At present, intelligent transportation systems (ITS) is increasingly relying on smartphone positioning, and its accuracy and availability may affect the quality of ITS services. To conduct a comprehensive analysis of the performance and accuracy of smartphone positioning, we released the open-source WHU-Smartphone dataset (the dataset is online at: https://github.com/CJQP/WHU-Smartphone- Dataset, the OSR data is available upon request by contacting the author), which is a large-scale, multi-scene, and multi-mode GNSS dataset collected by four Android smartphones from different manufactures and a low-cost GNSS module with high-precision ground truth. Our dataset contains several hundred driving routes with more than 2500 km from three large cities, including open-sky or shaded urban roads and highways. Multifrequency observation space representation (OSR) corrections of real-time kinematic (RTK) positioning with multiple GNSS systems, particularly those that supported the BeiDou Navigation Satellite System (BDS)-2 and BDS-3 systems, were implemented. We also applied some RTK quality control strategies for smartphone, which include optimizing the observation stochastic model, using the TDCP model to estimate precise temporal positions between adjacent epochs for one-step prediction of positional parameters, and prioritizing the fixing of satellite ambiguity parameters where the optimal and suboptimal solutions are consistent to address the issue of half-cycle slips. Compared with open-sourced software, our strategy significantly improves RTK positioning accuracy in complex environments. Based on this dataset, we have fully verified the positioning accuracy of smartphones and GNSS module in various traffic scenes, which can reach decimeter and centimeter level respectively in open scenes, providing an important basis and reference for smartphone application in intelligent transportation.
WHU-Smartphone: A Large-Scale Multi-Traffic Scene GNSS Dataset of Android Smartphones and a Low-Cost GNSS Module
IEEE Transactions on Intelligent Transportation Systems ; 25 , 11 ; 17993-18005
01.11.2024
5012716 byte
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
Application of UofC Model based Multi-GNSS PPP to Smartphones GNSS Positioning
British Library Conference Proceedings | 2021
|PEDESTRIAN NAVIGATION IN URBAN AREA USING GNSS RAW DATA WITH SMARTPHONES
British Library Conference Proceedings | 2020
|Learning GNSS Positioning Corrections for Smartphones Using Graph Convolution Neural Networks
DOAJ | 2023
|