基于智驾盒子数据的地图匹配方法及应用研究

发布时间：2018-04-05 01:30

本文选题：智驾盒子　切入点：地图匹配　出处：《北京交通大学》2017年硕士论文

【摘要】：随着车联网技术蓬勃发展,产生了记录车辆实时位置和行驶状态信息的"盒子"数据。通过对"盒子"数据的统计与挖掘,可以分析车辆健康状况与驾驶人行为特征,实现交通状态的感知与监测。而定位信息的地图匹配过程,是充分发挥数据价值的基础。基于此本文首先通过对电子地图进行格网划分,实现定位数据的快速初匹配过程,建立了包含全部定位点及其所有可能匹配结果的初匹配结果集。其次基于马尔可夫模型,提出了一种改进的权重最优匹配算法,将车辆航向和相邻定位点间实际行驶距离作为影响因子引入模型中。通过车辆航向与候选匹配点所在路段走向间的偏差角度、车辆定位位置与候选匹配点位置间的偏差距离来确定候选匹配点的观测概率;用两相邻定位点间车辆实际行驶距离与其对应的两候选匹配点在电子地图路网中最短路径长度的比值确定转移概率。用维特比算法实现由初匹配结果集到最优匹配路径的求解过程,并利用电子地图拓扑关系对匹配路径进行了链路修正,保证了匹配结果序列中相邻路段之间的可达性和匹配路径的合理性。最后利用匹配结果,进行了基于行程时间的路段平均速度测算,计算不同环路区间快速路、主干路和次干路均速的日变化情况。并将计算结果与出租车浮动车监测结果进行了对比,分析不同区域、不同等级道路上私家车与出租车速度的异同。分析结果表明,私家车与出租车路网速度及变化趋势基本一致,证明本文所提出的算法合理可靠;由于车况等因素影响,同等条件下私家车的平均速度略高于出租车平均速度。
[Abstract]:With the rapid development of vehicle networking technology, the "box" data is generated to record the real-time position and state information of the vehicle.Through the statistics and mining of the "box" data, we can analyze the health status of the vehicle and the characteristics of the driver's behavior, and realize the perception and monitoring of the traffic state.The map matching process of location information is the basis of giving full play to the value of data.In this paper, first of all, by dividing the grid of the electronic map, the fast initial matching process of the location data is realized, and the initial matching result set including all the positioning points and all possible matching results is established.Secondly, based on Markov model, an improved weight optimal matching algorithm is proposed, which introduces the vehicle heading and the actual driving distance between the adjacent positioning points as the influence factors in the model.The observation probability of the candidate matching point is determined by the deviation angle between the vehicle heading and the track of the candidate matching point and the deviation distance between the vehicle location position and the candidate matching point position.The transfer probability is determined by the ratio of the actual vehicle travel distance between the two adjacent positioning points and the corresponding two candidate matching points in the electronic map network.Viterbi algorithm is used to solve the problem from the initial matching result set to the optimal matching path, and the link correction of the matching path is carried out by using the topological relation of the electronic map.The reachability between adjacent sections and the rationality of matching path are ensured in the sequence of matching results.Finally, using the matching result, the average speed of the section based on the travel time is calculated, and the diurnal variation of the average speed of the expressway, the main road and the secondary trunk road in different loop sections is calculated.The results are compared with the results of taxi floating vehicle monitoring, and the similarities and differences of the speed of private cars and taxis in different regions and different grades of roads are analyzed.The results show that the speed and variation trend of private car and taxi network are basically the same, which proves that the algorithm proposed in this paper is reasonable and reliable, and the average speed of private car is slightly higher than that of taxi under the same conditions.
【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U495;U463.6

【相似文献】