Setting Course, an ABS Podcast

Autonomy in the Maritime Industry with ABB

• American Bureau of Shipping • Season 1 • Episode 1

Share episode

As vessels become increasingly complex with new technologies and fuels, the application of autonomous technologies to support human crew decision-making is gaining ground in the maritime industry.

On this episode of Setting Course, Chih Wei Lui, ABS Senior Managing Principal Engineer, and Kalevi Tervo, ABB Corporate Executive Engineer and Global Program Manager, join host John Snyder, Managing Editor of Riviera Maritime Media, to broadly discuss the impact of autonomous technologies on the maritime industry.

Lui and Tervo discuss specific applications, the evolving regulatory landscape, and the evolving role of crew onboard vessels with autonomous technologies.

If you like this episode, leave a review and share the episode with your friends.

Key Points

  • Autonomy in the maritime industry is in an exciting phase with a lot of work happening behind the scenes.
  • The focus is on the practical application of autonomous technology in vessel operations to improve efficiency.
  • Human beings are still crucial for decision-making, while machines handle tedious and data-intensive tasks.
  • These technologies are being piloted in various sectors of the maritime industry.


Guests
Chih Wei Lui is a Senior Managing Principal Engineer with ABS and leads the electrical plan review team in Singapore. Lui has been with ABS for 16 years working in various roles in Engineering plan review and Technology in Singapore and Dubai. He has extensive experience in maritime regulations and electrical and automation systems. He is a subject matter expert on smart and autonomous shipping and is actively involved in global efforts on this topic. He represents ABS on the IACS (International Association of Classification Societies) Expert Group on MASS (Maritime Autonomous Surface Ships). He was the lead author of the ABS Requirements for Autonomous and Remote Control Functions and the Advisory on Autonomous Functionality.

Dr. Kalevi Tervo received his D.Sc. (Tech.) degree in Control Engineering from Aalto University, Finland, in 2011, where he served in various positions during 2005-2011. At Aalto University, he led a research team involved in building the first IoT systems for mobile working machines with major Finnish working machine companies. Since joining ABB in 2011, he has led R&D projects focused on increasing ship automation and performance through data analytics, optimization and control. In his current role as a Corporate Executive Engineer, Dr. Tervo leads a major R&D program developing efficient, sustainable and intelligent shipping solutions.

[00:00:00] John Snyder: You're listening to Setting Course, an ABS podcast. Join us as we navigate the latest trends, developments, and challenges facing the rapidly evolving maritime and offshore industries. Catch every episode at www.eagle.org and podcast platforms everywhere. Hello and welcome to the show. I'm John Snyder, Managing Editor of Riviera Maritime Media, and I'll be your host.

Today we're going to talk about autonomy in the marine industry, including its uptake, the technologies making it possible, and regulations governing its use. To address these and other questions, we are fortunate to have two experts in the field. One is Kalevi Tervo. Kalevi is a Corporate Executive Engineer and Global Program Manager at ABB.

And our other guest is Lui Chi Wei. Lui is Senior Managing Principal Engineer, Global Engineering in Singapore for ABS. And as a way of background, ABS and ABB have a long history of working together on advancing new technologies. Among the projects the two organizations have worked on is a remotely operated harbor tug that received the world's first remote control navigation notation from ABS.

Gentlemen, welcome to the show. Lui I'm going to ask you to start us off. Give us an overview of the current state of autonomy in the maritime industry. Where are we? 

[00:01:40] Lui Chih Wei: Firstly, thanks for having us, John. So, autonomy is actually entering a very exciting phase in the maritime industry. So, if you recall a couple of years ago, there was a lot of hype.

But since then, to the people who are not involved directly in autonomous development, it may seem that the excitement has died down. There's a bit of lull in activities. But, don’t be mistaken. Behind the scenes, actually a lot of work is going on. So, in fact, at the International Maritime Organization, they have just completed a meeting which discusses the draft of the autonomous code. Various stakeholders, various technology developers and class societies like ourselves are also busy working behind the scenes.

So, we've actually gone past the introduction or the development of the technology. What we are looking at in detail now is actually the practical application of autonomous technology in vessel operations today. So, we are looking at autonomous technology from various viewpoints. It is not just the autonomous navigation, but what the industry is doing as a whole is actually to try to use digital technologies to, in a sense, automate wherever processes that are being taking place on board the vessels by human being. So, in fact, they are using such new technologies, digital and autonomous technology to automate and to improve the efficiency of vessel operations. 

