By Jordi Neves, Chief Digital Officer, Volvo Ocean Race
The Volvo Ocean Race is not only a huge event in the sailing world but one of the most extreme adventures out there. The competing boats circle the globe in often brutal conditions, with crews working around the clock for 28 days at a time in some of the most remote and inhospitable waters anywhere. For example, the vessels sail as far south as possible—beyond normal shipping lanes, where icebergs are a constant hazard—in an effort to catch the strongest winds.
For the 2017-18 race, they’ll set off from Alicante, Spain in October and, after stops in ten other host cities on six continents, cross the finish line in The Hague, Netherlands eight months and 45,000 nautical miles later.
But the challenges don’t only test human endurance. The event also pushes the boundaries of technology. And digital technology is crucial to the success of the race. It not only allows us to support teams operating in these perilous conditions but also enables the teams to share their stories in new ways with various audiences—from sailing enthusiasts to classrooms who use the race as a learning tool.
A Hostile Technology Environment
For the 2010 race, we took all the boats online, and it was one of the best decisions the race organizers ever made. The boats were equipped with integrated microphones and cameras, creating a veritable media machine on the seam capable of pumping information and images back to the race control center in Alicante at all hours.
For the upcoming race, each boat will be equipped with seven fixed cameras, two wireless cameras, a 360-degree “virtual reality” camera, and several microphones. Along with the sensors recording the boats’ performance, these will generate a telemetry packet with more than 150 variables coming from each boat every ten seconds.
We’ll be downloading roughly 100 terabytes of video and other content from the boats when they are at sea and as much as 1.5 petabytes when they’re in port. And that’s just the data from the boats; our models and other content generate another 12 to 14 petabytes average per leg.
But this cascade of data brings with it significant technical challenges. For one thing, the race takes place in an incredibly harsh technology environment. It’s not just extremes of temperature, humidity, and water pressure, but also G-forces, salt, and vibrations—when these boats are catching the wind, the vibrations on board are incredible. We work with world-class partners to make sure both the on-board hardware and the satellite network are as robust and resilient as possible.
Another unique requirement is the fact that the race’s rules prohibit crew members from having any direct connectivity with the outside world. That’s because a well-funded team could to use a robust computer model to guide the crew through the course.
So we have what is essentially a black box—a fiber network—between each boat and Race Control. Race officials monitor all content going to or from the boat to make sure that no one has access to information they shouldn’t have.
Beyond that, we’ve required that all boats carry the same amount of computer processing hardware. We don’t want crews to be building their own big data centers. We want them to have the same tools in making their navigational choices.
The idea of the race is to challenge the crews’ skills, decision-making, and risk taking. We give every team the same weather information as well as updates every six hours on how everyone is doing, but nothing more.
With this year’s race, we’re taking our digital narratives to the next level. It’s not just about the technology we have on board—although we are bringing on a host of cutting edge tools. We are trying to get more value of out the human aspect of story telling as well.
Besides our battery of on-board fixed and wireless cameras, we are continuing to use on-board reporters—we’ve expanded the team so that no one burns out—to capture and write the stories of the race.
In addition, each boat will also have two drones that can follow the boat, based on the GPS position. The vessels are in the middle of the ocean 90 percent of the time, far beyond the range of helicopters, and this will enable us for the first time to capture footage from the air—if we can manage the delicate task of landing the drones on a boat travelling at an average speed of 27 knots.
Our dream is to beam out live pictures from the middle of the Southern Ocean for the very first time—from drone to boat to satellite to Race Control and out to the world.
We will also be working with the crews to share more of their own stories with audiences directly, perhaps having the skipper live tweeting or some of the younger crew posting to Snapchat. We found a way to create a gateway so that crews can start a conversation with the audience. The aim is to allow our audience to experience what these individuals go through on board as they live it.
The Value of Self-Disruption
Much like our Volvo Ocean Race teams, we at Race Control can never allow any slack in our efforts during the race. We are always experimenting. And that's the beauty of the race. This isn’t a 90-minute football match. It’s all day, every day, for eight months. So we innovate not just leading up to the event, but during it.
As an agile development shop, we’ll sometimes send software updates to the boats when they’re at sea, making adjustments based on the information the boats are sending us. Sometimes, we lose communication with the boat, but we allow space for those mistakes. The only way to innovate is to take some risks.
We have also begun using machine learning capabilities to help us analyze big data—not just the telemetry data that can tell us about the performance of the boat, but also data about the content we produce. For example, we can identify people following the race online who have similar interests. During the last competition we found through analysis of social media that school teachers from around the world were using the content we produce to teach geography in their classrooms. This allowed us to create content packages specifically designed for this community of school users, and even arrange for students to have a live Q&A session with crew members at the end of the race.
We also see great potential in applying machine learning capabilities to better understand the performance of the humans on board the boats. Looking at biometric data can help us to understand the impact of sleep—or lack of it—on operations. Ultimately, in conjunction with academic partners, we may be able to come up with some models about, for example, the best crew rotations to optimize performance.
Planning for the Unplanned
We test everything we can before the race—the boats, the media, the technology, the reporters. But some approaches we have to reinvent as the race progresses. One big lesson I’ve learned is the importance of a Plan D, because the conditions of this race often make plans A, B, and C obsolete.
With the extreme temperatures and vibrations, just maintaining connectivity is difficult. But the biggest challenge is responding to the things you might never see coming. The only way to do that is to maintain extreme flexibility and to work with partners who are equally agile and adaptable.
During the last edition of the race, Team Vestas Wind crashed into a reef in the middle of the Indian Ocean, three days away from Mauritius, the nearest land. Using radio and GPS coordinates, they were able to report their positions to Race Control. It was the middle of the night in shark-infested waters. The cameras recorded the crash, but ultimately gave out as the electrical system was overrun with seawater. There was no way to send a plane, no way to send a helicopter. Eight hours after the crash, as the ship began to fall apart, the crew made their way through the dark waters to a life boat they’d earlier tied to a rock, then made their way to a dry part of the reef, where they were picked up by a fishing boat several days later. Thankfully, everyone survived.
As a result of climate change, the boats today are encountering more issues with ice. Ten years ago, we could tell them with some accuracy what latitude not to cross. Today, there’s much less certainty. We use our cloud back-end to help us look at the data in real-time for problems like ice or severe storms. We create models that enable us to proactively take care of the fleet. What’s more we share the data we’re collecting with organizations around the world trying to better understand the impact of climate change on our oceans and the weather.
What we do is incredibly exciting, and as we’ve seen, it can have real value even beyond the event itself. And I know that during the 2017-18 race we will come up against some obstacles we could not have foreseen but we will also uncover some innovations that otherwise would not have been possible.
HCL is the official IT services partner for the Volvo Ocean Race.