.The research found that the wireless probe estimated the vehicle speed to be too high at low speed,but it was opposite at high speed.If the speed with zero error was defined as the critical speed,the critical speeds of Wi-Fi and Bluetooth were 38 and 51 km/h respectively.Using the interval speed estimation method proposed in this paper,the time window error could be reduced by 63.2%,with a total reduction of 18.1%.