In the vast, unpredictable world of maritime navigation, human error remains a significant factor in collision incidents. Leszek Misztal, a researcher from the Faculty of Computer Science and Telecommunication at the Maritime University of Szczecin, Poland, has been delving into this issue, and his findings, published in ‘Applied Sciences’ (Zastosowane Nauki), are shedding new light on how technology can mitigate these risks.
Misztal’s research focuses on the communication breakdowns that often precede maritime accidents. He points out that “human errors accounted for the following: A total of 38.3% of all causes in 2019; A total of 46% of all causes in 2020; A total of 71.4% of all causes in 2021; A total of 64% of all causes in 2022.” These stats underscore the pressing need for improved communication systems at sea.
So, what’s the solution? Misztal and his team propose an automatic communication system that uses AI and metaontology—basically, a way to structure information so that machines can understand and act on it. The system translates natural language into a format that autonomous systems can process, and vice versa. This isn’t just about translating words; it’s about ensuring that critical information is exchanged accurately and efficiently, especially in high-stress situations.
The implications for the maritime industry are substantial. Imagine a world where ships can communicate directly with each other, reducing the chances of misinterpretation or missed messages. This could lead to fewer collisions, less stress for navigators, and ultimately, safer seas. The system could also be a game-changer for routine navigation, allowing for earlier communication and more time to analyze and respond to developing situations.
From a commercial perspective, the potential benefits are clear. Fewer accidents mean lower insurance premiums, reduced downtime, and fewer costly investigations. Additionally, the technology could be integrated into existing systems, providing a competitive edge for early adopters.
Misztal’s work is still in the prototype stage, but the results so far are promising. The system has been tested on various collision scenarios, and the initial findings suggest that it could significantly improve communication and decision-making in critical situations. The next steps involve refining the machine learning methods and expanding the database of natural-language sentences to better reflect real-world communication.
As Misztal puts it, “The application of metaontology for semi-autonomous as well as machine-to-machine communication systems may contribute to avoiding collisions (or significantly reduce collision consequences).” This isn’t just about avoiding disasters; it’s about creating a more efficient, safer, and more predictable maritime environment.
