Detection Method for 3D Lanes Based on Graph Relationship Optimization Integrating Point and Lane Features

Authors

• Yanji Jiang, School of Software, Liaoning Technical University, Huludao 125105, China; OpenSafe Lab, Utcer (Jiangsu ) Safety Technology Co ., Ltd, Suzhou 215100, China
• Xingyi Xiao, School of Software, Liaoning Technical University, Huludao 125105, China; OpenSafe Lab, Utcer (Jiangsu ) Safety Technology Co ., Ltd, Suzhou 215100, China
• Hao Dong, OpenSafe Lab, Utcer (Jiangsu ) Safety Technology Co ., Ltd, Suzhou 215100, China; Suzhou Automobile Research Institute (Xiangcheng ), Tsinghua University, Suzhou 215100, China; School of Automotive Engineering, Suzhou University of Technology, Changshu 215500, China
• Miao Yu, China FAW Group Research and Development Institute, Changchun 130000, China
• Jinshan Huang, Engineering & Technology Department, China FAW Group Co ., Ltd, Changchun 130012, China
• Daqian Liu, School of Software, Liaoning Technical University, Huludao 125105, China
• Bowen Fei, School of Software, Liaoning Technical University, Huludao 125105, China

Abstract

Under complex road conditions, the thin and elongated structure and small proportion of lanes lead to blurred visual features and insufficient positioning accuracy, which in turn threatens the road safety of autonomous driving. To address these issues, a 3D lane detection method or graph-based point and lane optimization network (GPLNet), based on graph relationship optimization integrating point and lane features, was proposed. Preliminary feature extraction was completed by the backbone network. 3D spatial positional coding with geometric constraints was obtained through a joint query embedding generation module. A graph relationship optimization network was utilized to perform graph relationship calculation and optimization modeling on point- and lane-level features of lanes to enhance the context awareness capability of lanes. Lane prediction and loss function calculation were achieved with a 3D prediction head. Experimental results indicate that the overall performance of the proposed method is superior to existing mainstream 3D lane detection algorithms, and the detection accuracy is higher in low- visibility lane scenarios, which verifies the effectiveness of the method.

Recommended Citation

Jiang, Yanji; Xiao, Xingyi; Dong, Hao; Yu, Miao; Huang, Jinshan; Liu, Daqian; and Fei, Bowen (2026) "Detection Method for 3D Lanes Based on Graph Relationship Optimization Integrating Point and Lane Features," Journal of System Simulation: Vol. 38: Iss. 5, Article 12.
DOI: 10.16182/j.issn1004731x.joss.25-0591
Available at: https://dc-china-simulation.researchcommons.org/journal/vol38/iss5/12

First Page

1303

Last Page

1319

CLC

TP391.9; TP311

Recommended Citation

Jiang Yanji, Xiao Xingyi, Dong Hao, et al. Detection Method for 3D Lanes Based on Graph Relationship Optimization Integrating Point and Lane Features[J]. Journal of System Simulation, 2026, 38(5): 1303-1319.

Corresponding Author

Dong Hao

DOI

10.16182/j.issn1004731x.joss.25-0591