[00:03:16] John Snyder: So helping in, say, for example, helping the crew or captain in the decision making process.

[00:03:23] Lui Chih Wei: Yes, decision making process and also to kind of a fill in the gaps. You know, some things are actually better done well by human beings. Whereas, for other parts of the operations, especially when we are thinking of trying to get the machinery on board to be working in its optimal zones and optimal operating characteristics, that's where the machine comes in better. 

So, what we're doing is we're trying to get a match, where human beings are good at using the cognitive intelligence to evaluate various conflicting information, the human beings will come in. But for things which are more tedious and which are data intensive, which are very critical and which require fast processing time, that's where the machines come in. So, we are seeing as we go forward, kind of using technology to optimize the human beings’ involvement in this operational process. 

[00:04:19] John Snyder: Kalevi, do you have anything to add to that? 

[00:04:21] Kalevi Tervo: I totally agree with Lui. I think the hype in it, to some extent, is over and now we're seeing practical implementations that can provide benefits and, specifically as Lui also pointed out, we're seeing kind of, really thinking of — not really thinking of where to replace the crew with machine, rather, seeing that what are the aspects that the computer can do better, where computer are at their best. And then also how do they kind of work side by side with the crew so that the crew can use their strengths whenever that's necessary.

So really getting the best out of both of those and that's, I think, it's a healthy development in the discussion and in the whole industry. 

[00:05:16] John Snyder: Now, Kalevi, what kind of technologies are making these autonomous functions possible? 

[00:05:24] Kalevi Tervo: Well, generally, I think that the marine industry has benefited a lot from megatrends related to automation, robotics and AI with the scientific research and R&D that has happened in other industries way before marine.

And now we're seeing, specifically, I would say, starting from data, so, the availability of edge to cloud technologies, enabling systematic data collection, and they're also enabling availability of data. From there, we can go to data driven modeling and deployment of models more easily.

We also have more affordable processing power on board that can enable running these algorithms. So now we're seeing more and more deployment of AI-driven perception systems and situational awareness systems, for example, that help to do the lookout task, monitoring and detecting navigational hazards from surroundings, and alerting the crew, and performing that relentlessly and continuously without the risk of human error.

That's one thing. And then the other thing is when you combine all of those data and bring more advanced algorithms on top of that to enable support and even autonomous, risk assessment, decision making, trajectory planning, and more advanced motion control of the vessel. So, it's a whole suite of let's say technologies that kind of enables this development.

[00:07:01] John Snyder: Now, Lui, do you have anything to add to what Kalevi said? 

[00:07:04] Lui Chih Wei: Yes, so, like what Kalevi mentioned, autonomous technology is not a singular technology. It's actually a convergence of various technologies — AI, for example, connectivity, sensor technologies, the data analytics. So, the industry, although we can see autonomous is actually in its infancy, it's also because all these technologies are trying to move in the same direction, but there will come a time that all the readiness of each of the individual technologies are ready, and it overcomes the barrier that is in front of it.

And once that's ready, we think that the industry will really make full use of autonomous technology to improve their operations and processes. 

[00:07:51] John Snyder: Lui, where are we seeing these technologies being piloted? 

[00:07:57] Lui Chih Wei: So, we see them in various, in fact, all sectors of the industry, because, in a nutshell, it can be divided into the domestic or short seas segment, and also at the other side, at the other end of the spectrum, we see them in the ocean going voyages. 

So where we see them being used are, for example, the very first technology that kinds of falls into the autonomous bucket — we see a situational awareness, autonomous navigation and collision detection and collision avoidance. We've seen it's already being trialed by various stakeholders, including by ABB for short sea vessels for tugboats.

They’re also being used increasingly in ocean going voyages, although the technology may not be used for autonomous navigation, but we see that situational awareness, which is one of the foundational core technology of autonomous navigation, they're actually being used today in conventional vessels to help give greater assistance to seafarers while navigating through the various busy waters.

[00:09:08] John Snyder: Now Kalevi, Lui mentioned ABB's projects. Could you highlight some of those for us? 

[00:09:15] Kalevi Tervo: Yes. In short distance vessels, one of the highlight projects that we ran a couple of years ago was the autonomous and remote controlled tug project in Singapore, where we demonstrated autonomous and remote operation in coastal, busy coastal waters and it was quite a successful project where we learned a lot. 

