Sea Machines Autonomous and Remote-Control Systems Enhance Safety

For most recreational mariners, circumnavigating Denmark would be a once-in-a-lifetime adventure. But when modern tug Nellie Bly safely completed that 1,000-mile passage in October 2021, the vessel’s commanding officers celebrated their success in Boston.

They didn’t fly to Beantown; they were already there.

Nellie Bly was equipped with a Sea Machines SM300 autonomous command-and-control system for a mission that the Boston-based technology firm dubbed “The Machine Odyssey.” While at least two professional mariners rode on board for all 129 operational hours of the circumnavigation, the SM300 autonomously piloted the vessel for 95-plus percent of the trip, including 14 port calls and 30-plus unscripted collision-avoidance maneuvers.

Despite not completing the journey themselves, Nellie Bly’s commanding officers did accomplish the longest autonomous voyage (to date) by a commercial vessel.

In the Beginning

Years in a leadership position at a marine-salvage company taught Michael Gordon Johnson about the kinds of accidents that can happen at sea. In 2017, he founded Sea Machines with the goal of making the marine environment safer. His vision was (and remains) potentially game-changing: to bring 21st-century autonomous and remote-control technologies to the marine space, where computers tackle navigational tasks and human operators provide oversight.

Sea Machines’ first commercial products—the SM200 and SM300—debuted in 2019. The SM200 wireless, remote-helm-control system allows users full bridge control from a little over 0.5 nautical miles away. Hardware-wise, it consists of a networked onboard cabinet (black-box computer) that interfaces with onboard systems and instruments, and a marinized remote user interface that comes fitted in a wearable belt pack.

The SM200 leverages industrial-grade components, including a Siemens-based programmable logic controller (PLC) for controlling a vessel’s steering, power and payload (say, a fireboat’s water pumps). Humans control the interface, which relays these commands to the onboard cabinet computer via dynamic frequency-hopping radios.

While the SM200 allows human operators to safely maneuver vessels around (or through) hazards from a safe vantage point or operating distance, the SM300 delivers range and autonomous mission-planning capabilities.

The SM300 also includes an onboard cabinet-style computer that networks to onboard controls, instruments, sensors, and systems, plus a ruggedized Windows-based laptop that serves as the remote user interface.

“It’s called the 300 because there are three main processors,” Johnson explains. “There’s the low-level control, which is PLC-based; there’s the autonomy computer, which makes decisions on path planning and collision avoidance; and the third processor is the user interface.”

The SM300 leverages the vessel’s own navigation and collision-avoidance systems for its eyes and ears. “We have a number of protections in our system to keep the vessel and the people on board safe” in a variety of scenarios from collision avoidance to low depth, Johnson says. “Our system is based around electronic charts, so the users can upload the charts for the location they’re operating in, and we bring in all the elements of that chart into our path planning and into our system.”

While Sea Machines directly sells the SM200 and SM300 (prices range from $50,000 to $100,000) to professional mariners and commercial-marine customers, the company is partnering with third-party entities, such as engine manufacturers (e.g., HamiltonJet and Rolls Royce), to sell its second-generation offerings as OEM-level equipment for a broader market, including recreational.

The first of these products, the Linux-based SM360, debuted this year when HamiltonJet unveiled its JETsense system. Johnson compared the SM360 to an ultrasmart autopilot (think Tesla). “It’s pilot assist,” he says. “It’s controlling steering and propulsion, and perceiving the domain and doing collision avoidance as necessary.”

One important way that the SM360 perceives its environment is via Sea Machines’ proprietary AI-ris, shorthand for artificial intelligence recognition and identification system. AI-ris employs networked cameras to detect, track, geolocate, and classify objects it “sees” into four main categories (e.g., sailboat or marine mammal). The SM360 then uses this information, plus data from other networked instrumentation and the vessel’s electronic chart system, to safely navigate.

Like its forebears, the SM360 employs a “human-on-the-loop” scheme, with three different modes of operation. These range from alerts that trigger humans to react (for example, if the system detects an object in the water) to full autonomous route planning.

So, if you’re pondering your next boating journey, consider what it would be like to be assisted by technology that never tires, loses concentration, or bogs down when confronted by confined waters and multiple targets.

Odds are good that joining the 99th percentile never sounded so rewarding.