.png){kind=icon}
.png){kind=icon}
Return
A multi-vector-field based inland ship path planning using an anisotropic fast marching method
.png){kind=icon}
PRE
AI
DOI：10.1016/j.oceaneng.2025.121303.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The safety and efficiency of ships navigating in inland rivers have been greatly emphasized due to the rapid development of the inland transportation. To enhance inland navigation safety, this paper aims to develop a multi-vector field path planning model for inland ships based on an anisotropic fast marching method (AFM). Firstly, a fast marching method (FMM) is employed to generate both repulsive and attractive fields. Subsequently, the gradients of the water depth and the current data are processed to derive the water depth and current fields. Next, the safety field can be obtained for a combined repulsive, attractive and water depth field. The safety field and the current field can then be integrated using an adaptive dynamic weighting method to generate a synthetic field. Finally, the AFM is utilized to compute the synthetic field to obtain a distance map, before an optimal path is obtained using a gradient descent method. The results demonstrate that the generated paths closely match actual AIS trajectories, providing valuable insights into inland navigation safety. The paper makes a significant contribution to inland shipping safety by offering feasible reference paths for complex environmental scenarios that currently lack practical solutions.
Keywords:
Intelligent ship navigation
Inland shipping
Ship path planning
Multi-vector field
Anisotropic fast marching
Journal
IF:
5.5
Papers:
5.8K
Citations:
7.6W