ISPRS International Journal of Geo-Information | Vol.6, Issue.12 | | Pages
An Automatic Road Network Construction Method Using Massive GPS Trajectory Data
Automatically acquiring comprehensive, accurate, and real-time mapping information and translating this information into digital maps are challenging problems. Traditional methods are time consuming and costly because they require expensive field surveying and labor-intensive post-processing. Recently, the ubiquitous use of positioning technology in vehicles and other devices has produced massive amounts of trajectory data, which provide new opportunities for digital map production and updating. This paper presents an automatic method for producing road networks from raw vehicle global positioning system (GPS) trajectory data. First, raw GPS positioning data are processed to remove noise using a newly proposed algorithm employing flexible spatial, temporal, and logical constraint rules. Then, a new road network construction algorithm is used to incrementally merge trajectories into a directed graph representing a digital map. Furthermore, the average road traffic volume and speed are calculated and assigned to corresponding road segments. To evaluate the performance of the method, an experiment was conducted using 5.76 million trajectory data points from 200 taxis. The result was qualitatively compared with OpenStreetMap and quantitatively compared with two existing methods based on the F-score. The findings show that our method can automatically generate a road network representing a digital map.
Original Text (This is the original text for your reference.)
An Automatic Road Network Construction Method Using Massive GPS Trajectory Data
Automatically acquiring comprehensive, accurate, and real-time mapping information and translating this information into digital maps are challenging problems. Traditional methods are time consuming and costly because they require expensive field surveying and labor-intensive post-processing. Recently, the ubiquitous use of positioning technology in vehicles and other devices has produced massive amounts of trajectory data, which provide new opportunities for digital map production and updating. This paper presents an automatic method for producing road networks from raw vehicle global positioning system (GPS) trajectory data. First, raw GPS positioning data are processed to remove noise using a newly proposed algorithm employing flexible spatial, temporal, and logical constraint rules. Then, a new road network construction algorithm is used to incrementally merge trajectories into a directed graph representing a digital map. Furthermore, the average road traffic volume and speed are calculated and assigned to corresponding road segments. To evaluate the performance of the method, an experiment was conducted using 5.76 million trajectory data points from 200 taxis. The result was qualitatively compared with OpenStreetMap and quantitatively compared with two existing methods based on the F-score. The findings show that our method can automatically generate a road network representing a digital map.
+More
positioning technology logical constraint rules directed graph laborintensive information automatic method producing road networks speed noise average road traffic volume acquiring comprehensive accurate and realtime mapping information road network construction algorithm field surveying digital map production flexible spatial temporal raw vehicle global positioning system gps trajectory data methods
APA
MLA
Chicago
Yongchuan Zhang,Jiping Liu,Xinlin Qian,Agen Qiu,Fuhao Zhang,.An Automatic Road Network Construction Method Using Massive GPS Trajectory Data. 6 (12),.
Select your report category*
Reason*
New sign-in location:
Last sign-in location:
Last sign-in date: