Preparing For Autonomy – The Importance Of Class Guidelines For Future Autonomous Operations
ClassNK’s Corporate Officer Mr. Fujinami, and Manager Mr. Yamada share insights on recently released class guidelines for Maritime Autonomous Surface Ships (MASS).
With your new guidelines, ClassNK is already providing a kind of future-proofing, as MASS have not yet entered the operational phase. Can you comment on the purpose of providing class guidelines for a future solution, and the inherent challenges?
In the case of MASS, technological developments are ahead of the development of regulations. In addition, innovative technologies like MASS are generally being developed as individual projects for the time being, and it is assumed that the target, degree, and scope of development will also vary from project to project.
Under such circumstances, ClassNK has released the guidelines for two purposes. The first is to understand correctly what kind of technology is currently being developed from the conceptual design phase and share it with relevant parties, including internally in ClassNK. By properly following this process, classification regulations can stay up to date and system owners can establish concrete business usage ideas, while system suppliers and system integrators work to clarify the direction of development.
The second purpose is to clarify the procedure for certification of autonomous vessels. This will make it easier for system suppliers, system integrators, and system owners to understand when to act and what to do, and should serve to accelerate practical implementation of autonomy. When deciding on procedures, particular attention was given to making sure that they do not deviate from conventional maritime industry methodology, in order to avoid confusion. The guidelines were created in a way designed to integrate the new concept of autonomous vessels with the existing methods.
The main challenge for the guidelines is the development of a concrete evaluation method. As the guidelines were created to accommodate various tasks onboard, their content is focused on organizing thinking rather than action, which results in many items being expressed in abstract terms.
From the practical point of view, having analyzed recent trends, we feel that the implementation of technologies for ship-maneuvering system tasks, including automated collision avoidance, automated berthing and un-berthing, and remote ship operation, will be feasible first.
In the future, we plan to target technologies that are close to practical application, establish methods and tools to evaluate their safety, and clarify standards. Ultimately, we aim to establish common procedures and indicators that can objectively evaluate the safety of various MASS.
ClassNK acknowledges the diversity of future solutions for autonomous vessels. Can you share some of the development scenarios you envision for the next five years? The next ten years? How will you prepare for these diverse scenarios?
When considering future development scenarios, two criteria are important: the onboard work of crew becoming subject to automation and autonomy, and the level of automation and autonomy by category.
From the perspective of cost-effectiveness and technical difficulty, developments targeting ship maneuvering will likely progress most quickly over the next five years. However, onboard work is diverse, demanding consideration of engine, cargo, and maintenance-related tasks. For example, the completely autonomous operation threshold could be greatly lowered by switching from reciprocating engine systems to batteries, but as this is dependent on the initial specifications of the ship, this topic is not part of our initial focus.
For coastal ships, there are many opportunities to use a land-based communication infrastructure to support remote vessel operations in congested areas, while a combination of track control systems (TCS) and remote monitoring can be applied in uncongested areas. These operational methods will likely be gradually established. However, data still needs to be collected to establish how current working practices will combine with greater levels of automation and remote control systems. Again, if the number of seafarers working on coastal ships is to be reduced or their qualifications changed, domestic laws covering their employment will need to adapt. Different countries will face different challenges on the amount of change that is possible in five years.
Where ocean-going vessels are concerned, the concept of the autonomous ship with crew is already under development. Automated maneuvering systems, including vessel collision avoidance maneuvering algorithms, have also been developed to a level that can be put to practical use, providing constant monitoring by the crew. Assessing the surrounding conditions with the same accuracy as humans is a challenge, but maintaining an environment that allows crewmembers to fall back on monitoring at any time allows implementation hurdles to be greatly reduced.
It is difficult to predict what the situation will be 10 years from now. Without wishing to be evasive, this will depend on what happens in the next five years. It is already clear what can be achieved using existing technology in terms of streamlining the onboard operation of coastal and ocean-going vessels. The rest depends on the extent to which domestic and international laws adapt to these developments. Implementation of the technology and its further development can be promoted if legislation keeps up; if it takes a long time to develop the laws, the pace of such technological development will inevitably slow down.