The other aspect that I'd like to highlight also, when it comes to the technologies and a little bit of the, let's say ship types, we're also focusing quite heavily in kind of enabling more advanced, higher-level of abstraction, automatic control. So, instead of going from using manual levers and an autopilot, but really have like a generic all speed motion control system that can enable easier manual control, but then also it's simpler to implement automation on top of that.

And then we are seeing, for example, in cruise vessels where we’ve deployed quite a few of these systems already where the user is able to operate the vessel in a higher abstraction level. And also then, on top of that, we can bring more automated functionalities, such as automatic crash stop, which enables to perform a crash stop in an optimal way while keeping the crew in control of the course of the vessel.

And then that's of course quite beneficial in those kinds of situations. 

[00:11:02] John Snyder: So certainly autonomy, in that instance, is increasing the safety factor then? 

[00:11:09] Kalevi Tervo: Yes, definitely. That's one of the main drivers, especially in bigger vessels that — to keep the vessel and the crew and the environment safe.

[00:11:24] John Snyder: And Lui, did you have anything to add to what Kalevi said? 

[00:11:27] Lui Chih Wei: Yes, John. Really, you've actually touched on a very important point. So, the development we see today, it is not just technology for technology's sake. Increasing safety is a very important key factor. So, various studies have also pointed out that quite a large percentage of incidents, of accidents on the high seas, are actually a result of human error.

So, what autonomous technology is also trying to do is basically, by utilizing such technology to, in a sense, lighten the cognitive load on the seafarers. As we've heard from various stakeholders, from various parties, we have heard feedback that we are asking a lot from the seafarers.

Not only that you have to focus on what the critical job is at hand, there is also a large number of administrative tasks which they need to do. So in a sense, by using technology to lighten the cognitive load on them, it allows them to focus on what they really need to focus on, which is the critical work on hand. And hopefully that will increase safety as we go forward. 

[00:12:44] John Snyder: Of course, there are hurdles to implementing autonomy in the industry. And I was wondering, Lui, if you could start us up, what hurdles do you see the industry has to overcome to really expand the uptake of autonomous technologies?

[00:13:05] Lui Chih Wei: So, really, the key two hurdles would be regulatory and standardization. Currently, in IMO regulations, most of the key regulations require a human being to be in charge of the functions. In fact, there's also a requirement that the human beings have to be on the vessel to keep watch. So we have to bear in mind that the maritime industry is a long industry.

Most of these regulations have been written decades ago, where the technologies we see today were not there yet. So, now that we have new technology coming out, which can do a better job, it would be fair or it would be prudent to see how we can leverage on such technology to carry out those operations.

So the IMO is aware of that. Various countries are aware of that. Which is why there's a key effort at IMO to try to draft what we call a MASS code, or Maritime Autonomous Surface Ships code, to give some sort of framework to allow autonomous operations within the maritime regulatory regime. And the other side of the coin is also without any regulations, without any form of standardizations, what we see is that various parties are utilizing different type of assumptions for them to carry out that work.

So, it's kind of — there's pros and cons in not having standardization at this stage of the development of maritime autonomy. Firstly, of course, the pro will be without rigid standards, without prescriptive standards, it gives a little bit of leeway for technology providers to try different technologies, to try different ways of doing things.

But on the other side of the coin, the owners and the other stakeholders are a little bit kind of confused on what the different terms used by different stakeholders actually mean. So that also is a little bit of a hindrance to allow greater adoption of such technologies. So in a nutshell, the industry is realizing that we need a practical regulatory framework to allow such use.

We also, as we develop further, and this could be the medium term, we need to have some sort of standard terminologies, standard descriptions of the functions and systems which we used to enable autonomous operations. 

[00:15:38] John Snyder: Now, Kalevi, as a technology provider, do you see that as a huge hurdle, the standardization that Lui has mentioned?

[00:15:47] Kalevi Tervo: It is a huge, huge hurdle, but I think that's exactly what is needed. I think it's quite important that everybody gets on the same page, and it will take time because commercial deployment of autonomous technologies, it'll require the ship owner and operator, it'll require technology vendors, classification societies, flag state authorities, insurance companies and all that.

Everybody needs to understand the technology, the operations with it, the limitations and the risks so that all stakeholders can do — let's say they're part of the process to document and to standardize. And that's the only way to get to a state where you can actually, if you think about the whole value chain, from the customer, who wants the owner, who wants to build a ship that they actually know what's available because there are standards and terminologies and all that. They know what's available and then the shipyards can inquire vendors those kind of capabilities and so on.

So everybody needs to kind of do their part. And at the moment, of course, the situation is a bit asymmetric that some stakeholders, of course, have a bit more information of these than some of the others. And it will take time. And we need to move step by step to really get there. 

[00:17:17] John Snyder: Now, Lui, earlier in our conversation, you mentioned the first draft of the MASS code was being reviewed at IMO. Where are we now in that process? 

[00:17:31] Lui Chih Wei: So, IMO has set a quite an ambitious target. They aim to have the MASS code enter into force by the 1st of January 2028. So, they are still in the drafting process. In fact, currently, the second draft is already out. So IMO has come out with a framework which details the foundational issues that they will need to look into for the implementation of autonomous remote control functions. In addition to that, they have also included two major chapters for the functions. One is the autonomous functions and second is remote operations. So the draft is out, the industry is poring over it trying to give advice, trying to influence how the final code will come out.

So it's kind of a race against time because we really need to get the code out by 2025 in order for it to enter the force by 2028. And it's like any endeavor where there's multiple parties. It will not be a walk in the park. But what we see is that there's really a lot of strong will from the industry, from the various stakeholders, from the flag states, port states and industry, to be able to come up with a goal based framework, which is practical and yet flexible enough which will not hinder the development of new technologies.

[00:19:05] John Snyder: Do you see any particular topics or concerns that you think that the code should address? Maybe if you could take that one first, Lui, and then Kalevi afterwards. 

[00:19:18] Lui Chih Wei: Yes, so, traditionally the maritime industry looks at safety from four factors or four fields. Firstly, you have hull and structures. Secondly, you have machinery. Third, piping. And fourth, electrical and automation. Now, as we move forward, what we see is that artificial intelligence algorithms are increasingly being used in the design of the vessel, design of systems, and also in the operations. So what we also feel that the code should be addressesing is how are we to, in a sense, check and how do we verify, AI technologies being put on board vessels as we go forward. 

[00:20:09] Kalevi Tervo: Yes, certainly. And I think also, when it comes to AI technologies, it's not only in verifying those when delivered, but also how do we maintain it? Because data driven modeling is something that you have to maintain on a continuous basis.

How do we maintain and how do we verify the integrity of those updates and all that? That's one thing. And maybe the other thing which I think should be also emphasized in the MASS code is kind of considering the ship as a system. Because now at the moment a different kind of subsystems on board vessels are on purpose, let's say, designed to be isolated so that you have fault tolerance from that perspective. But the challenge there is that now they have also been driven to a state that they need to be operated so that — human crew manages the interfaces and interplay between different subsystems, and when you go to higher level of automation, that is no longer an option.

You need to consider how the ship behaves as a system, how each and every subsystem should play together so that from a top level, the operations is safe and the integrity is guaranteed. And this is maybe something that I think there should be more focus and more discussion on this one as well.

[00:21:45] John Snyder: Lui, do you have anything to add to that? 

[00:21:47] Lui Chih Wei: Yes, so, in a sense, I agree with Kalevi. So, the ship traditionally has the systems operating, in a sense, in its silos. The ship's systems typically are also provided by different vendors in the supply chain. But as we go forward, looking at the overall aim of the industry, which is to decarbonize and get to a net-zero future, that is no longer an option.

We need to optimize the performance of the vessel. And one way to optimize the performance of the vessel is to ensure that the various systems work within the optimum range of each other. So the only way to go forward to that is to take a system of systems view of the vessel where you're not looking at the various systems individually, but it all has to work holistically together. As like what Kalevi mentioned, the whole ship as a system. 

[00:22:48] John Snyder: And speaking of systems, of course, probably one of the ship's most important safety systems is the crew. What kind of buy in do you have to have from the crew for autonomous or remote control functions to be successful? Maybe Kalevi, if you want to address that. 

[00:23:12] Kalevi Tervo: Firstly, I think it's quite important to emphasize that technologies are supposed to make people's lives easier. If we end up in a situation that we have, let's say, everything else that you have now on board and then on top of that, the autonomous things and it would just bring more work for the crew. That's a no go. You can't succeed with that. That will not succeed. So, of course we need to ensure that these technologies not only work as technologies, but so that they help the lives, improve the work and decrease the cognitive workload of the crew so that the technologies will kind of support and be more, let's say, responsible of those parts of the operations where they are designed to be responsible of, when that's kind of in the comfort zone of the technology, so to speak. And then ensuring that works smoothly so that we can, the crew can use their strength when needed and have more, let's say, meaningful time in the meantime, when they are not necessarily needed that much. 

