Professor Landsiedel, what is the aim at CAPTN Fjord 5G?

13. June 2024

CAPTN Fjord 5G is—just as the name suggests—focused on 5G mobile technology. But what else is this important project for autonomous shipping researching and developing? Prof. Dr. Olaf Landsiedel (Distributed Systems Research Group, Kiel University) explains.

CAPTN: What is the goal of CAPTN Fjord 5G?

Olaf Landsiedel: CAPTN Fjord 5G aims to operate an autonomous ferry via remote control or semi-autonomously across the Förde using 5G. That is the core objective. Additionally, there are some secondary goals: for example, we are also looking at how we can use 5G technology in other areas, such as port logistics or even sailing sports.

And what is your role in the project?

My role is, on the one hand, to lead the project from the university’s side. I am responsible for ensuring that the many participating research groups deliver what they are supposed to and need to deliver. Naturally, my research group is also actively involved in the project. Our main focus is ensuring the quality of data transmission remains stable. To transfer camera data from the ferry to the control center, we need a stable bandwidth or at least knowledge of how much bandwidth is available. It is our task to predict this bandwidth and respond accordingly if necessary.

What do you mean by bandwidth?

It’s similar to what happens at home when you stream movies or browse the internet. The ferry also needs a certain amount of bandwidth to upload or download data. If, at a given moment, we have fewer megabits or gigabits per second available, I may need to turn off a video source or send a video at a lower resolution to make optimal use of the limited bandwidth.

Doesn’t turning off a camera compromise safety?

No. The ferry has multiple cameras on board. If I am moving forward, for example, I absolutely need the rear-facing camera. But I can do without it for a few seconds. Alternatively, I can transmit the image in a lower resolution, using less bandwidth.

Besides camera data, what other types of data are transmitted?

Primarily camera data. Additionally, we can also transmit LiDAR data. This works like a type of radar signal, providing a 3D representation of the surroundings. Classic radar is also on board. Essentially, we transmit all the information that a captain would normally have on the bridge of a ship to the control center on land. This ensures that someone steering the ship remotely has all the necessary information to operate it safely and efficiently.

Is artificial intelligence (AI) a part of your work?

Yes and no. Our research group primarily focuses on intelligent data compression. Here’s an example: On the cameras, I can see the water and a few ships sailing. The dominant color is blue, which doesn’t necessarily need to be transmitted in full detail or high resolution. However, other elements in the image, such as docks or other ships, are crucial—I need to see them very clearly and understand how fast they are moving and in which direction. This is where artificial intelligence comes into play. The AI automatically determines which parts of the image are important. These are then transmitted with higher priority.

So, the project is only about data transmission and not about autonomous maneuvering?

In all our projects, we are generally working towards enabling autonomous maneuvers. However, this is not part of the CAPTN Fjord 5G project but rather belongs to CAPTN Fjord Areal. CAPTN 5G is primarily focused on optimal data transmission.

Which project partners are involved in CAPTN Fjord 5G?

Many research groups are participating in this project, primarily from computer science, but also from electrical engineering and business administration. The business administration group focuses mainly on port logistics. Other groups specializing in sensors and mobile communications are also involved. Additionally, local companies from Kiel, such as the internet and telecommunications provider Addix and the manufacturer of monitoring and control systems Anschütz, are part of the project. Vodafone is also on board.

The project has been running for over two years. What have the project partners achieved so far?

The goal is to develop a prototype for autonomous maneuvers—and we’ve made significant progress. This involves various considerations: What cameras does the ship need? What kind of intelligence is required on board? Which data is crucial? Then, the data needs to be transmitted to the control center. There, it must be received, displayed on a monitor, and a human operator should be able to remotely steer the ship using a joystick—similar to a video game.

Is it more challenging to collect such data in the Kiel Fjord compared to other waterways?

Navigating the Fjord is more challenging than sailing on the open ocean, where you are more or less alone. Our hypothesis is: If we can operate in the Kiel fjord, we can operate almost anywhere in the world. However, one advantage we have is mobile network coverage. In an urban area like our fjord, mobile coverage is naturally much better than in the open ocean, where there is no network coverage at all.

Is there a fallback option if the network suddenly fails? For example, is there Wi-Fi?

Yes, and the ferry is equipped with multiple mobile antennas. This means that if one provider’s network fails, the ferry automatically switches to another provider. This is crucial for safety. There is also Wi-Fi around the Kiel fjord, but that is being addressed in the CAPTN Förde Areal project.

You are also researching applications in sailing sports and port logistics. What exactly are the project partners doing there?

We are exploring several questions. One example from sailing sports: Today, coaches typically follow their sailing students in a motorboat, which is tiring and not particularly environmentally friendly. An alternative would be a live stream from the sailboat to a control center. The coach or coaching team could then provide immediate feedback and guidance, staying much closer to the action. At the same time, we are considering whether we could broadcast regattas using this method. If we can transmit data with very low latency and high efficiency, we could stream regattas live.

What exactly is latency?

When remotely controlling a ship, we primarily measure and monitor two key metrics. The first is bandwidth—the megabits per second available for transmitting large amounts of data. The second is latency, which refers to transmission delay. This is the time it takes for a signal to travel from the source, through the 5G network, to the control center. We need both high bandwidth and low latency, ideally in the millisecond range.

The funding for CAPTN Fjord 5G ends later this year. Have you achieved everything you set out to?

We are already where we aimed to be. The main goal—remote controlling the ferry—is working. Now, in the final months, we are focusing on implementing improvements and optimizations. However, our long-term goal remains to develop a fully autonomous ship. That means we don’t just want to remotely control the vessel; we want it to execute maneuvers on its own. This is a complex challenge because the system must reliably detect all objects in the water—from swimmers and kayakers to cruise ships and even submarines. Additionally, there are collision avoidance and docking maneuvers to consider. These are still major research topics. Another open question is regulatory approval, which remains completely unclear. So, we will be working on this for many more years.

Thank you for the interview!

(Translated with the assistance of ChatGPT)