Work

SMART-SEA: Ship collision avoidance of stationary structures through integrated machine learning radar image detection and high fidelity maneuvering models

Year: 2025

Type: article

Abstract: Collisions and near-misses between marine vessels and stationary offshore platforms are an increasingly prevalent problem that can result in both property damages and, in extreme cases the loss of life. The majority of collisions with stationary platforms are due to human error such as lack of proper attention, accident misturns, or no staff being present at the bridge. The commercial aviation industry has both Airborne and Terrain Collision Avoidance Systems (ACAS and TCAS, respectively); no such system exists for marine vessels. The SMART SEA project aims to develop a system for marine vessels through the creation of an all-weather hazard avoidance system to help assist ship captains and crews navigate through different weather and stationary hazards that can occur at sea. The system incorporates object detection and classification from raw radar data, a mariner-sourced hazard avoidance database and maneuvering advice based on a 3-tiered maneuvering model. These are presented in a user-friendly manner on a decision dashboard. This paper describes the current progress in developing a prototype of the SMART-SEA system and its components.

Source: Process Safety and Environmental Protection

DOI: 10.1016/j.psep.2025.107688

Authors: Andrew Deng, Yijun Sun, Björn Windén, Freddie Witherden, Ryan Vechan, Mirjam Fürth

Institutions: Texas A&M University, Texas A&M University at Galveston

Language: English

Topic: Maritime Navigation and Safety

Subfield: Ocean Engineering

Field: Engineering

Domain: Physical Sciences

SDG: Industry, innovation and infrastructure

Open Access status: hybrid

Estimated APC paid: $3,500