Many annual student competitions purposefully make the requirements not just different, but tougher each year keeping success just beyond reach. With this strategy, organizers hope to both develop top-rate engineers and move the technology forward step by step.
"It's hard enough to develop a land-based autonomous vehicle but the underwater environment adds extra challenges."
The Autonomous Underwater Vehicle (AUV) Competition follows this strategy. One team completed the course in 2009, so the organizers have promised it will be more difficult this year, its 13th competition, according to Micah Boswell (CPE '11), leader of Virginia Tech's team.
From left: Joe Ball, Ben Guzzardi, Richard Stroop, and Bryant Ferguson.
The competition, sponsored by the Association for Unmanned Vehicle Systems International (AUVSI) and the Office of Naval Research (ONR), focuses on realistic missions underwater.
With vehicles no longer than 6 feet and less than 3 feet high or wide, teams seek to complete a number of underwater tasks. The AUVs must drive through a validation gate, follow an orange path, touch a red buoy, jump over bars, drop markers in bins, fire a torpedo, grab a PVC structure and surface in a specified octagon all in 15 minutes or less.
It's hard enough to develop a land-based autonomous vehicle to complete such locomotion with visual and acoustic recognition tasks, but the underwater environment adds extra challenges, Boswell said. "ECEs generally don't have to operate underwater."
The electronics must be small and dense so the system is the same weight as water, which is surprisingly heavy, he noted. The team will add weights, if needed, so the AUV can submerge. Powering the system underwater is also tricky, he said. "We decided to use lithium polymer batteries, but they don't react well when combined with water. We need numerous failsafes so that if the system shorts, the battery is protected and doesn't blow up the vehicle."
During the July competition in San Diego, the teams will not operate their own AUVs, but will put them in the hands of Navy divers. This means the operating system must be user-friendly. With that in mind, the ECE students make sure the other students can understand and operate the vehicle and are continuously improving the software so that all of the systems can interact well through a simple interface.
This is the second year Boswell has participated. In 2009, Virginia Tech floated its first AUV team in several years. "Last year none of us had any idea what to expect," he said. "Just the fact that we were able to get an AUV put together and score points I think it's pretty amazing that we were able to do that given that we started with no idea of anything."
Last year's team was six strong, with three CPE students. This year, with many returning members, a much larger team, and sponsorship from Lockheed Martin, the team has dreams of greater success.
Some of the challenges the students face stem from the team's growth. This year's team has 22 members, with eight ECEs. The team is advised by ECE's Dan Stilwell and Craig Woolsey of aerospace and ocean engineering.
With the larger team, the students have been able to rebuild, modify and add systems for higher value tasks at the competition. One new system is passive sonar. The team had hoped to buy a functional system off the shelf, but balked at the almost-$20,000 price. Instead, the team identified a specialized sonar group that is working with Stilwell's laboratory to develop a cost-effective system.
Once a subsystem works independently, however, it must be integrated into the whole system, which can be a challenge. "We're dealing with more complexity integrating the work of a much larger group," explained Boswell.
The software is critical to the system integration. Mathworks, a sponsor of the Ware lab that houses the team, has provided software, including Simulink, which the students use for autonomous-level code. "We integrate that with our own C++ code," Boswell said. "Simulink gives us a graphical programming interface, which makes it easier for those who are not strong programmers to look at the code and understand how it works," he added. "This helped us develop the software very rapidly." Last year, the team hadn't started programming until about three weeks before the competition, he said.
Although the larger team helps with technology development, it makes it more difficult to ensure strong team interactions. Boswell described a recent underwater test. "It was the first time we had several different groups at the same event, and it was an interesting experience to have groups that have been working on the same project for weeks meet each other for the first time. As the team has grown, it has become harder and harder for everyone to know each other, and a challenging task for the overall coordinator (me) to keep track of everyone's work," he said. "I believe this experience is invaluable for anyone working on a large team, and should be an integral part of a good engineering education."