[00:24:38] John Snyder: And Lui, did you have anything to add to that? 

[00:24:40] Lui Chih Wei: I second what Kalevi is saying. So, the best way to get the buy-in from the seafarers is if these technologies that we are making them use, or that we're putting on board, make their lives easier and also, in a sense, create a safer, uh, working environment for them. So that would be — once seafarers have tasted the easier life or easier working conditions brought by such technologies, I believe they will hop on board the train and upgrade themselves, upskill themselves and use such technologies to their benefit.

[00:25:19] John Snyder: And do you see, certainly we've just come out of the pandemic, do you see that has created more of a receptive atmosphere for using autonomous and remote control applications? Do you see the industry more ready to take up the technology? 

[00:25:41] Lui Chih Wei: Yes, John, definitely. So, the COVID pandemic has taught the whole world and, of course, in the context of a corporation from top management down to their middle managers down to the engineer and working level, that you can actually carry out work remotely and this work is still efficient and still effective. And in addition to bringing about and kind of educating the whole of the workforce, how to carry out work remotely, with the pandemic, we've forced — it also has impacted the risk assessment that each company, each organization is carrying out to enable their operations to go along unhindered. So, on board a vessel, if we have to take into account that in the future, somebody on board a vessel may get COVID or certain other type of infectious diseases. So, in a risk assessment type, we have to take into account that in case this happened, the entire vessel will in a sense be considered down.

So with remote operations, with autonomous operations, we still can allow operations to carry on with control or supervision from a remote station or an onshore station. So, yes, the pandemic has really given us a big boost to the development of autonomous and remote control technologies.

[00:27:12] John Snyder: Do you agree with that Kalevi? 

[00:27:14] Kalevi Tervo: Certainly. I think during the pandemic, everybody had to find new ways to do and to maintain the business and deliver the projects, especially in the maritime industry. I'm more in the global industry and from ABB's perspective, for example, we've delivered ships all around the world.

So during the pandemic, of course, when you can’t travel, you have to find new ways to do commissioning more remotely, to do more maintenance more remotely, and also now classification societies such as ABS are also enabling, doing a remote survey for the system so that crew doesn't and people don't have to travel, for example, a factory acceptance tests for the systems. It's quite a lot of benefits for the whole industry to reduce traveling and then maintain the operations.

And maybe the other thing also that, even if they are not necessarily seeing in a near future, like ocean-going remote control vessels, but what we are seeing already today is that more and more work and tasks are done in collaboration with the onshore crew and the onboard crew, such as planning and managing the voyage, certain risk assessment, different kinds of operations that can be more efficient to be done from onshore.

And this is, of course, what we're seeing happening more and more in the industry. 

[00:29:05] John Snyder: Now, we're just about out of time for the podcast, but I thought if there was anything you wanted to leave as far as final thoughts for our listeners. Maybe Lui, if you had any final key takeaways for the listeners regarding autonomous technology.

[00:29:25] Lui Chih Wei: Yeah, looking at autonomous technology, the key takeaway is it is not technology for technology's sake. There are real drivers that demand such technology, or there are real drivers that require optimum use of such technology, which is to improve safety. So, going forward, what we'll see is that the industry is currently in the midst of great transformation. Like Kalevi has alluded to, we are moving with the new technologies coming up, the new requirement to increase safety. We are seeing that there's a need of a mindset change to be able to adopt and implement such technologies efficiently to the benefit of industry and the benefit of all.

[00:30:13] Kalevi Tervo: Yes, to add on that — of course, I agree with what Lui said — I think we're already seeing today that those, let's say first movers, who invest in these autonomous technologies today, when it comes to situational awareness, when it comes to more advanced control, collision detection and collision avoidance, even to some extent, maybe in short distance remote control, they are gaining a competitive advantage and improving their, let's say, business by means of these technologies. And we will see this more and more in the future, especially when the regulatory picks up and the whole industry takes the next steps.

So I think we're just in the beginning, but already at speed, I would say. 

[00:31:06] John Snyder: Well, safety and competitive advantages, I think you've hit key issues certainly for vessel owners. Gentlemen, thank you both for joining me today on Setting Course, an ABS podcast. Thank you very much.

[00:31:21] Kalevi Tervo:  Thank you very much. 

[00:31:22] Lui Chih Wei: Thank you very much. 

[00:31:23] John Snyder: Thank you for joining us today on Setting Course, an ABS podcast. If you're interested in learning more about today's topic or listening to more episodes, visit www.eagle.org.