Headed by Melissa Schiele, a PhD researcher in Loughborough University’s School of Mechanical, Electrical, and Manufacturing Engineering, the aim of the project is to develop simple methods using drones to gather images of ocean and beach plastics. Schiele said these will be used to build a baseline database of aggregation, deposition rates and classification.
The project is being undertaken in collaboration with Oceans Unmanned, the Marine Research and High Education Center (MaRHE – University of Milano-Bicocca), and supported by The Ritz-Carlton Maldives, Fari Islands.
The Maldives project follows similar work undertaken by Schiele in Belize that explored the use of a novel fixed-wing UAV as an enforcement and monitoring tool in the Turneffe Marine Atoll.
Common to both projects is the Aeromapper Talon fixed-wing UAV. The system being used in the Maldives was donated by the non-profit organization Oceans Unmanned and features a Sony RX0 camera controlled via ArduPilot’s Mission Planner software.
“Although this system is ultimately the same as the one in Belize - though it no longer has a live-link nose camera - the system I am developing for my PhD independently addresses key requirements from the system, for example, flying beyond 10km range,” said Schiele.
Schiele explained that the drone’s flight altitude will depend on how much area needs to be covered and the GSD (Ground Sampling Distance) required.
“We flew the Talon at 80m and 100m altitude and were impressed with the orthomosaics produced in Agisoft Metashape [photogrammetry software] of the seagrass beds in Magoodhoo island,” she said. “The drone flies at around 55kph which is a little nippier than our Mavic and Phantom [drones] we were also flying.”
She added: “We also uploaded images to DroneDeploy and look forward to analysing and comparing the two outputs in more detail.
“Currently we are working on methods for analysis and will be inviting local academic stakeholders to join our research group officially. The project stakeholder map is growing large, very quickly, and building capacity in situ is a core objective for us.”
The Maldives is home to the world’s seventh-largest coral reef system and more than 1,100 species of fish and 180 species of coral. Discarded or lost fishing nets – ghost nets – are a problem in the region, but this could be alleviated with a machine learning algorithm developed by Dr Sol Milne, former resident naturalist at the Ritz-Carlton, to detect the nets from drone images.
“He has net data from around the world, and he and I worked to get net data from around the resort in the Maldives,” said Schiele. “It's an important part of our work in the Maldives as nets are everywhere and spotting them is difficult until they are upon you.
“Kat Mason, another former [Ritz Carlton] naturalist, intercepted an Olive Ridley turtle in a ghost net on a routine drone flight - a real proof of activity concept if I ever saw one! The turtle was rescued.”
More information on the partnership can be found on the Oceans Unmanned webpage, and more information on Schiele’s research at Loughborough University can be found in her interview with VOLUME magazine.