- Dutch startup Captain AI Rotterdam is engineering self-driving boats and ships with the deep learning algorithms and a computer simulator.
- Captain AI successfully demonstrated a self-driving version of one of Rotterdam’s popular water taxis in late 2017.
- The 57-foot vessel has surveillance cameras mounted on the bridge with a digital GPS system to provide accurate location readings than standard GPS.
- There are specialized sensors for monitoring and controlling the engine and steering. All of the data about the ship’s operations is fed to computers that are monitored on board and on land.
According to NBC news, Rotterdam is working with tech firms to develop robotic vessels that promise greater safety, reduced traffic and a fix for labour shortages.
Is the shipping industry ready?
As the world gears up for driverless cars, the shipping industry is working to adopt similar technology — with the hope that replacing traditional crewed vessels with autonomous ones might help prevent collisions at sea, smooth the flow of traffic in congested ports and ease labour shortages.
Here at Europe’s largest port, the Dutch startup Captain AI is at the forefront of this transition. Using a high-tech port patrol ship, deep learning algorithms and a computer simulator originally designed to train captains, the company is engineering self-driving boats and ships.
Vincent Wegner, CEO of Captain AI, said of the development of autonomous vessels: “It’s like a puzzle with all the pieces already there. All the hardware and software, it’s all out there, it’s all being developed and getting cheaper, open sourced, so we can really reuse it and really leapfrog this whole development.”
Captain AI successfully demonstrated a self-driving version of one of Rotterdam’s popular water taxis in late 2017. This summer, the company will send the Floating Lab Rotterdam, a port authority ship, out to sea without a skipper at the helm.
From the outside, the 57-foot vessel looks unremarkable except for the surveillance cameras mounted on the bridge. Inside, there’s a digital GPS system, which provides more accurate location readings than standard GPS, along with sensors monitoring and controlling the engine and steering. All of the data about the ship’s operations is fed to computers that are monitored on board and on land.
The port authority decided to repurpose the vessel as a test bed for universities and startups, like Captain AI, to help accelerate the transition to automation and to explore other ways to optimize the global shipping industry while minimizing its carbon footprint.
“We feel responsible for not just creating economic value, but also societal value,” Marjolein Boer, the port’s innovation manager, said. The port has been a global leader in many capacities, such as creating partnerships aimed at reducing emissions, she added, and it sees the opportunity to work with startups to set new standards for industry innovations.
The Floating Lab’s maiden autonomous voyage will be one of a series of tests that will increase in complexity, Wegener said. There will have to be someone on board to adhere to local maritime laws, but the ship will drive itself a few hundred feet at first and then longer distances within the port.
Captain AI isn’t the only company in the race to launch self-driving vessels. In Norway, the world’s first fully autonomous electric container ship, the Yara Birkeland, is slated to begin operations next year while the Norwegian company Massterly aims to offer a range of autonomous shipping products from control systems to vessel design.
What Captain AI has set out to create is a system that can be adapted to any type of vessel, big or small. The maritime simulator it uses, built by the Dutch developer VSTEP, includes specifications for port locations worldwide, as well as for vessels ranging from tugs to tankers to naval ships.
“We already have a whole catalogue of ships in there that are realistic; how they operate at sea is pretty realistic, so we can scale our model,” Wegener said.
The wide applicability of the technology means all it would take for a real-world ship or boat to become captainless would be installing the necessary hardware.
Upgrading a vessel to be fully autonomous costs roughly $20,000 to $35,000, Wegener said. It’s a relatively small investment if it eliminates the need for a captain or crew.
Such talk raises the prospect of lost jobs, but Wegener said new jobs would emerge. Shipping operators will need technical specialists for both hardware and software, and ports will need on-shore captains to monitor traffic and coordinate voyages.
Tankers and other large ships making cross-ocean hauls will still need crews to maintain engines, he added. But smaller, short-haul vessels could become fully autonomous and living quarters could be redesigned for more cargo storage.
Captainless vessels to be a reality
In the short term, Captain AI envisions automation helping small vessels like Rotterdam’s water taxis operate in foggy conditions. Vessels used for transferring waste or other materials could also become fully automated.
Captainless oceangoing ships could be a reality within the next two years, according to Wegener, but international regulations that require seafarers on board are likely to slow the adoption globally. If Captain AI’s model proves effective and local laws are quick to adapt, Wegener said Rotterdam could be the first site for unmanned vessels to become the norm.
Did you subscribe to our daily newsletter?
It’s Free! Click here to Subscribe!