Towards robust intelligent shipping via graph-informed representation, generative routing, and certified maneuvering

• Hierarchical Spatio-Temporal Representation. We propose a graph-based maritime representation that integrates AIS trajectories, bathymetry, and regulatory constraints into multi-scale embeddings, enabling preference-conditioned and structure-aware route generation. • Constraint-Embedded Generative Routing. We design a generative routing network with in-loop feasibility projection, ensuring that candidate routes are feasible by construction with respect to geometric, dynamic, and regulatory manifolds. • Certified Robust Maneuvering. Develops a maneuvering controller that combines Model Predictive Control (MPC), robust invariant tube sets, and control barrier functions, providing formal guarantees of safety and robustness under environmental disturbances and multi-ship interactions. Intelligent shipping requires route planning frameworks that jointly capture operator preferences, vessel maneuverability, and verifiable safety under environmental disturbances. For this, we propose a unified data-to-control framework for maritime routing in which representation, generation, and execution are tightly coupled rather than assembled in isolation. A preference-conditioned hierarchical spatio-temporal graph fuses AIS traffic, bathymetry, and regulatory layers to gate the generative search; a constraint-embedded generative router performs in-loop feasibility projection onto a joint geometric-dynamic-regulatory manifold, producing feasible-by-construction waypoint-speed sequences; and a certified maneuvering layer attaches tube model predictive control and control-barrier constraints, yielding verifiable artifacts (tube membership, barrier margins) that form a route-to-execution contract and close the planning-execution loop. Extensive experiments demonstrate that the framework generates preference-aware, maneuverability-consistent, and provably safe maritime routes, establishing a solid foundation for large-scale Intelligent Shipping applications.

Authors

Institutions

Publication Details

Journal
Ocean Engineering
Published
2025-12-26
DOI
https://doi.org/10.1016/j.oceaneng.2025.123957
Primary Topic
Maritime Navigation and Safety
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Towards robust intelligent shipping via graph-informed representation, generative routing, and certified maneuvering

Xiangjun Yin, Xiongfei Geng, Qiaochan Yu, Hui Feng et al.
Ocean Engineering
Maritime Navigation and Safety
article

Towards robust intelligent shipping via graph-informed representation, generative routing, and certified maneuvering

Xiangjun Yin, Xiongfei Geng, Qiaochan Yu, Hui Feng, Huihui Yue
article en

Abstract

• Hierarchical Spatio-Temporal Representation. We propose a graph-based maritime representation that integrates AIS trajectories, bathymetry, and regulatory constraints into multi-scale embeddings, enabling preference-conditioned and structure-aware route generation. • Constraint-Embedded Generative Routing. We design a generative routing network with in-loop feasibility projection, ensuring that candidate routes are feasible by construction with respect to geometric, dynamic, and regulatory manifolds. • Certified Robust Maneuvering. Develops a maneuvering controller that combines Model Predictive Control (MPC), robust invariant tube sets, and control barrier functions, providing formal guarantees of safety and robustness under environmental disturbances and multi-ship interactions. Intelligent shipping requires route planning frameworks that jointly capture operator preferences, vessel maneuverability, and verifiable safety under environmental disturbances. For this, we propose a unified data-to-control framework for maritime routing in which representation, generation, and execution are tightly coupled rather than assembled in isolation. A preference-conditioned hierarchical spatio-temporal graph fuses AIS traffic, bathymetry, and regulatory layers to gate the generative search; a constraint-embedded generative router performs in-loop feasibility projection onto a joint geometric-dynamic-regulatory manifold, producing feasible-by-construction waypoint-speed sequences; and a certified maneuvering layer attaches tube model predictive control and control-barrier constraints, yielding verifiable artifacts (tube membership, barrier margins) that form a route-to-execution contract and close the planning-execution loop. Extensive experiments demonstrate that the framework generates preference-aware, maneuverability-consistent, and provably safe maritime routes, establishing a solid foundation for large-scale Intelligent Shipping applications.

Ocean EngineeringVol. 348
Tianjin University (CN), National University of Singapore (SG), Nanyang Technological University (SG), Nankai University (CN), China Waterborne Transport Research Institute (CN)
Life below water
Openalex Percentile: Top 36%
Maritime Navigation and Safety
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.