Not Rocket Science?



Visibility into the maritime domain really is based on rocket science, but new kinds of entrants and the move to common web technologies is unlocking a new world of uses for maritime data, explains Nick Allain of Spire Global.

You would be forgiven for not immediately associating aerospace with ship traffic. Fishermen leaving the Port of Chimbote in Peru for a week long journey are unlikely to have satellites uppermost in their minds when they climb onboard, and an oil tanker in the Gulf of Mexico, despite its proximity to NASA, isn’t wondering whether astronauts have found the time to deploy the next generation of ship tracking instruments into space.

So it may be a surprise to some that a revolution in nanosatellites—tiny wine-bottle sized spacecraft—is beginning to revolutionize the way ships are tracked. Deployed into Low Earth Orbit in batches of about four or eight at a time, they are now collecting millions of data points each day about ships in the ocean. These satellites make tracking in the most difficult places, like the Arctic or middle of the ocean, much easier.

In 2012, I too had very little understanding of aerospace—let alone ship tracking. That’s when I was connected with a small company starting up in Silicon Valley and before long, I was knee-deep in satellites. Just five years later, Spire has 48 ship tracking satellites and a team dedicated to the collection and production of knowledge about ships in the Earth’s oceans.

At the core of what we do though, is the customer and key to building customer propositions is taking billions of data points and translating them into something that people can use. That role falls to Product Manager Kyle Brazil, who describes himself as someone who, “sits at the intersection of business, engineering, and design. Our primary responsibility is to the customer, making sure that we fully understand their problems then building and delivering solutions to solve them,” he explains.

“In the maritime world, that involves a lot of time talking to people in a range of industries who spend their days steeped in data, trying to produce actionable information from it. My goal is to create products that make their lives easier. Once new products are out in the wild we then measure their success based on how well they do at solving the problems we intended them to. If they fall short, we adapt our approach for the next iteration. If they’re successful, then we move on to the next big problem.”

The move towards common web technologies like REST APIs, and away from specialised geospatial data standards that only a handful of GIS experts know how to use, is an important trend.

Taking advantage of these new products requires a minor but important shift in how some users access ship data and interact with it. Rather than being handed a never-ending stream of information to process, they use a question and answer based approach that leaves the sifting up to the service provider.

That new API approach, standing for Application Programming Interface, unlocks a new world of uses for maritime data. “The move towards common web technologies like REST APIs and away from specialized geospatial data standards that only a handful of GIS experts know how to use is an important trend,” Kyle believes. “This massively lowers the barriers to entry for many more customers in many more industries. When developers don’t have to go out and learn a whole new standard just to work with a new dataset it opens up the potential for them to integrate new services into their applications and analytics toolchains.”

If that sounds very ‘Silicon Valley’, that’s because the same trends gripping the technology industry are set to affect how the maritime domain is understood. Silicon Valley’s self-driving cars and robots may seem to be leading the way with artificial intelligence (AI) but it’s about much more than just things with four wheels or robotic arms. Companies are hard at work on autonomous ships that can navigate on their own and vessel route prediction that knows what a Master will do hours before he does it. It requires an incredible amount of data, which happens to be a strong suit of Spire’s satellites.

“There aren’t a whole lot of companies who collect and process this volume of data in a constant stream, 24/7” confirms Kyle, “I call it the Twitter of ships. There’s a lot of noise so we spend a lot of time thinking about how to best make signal out of that using machine learning, artificial intelligence, and careful filtering. There are a lot of exciting developments in the AIS world—like VDES—but the real exciting stuff, to me, is being able to ride the wave and take advantage of the massive advances in computing power and machine learning. Global-scale geospatial data has always been difficult to work with because it’s impossible for a human, or even a large of team of them, to be looking at all points on Earth around the clock. We are now at the point where we can build intelligent passive monitoring systems that watch for events, detect patterns, and only notify humans when they need to take action. This will only accelerate in the future.”

Despite all the advances and excitement around satellites, machine learning, and new APIs Kyle is quick to point out that no matter how advanced the technology, the underlying importance of the shipping industry shouldn’t be underestimated as it’s critical to our daily lives. “One of our biggest challenges as we move into new markets is showing people how the movement of ships impacts their everyday life,” says Kyle.

“From the food you eat to the iPhone you use to the geopolitical environment, you want decision makers to have the most robust data and insights possible. This data can save lives through protecting our food supply, helping prevent human trafficking, assisting in the search for lost vessels, and more. It’s good to never lose sight of that.”


Images courtesy © Getty Images

This article appeared in the Q4 2017 issue of Futurenautics.


free of charge as a PDF

Author: admin

Share This Post On