This post is also available in:
עברית (Hebrew)
Autonomous maritime systems are advancing faster than the frameworks designed to prove they are safe. Unmanned surface vessels increasingly rely on AI-based perception to detect obstacles, identify other vessels, and make navigation decisions in complex sea states. While the technology is improving rapidly, regulators and operators still face a central challenge: how to verify and validate that autonomous behavior will remain reliable in unpredictable real-world conditions.
A new international collaboration aims to tackle that problem at its root. Singapore’s Defence Science and Technology Agency and the Republic of Korea’s Korean Register have agreed to jointly develop a structured verification and validation framework for autonomous maritime technologies, with a particular focus on AI-driven perception systems used aboard unmanned surface vessels. The effort is designed to move beyond ad hoc testing toward a consistent, repeatable way of assessing autonomy safety.
According to Maritime Executive, the framework will examine how AI perception algorithms perform across a wide range of scenarios. By applying systems engineering principles alongside maritime certification standards, the framework is intended to provide clearer evidence of whether an autonomous vessel can be trusted to operate safely.
A key goal of the initiative is standardization. Today, autonomous vessel developers often rely on custom testing methods that vary widely between projects and countries. The joint effort seeks to align technical evaluation with policy and certification requirements, creating a foundation that can be adopted internationally. This would allow regulators and operators to compare systems more easily and reduce uncertainty around approval and deployment.
Navies are increasingly deploying unmanned surface vessels for surveillance, patrol, logistics, and mine countermeasures. These platforms often operate with minimal human oversight and in contested or congested environments. A robust verification and validation process helps ensure that autonomous systems behave predictably, reducing operational risk and increasing confidence in their use alongside crewed vessels.
The collaboration also reflects the dual-use nature of maritime autonomy. Commercial shipping, port operations, and offshore infrastructure face many of the same safety challenges as military systems. Establishing a trusted evaluation framework benefits both domains by accelerating adoption while maintaining high safety thresholds.
As autonomous ships move from trials to routine operations, proving that AI can be relied upon at sea will be just as important as the technology itself. By focusing on how autonomy is tested and validated, this initiative addresses a critical gap that could shape the future of unmanned maritime operations worldwide.
