Figure 5 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
The segments in the OSM map. The red lines represent the roads of OSM, the blue points are the start or end points of segments, and ei means the segment of road.
Source publication
The data collected by floating cars is an important source for lane-level map production. Compared with other data sources, this method is a low-cost but challenging way to generate high-accuracy maps. In this paper, we propose a data correction algorithm for low-frequency floating car data. First, we preprocess the trajectory data by an adaptive d...
Context in source publication
Citations
... The methods of clustering include (1) clustering based on the density of GPS points (Biagioni and Eriksson 2012;Li et al. 2018) and (2) clustering based on the direction and distance features of GPS traces (Tang et al. 2016;Deng et al. 2018;Li et al. 2012;Liu et al. 2012). The kernel density method is the most commonly used way to build the probability function of similar GPS points for clustering (Biagioni and Eriksson 2012). ...
Vehicles have been increasingly equipped with GPS receivers to record their trajectories, which we call floating car data. Compared with other data sources, these data are characterized by low cost, wide coverage, and rapid updating. The data have become an important source for road network extraction. In this paper, we propose a novel approach for mining road networks from floating car data. First, a Gaussian model is used to transform the data into bitmap, and the Otsu algorithm is utilized to detect road intersections. Then, a clothoid-based method is used to resample the GPS points to improve the clustering accuracy, and the data are clustered based on a distance-direction algorithm. Last, road centerlines are extracted with a weighted least squares algorithm. We report on experiments that were conducted on floating car data from Wuhan, China. To conclude, existing methods are compared with our method to prove that the proposed method is practical and effective.
... Current research in the field of map matching merges from two complementary main views of the problem. One view is about optimizing processing speed [e.g., 49,46,70,5,85,79,24,26], the other one is about optimizing accuracy of the output of map matching algorithms [e.g., 33,35,57,53,30,74,81,82,7,48,40,15,31]. More accurate results are usually enabled by using technologies with a higher time complexity. ...
... When calculating the probabilities or rewards existing data from the recorded track and given road network can be used. There are for example distances and routes [53], bearings [58], trajectories [40], and velocity and timings [46] that can be computed. Typically, routes in the road network are calculated with Dijkstra's shortest path algorithm [19]. ...
Map matching is about finding the best route of a given track in a road network. This can be useful for many statistical analyses on mobility. With increasing spread of modern cars and mobile devices many tracks are available to match. The difficulty in map matching lies in the geospatial differences between tracks and road networks. Current technologies resolve such differences with Hidden Markov Models and Viterbi algorithm. They majorly vary concerning the used metrics for the probabilities in the models. In this research we improve map matching technology by refining the underlying algorithms, models and metrics. We will introduce Markov Decision Processes with Value Iteration and Q-Learning to the map matching domain and we will compare them to Hidden Markov Models with Viterbi algorithm. Markov Decision Processes allow to use active decisions and rewards, which are not available in previous methods. Combined with improvements concerning the preparation of tracks and the road network, and various technologies for improved processing speed, we will show on a publicly available map matching data set that our approach has a higher overall performance compared to previous map matching technology. We will eventually discuss more possibilities we enable with our approach.
... Based on the vehicle trajectory with a low-sampling rate, Yuan et al. [17] considered the road distance, direction, speed and topology, proposing a global map-matching algorithm. Li et al. [18] proposed a calibration algorithm based on low-frequency sampling data, and used adaptive density optimisation to pre-process the trajectory data; Chen et al. [5] proposed a multi-criteria dynamic programming (MDP) mapping matching algorithm, which uses MDP technology to ensure that a suitable matching path is determined while minimising the number of candidate routes for each GPS point. The aforementioned algorithm can improve the matching accuracy of the algorithm at a low-sampling frequency. ...
Global positioning system (GPS) trajectory map matching projects GPS coordinates to the road network. Most existing algorithms focus on the geometric and topological relationships of the road network, while did not make full use of the historical road network information and floating car data. In this study, the authors proposed a deep learning enabled vehicle trajectory map‐matching method with advanced spatial–temporal analysis (DST‐MM). The algorithm mainly focused on the following three aspects: (i) analyse the spatial relevancy from the prospective of geometric analysis, topology analysis and intersection analysis; (ii) to make full use of the historical and real‐time data, a deep learning model was conducted to extract the road network and vehicle trajectory features and (iii) establish a speed prediction model and nest it in the temporal analysis structure. It narrows down the path search range through establishing the dynamic candidate graph. Experimental results show that the proposed DST‐MM algorithm outperforms the existing algorithms in terms of matching accuracy for low‐sampling frequencies GPS data, especially in the central urban area.
... Astro-inertial systems are considered the most accurate among other navigation systems [8,9], while in another aspect it is recognized that these astro-systems have a lower noise immunity. During the operation time of autonomous INS with sufficiently long intervals, errors can reach unacceptably large values [10]. It is necessary to compensate for the errors of an autonomous INS applying the internal connections of the system. ...
This paper presents new algorithmic methods for accuracy improvement of autonomous inertial navigation systems of aircrafts. Firstly, an inertial navigation system platform and its nonlinear error model are considered, and the correction schemes are presented for autonomous inertial navigation systems using internal information. Next, a correction algorithm is proposed based on signals from precession angle sensors. A vector of reduced measurements for the estimation algorithm is formulated using the information about the angles of precession. Finally, the accuracy of the developed correction algorithms for autonomous inertial navigation systems of aircrafts is studied. Numerical solutions for the correction algorithm are presented by the adaptive Kalman filter for the measurement data from the sensors. Real data of navigation system Ts-060K are obtained in laboratory experiments, which validates the proposed algorithms.
The GPS map matching technique matches a series of geographic coordinates to a road network. However, most existing algorithms mainly consider the geometric and topological relationships of the GPS observations, disregarding information in the historical floating data. This study developed an enhanced spatial-temporal matching (EST-matching) algorithm effective with low frequency GPS data. The proposed algorithm uses the possible minimum travel time to filter out the unrealistic route and build a graph of the candidate paths. (2) It considers the initial matching probability between candidates and the historical edge data to identify the actual vehicle route. (3) For vehicles at intersections, we introduce direction analysis to increase algorithm accuracy. The EST-matching algorithm was tested against the stMM algorithm to verify its performance at various data collection frequencies and matching radius. The proposed algorithm outperforms the stMM algorithm in terms of matching accuracy based low sampling frequencies, especially in central urban areas.