However, even if rules development progresses smoothly, it is debatable whether the potential for technological advances will itself change significantly in the next 10 years.
Instead, we expect the more significant movement to take place in ‘social implementation’, including rules development, rather than the sophistication of technology. Time will tell whether the ratio of automated ships comes to represent a low percentage of those in service or rises to more than 50 percent
From the point of view of social acceptance, the anticipated role of class is significant. We should proceed by carefully identifying which of the diverse onboard tasks are suitable for automation or remote control while considering the direction of technological development and identifying and maintaining safety requirements that are essential for social acceptance.
Class NK’s autonomy and automation requirements cover each stage in the life of a vessel, from design and development, to installation, and operation and maintenance. The operational life of a vessel could potentially extend to 20 or 30 years. How will you address the need to safely upgrade the level of vessel autonomy during the operational stage?
Many advanced systems are already installed on ships in service, and we believe that basic efforts will remain the same even for automated navigation systems, including automation and remote control systems. In other words, if a system is updated, the supplier will contact the user and the user will follow the supplier’s instructions to carry out any required processing.
ClassNK’s guidelines also clarify the roles that system suppliers and system owners should fulfill after operation. One example involves system suppliers. In general, the system supplier should carry out the version upgrade of the automation system and deal with any bugs that are found throughout the life cycle of the automation system.
System owners present another case. The system owner should establish procedures for operating the automated system and implement all software changes and updates according to the procedures. In addition, before making any changes that may affect the reliability or safety of the automated system, they shall seek approval from their class society in advance.
However, for automated navigation systems, including automation and remote control systems, there are cases where multiple advanced subsystems are combined, which may complicate maintenance in the operational phase. From this perspective, the system owner may need to have a certain level of capability, and cooperation from the system integrator engaged in the installation will be needed when performing such complicated measures.
It is important to clarify procedures, including maintenance and management, and the roles and responsibilities of the system supplier, system integrator, and system owner, from the time the system is installed.
In addition, as technology develops, requirements in the ClassNK guidelines may also change. By accurately grasping the trends and adapting accordingly, we will provide survey and inspection services suitable for the situation.
How does ClassNK view the role of Remote Operation Centers in the future of MASS?
We believe they will have three major roles: remote monitoring of autonomous vessels with crew onboard, remote control of vessels, and remote monitoring of crewless vessels. Initially, progress towards the first two will likely be carried out in parallel, in preparation for implementing the third role.
The expectation is that safety at sea will be improved thanks to enhanced support from land. However, caution is advisable in relation to the stability of communications. Advanced results can be achieved in an ideal communications environment, but technology that is fully capable of overcoming communication instabilities has yet to be developed. In these circumstances, it must be feasible to operate remote-control facilities even when faced with such instabilities. Careful consideration is needed on how to secure maritime safety in such situations.
On a related note, it is also fair to note that remote operation facilities will likely create new job opportunities, some of which may be attractive to seafarers.
How can MASS contribute to a sustainable future for shipping?
Conventional vessels do not necessarily have serious shortcomings in terms of safety and environmental preservation, but the shortage of seafarers is an urgent issue for the maritime industries. As well as envisaging smaller crew numbers, MASS is being developed in the hope that it will contribute to the safer navigation of ships and a lower frequency of accidents by reducing the burden of those onboard ship.
Any other thoughts on automated and autonomous vessels that ClassNK would like to share with Generation’s readers?
Discussions on MASS thus far have been highly abstract in their nature. For example, when we started to create our guidelines for autonomous ships in 2017, we could not quite envisage the system that should be subject to class certification, and it felt like we were chasing shadows.
However, from around the second half of 2018, trials were underway – mostly in Europe – while various trial projects have been launched in Japan since 2019, suggesting that technological development is accelerating.
Accordingly, it feels like we do not have to do anything in particular to force the development of MASS. By carefully identifying the technical differences from conventional ships, the necessary and sufficient requirements will become apparent and the issues clear. As class alone cannot resolve all of the issues, the development of a foundation that can be tackled by the entire maritime industry is necessary. All stakeholders, including developers of specific technologies, are invited to review ClassNK’s “Guidelines for Automated/Autonomous Operation on ships” and share their feedback with us. Our door is always open.