Generating lane-level road networks from high-precision trajectory data with lane-changing behavior analysis

Abstract

ABSTRACT–Recent advances in mobile mapping systems have facilitated the collection of high-precision trajectory data in centimeter positioning accuracy. It provides the potential to infer lane-level road networks, which are essential for autonomous driving navigation. This task is challenging due to the complicated lane merging and diverging structures as well as the lane-changing patterns in trajectory data. This paper presents a lane-level road network generation method from high-precision trajectory data with lane-changing behavior analysis. Trajectories are firstly partitioned by detecting road intersections and changes in lane structure. Subsequently, in regions with consistent lane structure, a principal curve fitting algorithm is developed to extract lane centerlines. Erroneous lanes generated by lane-changing behavior are pruned based on a constructed lane intersection graph. In regions with merging and diverging lanes, a lane-group fitting algorithm is designed. This algorithm estimates lane locations by incorporating a Gaussian mixture model with lane width prior knowledge and then infers lane-level topological structures using trajectory flow information. The proposed method is evaluated on a real-world high-precision trajectory dataset. Comprehensive experiments demonstrate that it outperforms state-of-the-art methods in four metrics. Under complex scenarios, the method is capable of generating lane-level road networks with higher completeness and fewer fragments.Keywords: Lane-level road networkhigh-precision trajectory datalane-changing behavior AcknowledgmentWe thank the editor and anonymous reviewers for their constructive comments.Data and codes availability statementThe data and codes that support the findings of this study are available at the link: https://doi.org/10.6084/m9.figshare.23529336. A subset of the data is shared for demonstration purposes.Disclosure statementNo potential conflict of interest was reported by the author(s).Notes1 We use the term lane-changing to describe the driving behavior in trajectory data, while lane transition refers to special road network structure where lanes merge or diverge.Additional informationFundingThe work was supported by the Chongqing Technology Innovation and Application Development Project [Grant No. CSTB2022TIAD-DEX0013]; funding from the State Key Laboratory of Intelligent Vehicle Safty Technology and Chongqing Changan Automobile Co. Ltd; and the Fundamental Research Funds for the Central Universities [Grant No. 2042022dx0001].Notes on contributorsMengyue YuanMengyue Yuan is an M.S. student in the School of Remote Sensing and Information Engineering at Wuhan University. Her research interest is geospatial data mining and transportation geography. She contributed to the idea, study design, methodology, implementation, and manuscript writing of this paper.Peng YuePeng Yue is a professor at Wuhan University. He serves as the deputy dean at the School of Remote Sensing and Information Engineering, the director at the Hubei Province Engineering Center for Intelligent Geoprocessing, and the director at the Institute of Geospatial Information and Location Based Services. He supervised this study and contributed to the study design, methodology, and manuscript writing.Can YangCan Yang is a postdoctoral researcher in the School of Remote Sensing and Information Engineering at Wuhan University. His research interest is in trajectory pattern mining and recognition. He contributed to this paper’s idea, study design, methodology, and manuscript writing.Jian LiJian Li is an engineer at the State Key Laboratory of Intelligent Vehicle Safty Technology and Chongqing Changan Automobile Co. Ltd. He contributed to this paper’s idea, study design, and methodology.Kai YanKai Yan is an engineer at the State Key Laboratory of Intelligent Vehicle Safty Technology and Chongqing Changan Automobile Co. Ltd. He contributed to this paper’s idea, study design, and methodology.Chuanwei CaiChuanwei Cai is a Ph.D. student in the School of Remote Sensing and Information Engineering at Wuhan University. He contributed to this paper’s study design and methodology.Chongshan WanChongshan Wan is an M.S. student in the School of Remote Sensing and Information Engineering at Wuhan University. He contributed to this paper’s implementation.