In the bustling waters of coastal regions, where ships navigate in close proximity, the risk of collisions is a constant concern. A recent study, led by Hee-Jin Lee from the Division of Education and Planning at the Korea Institute of Maritime and Fisheries Technology in Busan, aims to tackle this issue head-on. Published in the Journal of Marine Science and Engineering, the research proposes a novel system that integrates the Computed Distance at Collision (CDC) method with an Adaptive Neuro-Fuzzy Inference System (ANFIS) to predict collision risks more accurately.
So, what does this mean for maritime professionals? Traditionally, collision risk has been assessed using the Distance at Closest Point of Approach (DCPA), which relies on the position of GPS antennas. However, this method doesn’t account for the actual shape of the ship’s hull or the potential collision point. Enter CDC, which directly reflects the hull geometry and boundary conditions specific to different ship types. “The CDC method enables a more realistic assessment of collision risk,” Lee explains, “by accounting for the hull geometry and boundary conditions specific to different ship types.”
The system was put to the test using ship motion simulations involving bulk and container ships at varying speeds and crossing angles. The results were enlightening. The time to collision decreased with higher speeds and increased with wider crossing angles. Interestingly, the bulk carrier exhibited a wider collision-prone angle range and a greater sensitivity to speed changes than the container ship. This highlights the differences in maneuverability and risk response between ship types.
The proposed system uses four input variables—ship speed, crossing angle, remaining time, and remaining distance—to infer the collision risk index (CRI). This allows for a more nuanced and vessel-specific assessment than traditional CPA-based indicators. The system’s real-time applicability and accurate risk differentiation across scenarios make it a promising tool for enhancing situational awareness and proactive risk mitigation.
For the maritime sector, this research opens up opportunities for improving safety in Maritime Autonomous Surface Ship (MASS) and Vessel Traffic System (VTS) environments. By providing a more accurate and vessel-specific assessment of collision risk, the system can support real-time decision-making and help prevent accidents. As Lee notes, “Future work will focus on real-time CDC optimization and extending the model to accommodate diverse ship types and encounter geometries.”
In the ever-evolving landscape of maritime technology, this research is a significant step forward. It’s not just about predicting collisions; it’s about understanding the nuances of ship behavior and using that knowledge to enhance safety and efficiency. For maritime professionals, this means a tool that can provide more accurate risk assessments, leading to better decision-making and ultimately, safer waters.