Students win contests designing delivery drones that respond to real-time wind change

Body

Part of the charm of tiny Tangier Island, located smack dab in the middle of the Chesapeake Bay, is that in some ways it seems stuck in time, with residents speaking an outdated dialect and relying on crabbing for a living. What isn’t so endearing to residents about their simple life is the need to take a ferry to the Eastern Shore to obtain necessities, such prescription medications. Such a trip is an all-day affair, often causing people to miss work and lose wages.

But now Riverside Health System and DroneUp, a delivery service based in Virginia Beach, are working with a team of George Mason University students to deliver medications via a small drone. But even that poses challenges.  

“The route is 15 nautical miles (NM) from Onley, Virginia, to Tangier Island,” said Darius Jack, a student in the George Mason's College of Engineering and Computing’s Systems Engineering and Operations Research Department. “The route traverses both land and terrain resulting in significant wind shifts, including onshore and offshore winds, depending on the time of day.”

Depending on a drone’s design, it can be very sensitive to wind, requiring extensive analysis. Team member Grean Ramos said, “We developed an aerodynamic simulation of the drone and its batteries. The simulation showed that the 15 NM route is close to the drone's range limit. Any unexpected headwind could result in the drone running out of battery and needing to land before reaching Tangier Island.”

“There are weather stations at Onley and at Tangier Island airports, but no weather stations in-between," said team member Diana Machuca. "We analyzed one year of hourly wind data from the airports and found that headwinds would prevent the drone from reaching the clinic 32 percent of the time.”

Four college student contest winners hold a check
Team members Gurkirt Tahim, Darius Jack, Grean Ramon, and Diana Machuca

So the team developed a Drone Range Decision-support System (DRDS). The DRDS uses the aerodynamic simulation described above to continuously monitor the range-remaining as the drone flies. If the simulation estimates that the drone has experienced an unexpected wind and will no longer have enough battery power to reach the destination, it recommends the drone operator to return home (if it can) or to land safely in designated landing zones. Simulations showed winds would require the drone to return home 17 percent of the time.

The team’s plan won First Place in the STEM Track in the Patriot Pitch Competition, a annual contest sponsored by the Costello College of Business at George Mason, and First Place in the Systems and Analytics Track at the Sage Memorial Design Competition, a student-focused international forum for the design of complex systems.

“Working on this project was an amazing experience,” said Jack. “We had to use all the system engineering knowledge and skills we learned over the past three years. Building systems in the real-world is hard, but perseverance and creativity pay off.”