Graph-Network-Based Vessel Path Planning for Traffic Separation Schemes

Expand
  • a. School of Navigation;b. Hubei Key Laboratory of Inland Shipping Technology;c. State Key Laboratory of Maritime Technology and Safety, Wuhan University of Technology, Wuhan 430063, China

Online published: 2026-07-06

Abstract

To address the poor adaptability of traditional path planning algorithms in complex traffic separation scheme waters, this study establishes a unified path planning modelbased on navigation rules and good seamanship requirements. The model incorporates collision-free constraints, traffic flow direction constraints, and special area navigation rules in-to the solution framework. A two-layer graph network is constructed using Electronic Navigational Chart data: the upper layer represents route topology and regulatory constraints, while the lower layer represents the navigable space defined by safe water depth, obstacles, and prohibited areas. An autonomous digital environment generation method adapted to the two-layer graph network model is proposed. A two-layer hybrid A* search algorithm achieves hierarchical planning, handling heading change limitations and spatial constraints in special areas such as roundabouts through geometrically constrained path gene-ration. Simulation results demonstrate that the proposed method improves planning efficiency while ensuring compliance and collision avoidance safety.

Cite this article

贺益雄a, b, c, 刘鸿昌a, 赵兴亚a, 薛海东a, 张坤a, 段少玺a . Graph-Network-Based Vessel Path Planning for Traffic Separation Schemes[J]. Journal of Shanghai Jiaotong University, 0 : 1 . DOI: 10.16183/j.cnki.jsjtu.2026.038

Outlines

/