“当前”模型自适应卡尔曼滤波在有轨电车定位中的应用

The Application of Self-adaption Kalman Filtering Algorithm Based on the Current Statistic Model in the Tram’s GPS Positioning

有轨电车常与地面其他交通工具混行,因此其准确定位是列车调度指挥和运输安全的重要保障。为节约成本,有轨电车系统一般采用GPS结合速度传感器等方式进行定位,而非大量铺设应答器,因此,GPS信号的处理对于列车定位精度的提高具有重要作用。从简化系统、节省费用、计算难度和精度等方面综合考虑,采用卡尔曼滤波方法对GPS定位信息进行修正。同时,为更好地描述列车运行状态,使其更贴近列车真实运动情况,采用机动加速度"当前"统计模型结合自适应卡尔曼滤波方法对有轨电车位置进行最优估计处理。通过Matlab仿真对在试验场地采集的GPS定位信息进行最优估计处理。仿真结果表明:采用"当前"模型下的自适应卡尔曼滤波算法后,列车定位结果能够更好地跟随其运动的真实轨迹,更接近其真实值,很好地提高了定位精度。

Trams often mix with other vehicles on the ground, so their accurate positioning is important for train dispatching and transportation safety. The method of combining speed sensors and GPS is usually adopted by the streetcars system in positioning area. In order to simplify the system, save cost, calculate difficulty and accuracy, Kalman filter method was used to correct the GPS positioning information. Meanwhile, the current statistic model of maneuvering acceleration was adopted to describe the tram state, which was closed to the real movement of the tram. Through Matlab simulation, the GPS positioning message collected in the test site dealt and estimated by the Kalman filtering based on the current model. The result shows that the positioning result can follow the real movement of the tram better. In other words, the Kalman filtering based on the current model can improve the precision of positioning and the result may closer to the real value.