We’ve talked about aerial drones quite a bit on MAKE, but yesterday, Aug. 26, it was a nautical drone that made history.
Back in 2010, a group of college kids started working on an unmanned robotic motor boat that could cross the Atlantic from Rhode Island to Spain. Why? Because a robotic boat that motors around a pond or lake is cool, but one that can cross the ocean is an inspiration to everyone.
The crew consists of Dylan Rodriguez (engineering, Worcester Polytechnic Institute), Dan Flanigan (civil engineering, Bucknell University), Max Kramers (mechanical engineering, University of Rhode Island), Ken Muller (mechanical engineering, Worcester Polytechnic), Brendan Prior (liberal studies, Endicott College), Mike Flanigan (aerospace engineering, Notre Dame), and David Pimental (computer science, Northeastern).
They started construction on the final version of their boat, called Scout, in April 2012. It launched on Aug. 24 and the crew has already broken the record for distance traveled by an autonomous transatlantic water craft. Scout had traveled about 100 miles out to sea by yesterday evening. They beat a previous record of about 60 miles set by a team from Aberystwyth University in the 2010 Microtransat Challenge.
Initially the crew wanted the solar-powered vessel to begin its journey earlier in the summer, to take maximum advantage of long sunny days. They were ready to launch from a point on Rhode Island on June 29, and made their first attempt. Bad weather prevented Scout from being able to collect enough power during the day to make it out to sea. Rather than risk the craft being washed up on shore, the crew collected Scout to try again.
A second attempt was foiled due to a faulty rudder servo. If you don’t succeed, try, try, again. Scout was successfully launched on the third attempt in the wee hours of Saturday morning on Aug. 24.
The Scout Transatlantic project has a very nice feature that allows you to track Scout’s progress from your computer or mobile device. I’ve been stuck to the the automatically refreshing page like glue since I learned about the project this week.
The crew worked to design around every risk they could imagine. The hull was designed to be strong, but lightweight, reinforced with several sturdy bulkheads. They maximized the number of solar panels they could fit by removing the aluminum frames, and tilted the deck to face south for greater exposure to the sun. When power is low, Scout enters a low power drift mode to save batteries. An Arduino Mega microcontroller handles navigation, and a second Mega handles sensors and communications through an Iridium satellite transceiver. The software is reset every 12 hours to avoid pesky memory leaks or corruption from long term use. A weighted keel will right the boat if it should capsize. The boat is as water tight as they could make it, but it is also equipped with bilge pumps.
The route was planned to cross shipping lanes at right angles, minimizing chances of a collision. They planned to take Scout south into the Gulf Stream and cross the Atlantic as quickly as possible. The longer the voyage takes, the more chances there are that something will go wrong.
Although the crew worked hard to test and re-test the system, Murphy did manage to throw in one surprise. Scout included code to avoid having to backtrack to a waypoint that it had floated past while in drift mode. Instead the code caused Scout to bypass most of its planned waypoints. The crew has analyzed the error and understand what happened. They believe that Scout is now navigating to waypoint 44, near the coast of Spain. Once Scout reaches that waypoint, it should continue normally through the last sixteen or so waypoints to the shores of Sanlucar de Barrameda, Spain, where Columbus began his second journey to the New World.
This project didn’t get made as research for some corporation or government. It wasn’t made to win a contest, or for a school project. These guys built Scout because they wanted to, and they haven’t let anything stop them. Early on, one of the schools denied their request for lab space for the project. Due to legal concerns, they said. The crew was undeterred. They did their own research and consulted with a family friend with experience in marine law and decided to proceed.
Scout ended up costing about $6,000 to build. That’s not a lot of money as this sort of project goes, but quite a bit to come up with if you don’t have help. The crew has been very happy with their major sponsor, Jamestown Distributors, a family-run marine and building supply business. There are also many enthusiastic individual contributors who have donated money or just cheered for the crew on their Facebook page.
Through many, many hours of work, they built a boat that has so far succeeded where others have failed. Despite many technical challenges, the team persevered. They even had a minor run in with local police while testing Scout at night. Several teens hanging out on a dock at night may have looked suspicious, and the police were not inclined to believe their story about a robotic boat floating nearly invisibly out in the water. Would you? Luckily, they were able to show the police Scout’s blue marker light moving around and explained that the unmanned boat would come back to dock by itself in fifteen minutes. The police didn’t bother to wait around and left them to it.
Now that Scout is off to sea, the team actually has little to do but watch and wait. There is no way to give Scout commands, so it’s just Scout’s programming, systems, and hull against the ocean. Scout’s journey may take about three months, but the team will be able to make a much better prediction as time passes and more data is collected.
They are feeling confident that Scout will reach its destination, but even if it doesn’t, they’ve already succeeded. If Scout fails, they’ll build a new boat and make it better using what they learned the first time. In the meantime, they’ll keep watching http://www.gotransat.com/tracking/ with the rest of us.