From Villanova Magazine, Winter 2015 issue
A team of eight Villanova engineers arrived in Marina Bay, Singapore, during the University's fall semester. Invigorated by nerves, excitement and nearly a year’s worth of intense preparation, they competed in one of the highest-level autonomous marine system challenges ever posed to university researchers. Working alongside partners from Florida Atlantic University (FAU), they had three days to test and debug their 16-foot Wave Adaptive Modular Vessel (WAM-V®) before the qualifying rounds of the inaugural Maritime RobotX Challenge. Each team member would have to work as many as 90 hours during the week of the competition, feverishly fine-tuning the sensors, controls and software algorithms that would enable the boat to complete five complex tasks without human guidance or remote control.
The elite field of RobotX competitors included two other teams from the United States—a joint team from MIT and Olin College, and a team from Embry-Riddle Aeronautical University—and three teams each from Australia, Japan, Singapore and South Korea. A grand prize of $100,000 was on the line, but the financial prize mattered much less to those on the Villanova team than did their pride in their collaborative innovations and their contagious excitement about sharing their knowledge with their international counterparts.
ASSEMBLING THE TEAM
At the helm of the RobotX project was C. “Nat” Nataraj, PhD, professor and chair of the Department of Mechanical Engineering, and the Mr. and Mrs. Robert F. Moritz, Sr. Endowed Chair in Engineered Systems, who heeded the call for proposals from the Office of Naval Research (ONR) when ONR and the Association for Unmanned Vehicle Systems International (AUVSI) announced the competition in 2013. After eight years’ participation in the AUVSI’s RoboBoat Competition, Dr. Nataraj knew RobotX would be a unique opportunity to elevate his students’ real-world engineering experience.
Dr. Nataraj also identified the ideal partners in FAU and Karl von Ellenrieder, PhD, associate professor in FAU’s Department of Ocean and Mechanical Engineering. At the 2012 RoboBoat competition, Dr. Nataraj recognized a complementarity between FAU’s and Villanova’s strengths. With on-ocean facilities, FAU would be able to store the team’s WAM-V, test it in trials of the on-water RobotX tasks, and contribute to its systems engineering and low-level control. Meanwhile, the autonomy experts at Villanova could manage high-level control, vision-based navigation, localization and mapping.
To assemble the Villanova dream team, Dr. Nataraj and his colleague in Mechanical Engineering, Associate Professor Garrett M. Clayton, PhD, called first on graduate students Anderson Lebbad ’12 COE, ’14 MS and J. “Wes” Anderson III ’13 COE, ’15 MS.As the team’s anchors, they would devote the most student lab time to the development of the boat’s systems. But the RobotX team could not consist solely of mechanical engineers. “The specific challenges of RobotX required that we include students specializing in computer science, electrical engineering and mechanical engineering who could all work together,” says Dr. Nataraj.
Dr. Nataraj completed his lineup with Chidananda Matada Shivananda ’14 MS, Edward Zhu ’15 COE, Dylan DeGaetano ’15 COE, Priya Shah ’16 COE, Gin Siu Cheng ’15 COE, Michael Benson ’15 COE and Gus Jenkins ’16 COE.
INNOVATION AT SEA
The Maritime RobotX judging rubric laid out clear standards for a high-scoring WAM-V® mission. First, the boat had to successfully navigate the course autonomously. Second, it had to be able to locate an active pinger to a high degree of accuracy and report the color of the buoy in that quadrant. Third, the vessel would receive points just for docking and substantially more points for landing in the correct dock. Fourth, the boat had to report a light sequence as broadcast from a buoy. Finally, it had to detect and avoid obstacles, with point detractions for each collision.
All of these tasks depended on the Villanova team’s development of a new vision system for the boat. In past competitions such as RoboBoat, reports Lebbad, “we were using heuristic methods of color identification. To the question ‘Is the buoy red?’ the vision system would basically decide yes or no.” But if something in the environment changed—if it got brighter or darker, or if it was raining or the boat was moving—then the analysis was thrown off. The solution the RobotX team devised was to use a probabilistic approach to determining color. Lebbad explains: “It was a lot more scientific. Now, we knew the colors that it could be, and the system weighed the possibilities over time and multiple samples.”
A further challenge was getting all the different systems—the high-level planner, the acoustic system and the navigation system—to “play together.” Humans, Dr. Nataraj explains, have no problem processing two or more kinds of sensory input at the same time to make judgments about their environments. Creatinga similar fusion of visual, locational and other information for an autonomous boat is extremely challenging.
BEYOND THE SCIENCE
When asked about the most exciting aspect of the project, the students agreed that just the opportunity to travel to Singapore was a highlight. But they also concurred that the learning experience was beyond compare. For Ed, the multidisciplinary nature of the team was a standout feature of the year. “It was fascinating to see how people from different disciplines brought different skills and strategies to the table. We couldn’t have accomplished nearly as much work without that collaboration.”
Lebbad stresses that, ultimately, “it was a lot of fun. It was an awesome way to get hands-on with your education. When you’re learning in the classroom, there’s rarely an opportunity to demonstrate this concretely what you’ve learned.”
Projects like RobotX are not cheap. In fact, the team benefited from about $340,000 in funding from such contributors as the Naval Engineering Education Center, the Naval Sea Systems Command, the Air Products Foundation, United Technologies, Speedgoat and VideoRay. “We’ll definitely plan to apply for the second RobotX competition two years from now, and we’ll need resources,” Anderson says. His team laughs as he floats another idea: “Or, you know, every Friday we have a meeting with pizza.”