殷吉崇, 武芳, 李安平, 杜佳威, 刘呈熠. 顾及速度与航向信息的轨迹匹配方法[J]. 武汉大学学报 ( 信息科学版), 2021, 46(11): 1774-1781. DOI: 10.13203/j.whugis20210058
引用本文: 殷吉崇, 武芳, 李安平, 杜佳威, 刘呈熠. 顾及速度与航向信息的轨迹匹配方法[J]. 武汉大学学报 ( 信息科学版), 2021, 46(11): 1774-1781. DOI: 10.13203/j.whugis20210058
YIN Jichong, WU Fang, LI Anping, DU Jiawei, LIU Chengyi. Trajectory Matching Considering Speed and Heading Information[J]. Geomatics and Information Science of Wuhan University, 2021, 46(11): 1774-1781. DOI: 10.13203/j.whugis20210058
Citation: YIN Jichong, WU Fang, LI Anping, DU Jiawei, LIU Chengyi. Trajectory Matching Considering Speed and Heading Information[J]. Geomatics and Information Science of Wuhan University, 2021, 46(11): 1774-1781. DOI: 10.13203/j.whugis20210058

顾及速度与航向信息的轨迹匹配方法

Trajectory Matching Considering Speed and Heading Information

  • 摘要: 现有轨迹匹配方法具有一定局限性,其匹配精度无法满足车辆导航定位的需求。针对拓扑匹配方法对于起始匹配位置的依赖性,提出了顾及速度与航向信息的轨迹匹配方法。该方法综合利用速度、距离和航向约束改进起始匹配路段和起始位置的判定,并通过后续时刻车辆的正确匹配位置修正起始位置,避免了传统拓扑匹配方法中起始位置匹配错误导致的误差传递累积,提高匹配路段的准确性。同时以起始位置为基础,速度与时间信息为约束确定匹配点。为验证所提方法的有效性,选取多条复杂程度各异的路线进行实验,并与现有的几何匹配方法和拓扑匹配方法进行比较。实验结果表明,该方法在不同复杂程度的城市路网下具有良好的匹配效果,准确率可达93.53%。在匹配准确率方面优于现有的两种方法,在匹配效率方面能满足定位导航的需求。

     

    Abstract:
      Objectives  Trajectory matching is the process of matching location data to the corresponding position in the road network. It is a necessary processing step for many related applications based on global navigation satellite system (GNSS) trajectory data analysis and position analysis.
      Methods  We propose a trajectory matching method considering speed and heading information to solve the problem that existing trajectory matching methods have low accuracy and efficiency for complex road network data. Firstly, we consider constraint criteria such as distance, heading, speed, topology and mileage to optimize the selection process of matching sections. Secondly, considering the continuity of the trajectory in the construction of the candidate road set, a rectangular buffer is added to avoid missing matches. Moreover, the proposed method improves the judgment criterion of initial position and uses the subsequent matching position to correct the judgment of initial position, and to avoid the accumulation of error transmission caused by the matching error of initial position in traditional topology matching method, and improves the matching accuracy. Finally, the speed and time information are used to determine the matching position in the matching section and to improve the matching efficiency. To evaluate the performance of our proposed method, a number of real route data sets with different complexity are used to compare the proposed method with the geometric matching method and topological matching method.
      Results  Experimental results show that the proposed method has a good matching effect for urban road networks with different levels of complexity. It can match the correct road for special road sections such as overpasses, intersections, and parallel sections. And the accuracy of the method can reach 93.53%, which outperforms the existing two methods.
      Conclusions  Our proposed method can meet the needs of positioning and navigation in terms of matching efficiency.

     

/

返回文章
返回