(Above: UC Merced graduate student Brendan Smith goes knee deep to demonstrate the capabilities of a drone he created.)
by Gordy Slack
Drones may have a sinister reputation, but they are finding a useful niche in academic research in the service of environmental science, resource management, and precision agriculture. Researchers at UC Merced’s new Mechatronics, Embedded Systems and Automation (MESA) Lab explore innovative ways to use Unmanned Arial Vehicles (UAVs) to gather data from places too expensive, rugged, or remote to reach by conventional means.
Small, inexpensive UAVs can give ecologists access to remote areas where collecting samples or data has been difficult or impossible. The “data drones” that Professor and MESA Lab director YangQuan Chen has developed for the past five years are small, but their list of capabilities is long. They can collect high-resolution thermal, near infrared (NIR), and red-green-blue (RGB) standard video and imagery; identify plant species along with their nutrient and hydration levels (key measurements for monitoring both forests and crops); collect air quality samples; track pipelines and shale fracking sites for gas leaks; trace birds and other wide-ranging animals fitted with tracking devices; and even monitor biodiversity.
Monitoring aquatic biodiversity is the focus of a new collaboration between Chen and UC Davis’s Michael Miller. The team plans to use Chen’s drones to conduct surveys of mountain rivers and other hard-to-reach aquatic ecosystems. Chen’s part of the job is to develop a UAV that can land on moving water (even in high winds and foul weather) and retrieve a small sample. Miller is honing a procedure that extracts and sorts DNA to identify all the species living in the body of water from which the sample is taken. The method, which separates the environmental DNA (eDNA) from the water, requires far smaller sample sizes than older methods. An Assistant Professor of Population, Quantitative Genetics, and Genomics at UC Davis, Miller believes he can reduce the required volume of sample to less three tablespoons.
Chen is developing a prototype of the aircraft, which is two feet long and weighs about three pounds. It has four rotors, and the entire frame is a pontoon that enables it to land and take off directly from the water’s surface.
Another planned use of the water-drone would be to conduct biological surveys during and after disasters, such as floods, chemical spills, or fires. The vehicles will need to be able to fly, land, and maneuver on moving water, and take off from that water, all in wet and windy weather. These control challenges, says Chen, are at the center of his project.
“You do not need to be able to hold the aircraft in exactly the same spot, but you do need to have active maneuvering of the aircraft while it is on the water,” says Chen. If you land on water and surrender to the river flow, the drone could be carried into brush or hit a rock.
In addition to exploring the use of drones for environmental monitoring, MESA Lab researchers also focus on drones that can gather data for farmers. Precision agriculture, a hyper-responsive and efficient farm-management approach, tailors watering and fertilization regimens to real-time crop data that can be inexpensively gathered from above.
“In the next 20 years, I believe the San Joaquin Valley and Central Valley will become the ‘Data Drone Valley,’” says Chen. “By 2020, Merced should have a world-class drone center to provide agriculture with the aircraft and analysis that the new field needs.”
As water for agriculture becomes less abundant in California, and as the world’s food demands grow, farmers will need to extract higher yields from fewer acre-feet of water. One important way to do that, says Chen, is to closely monitor real-time crop conditions in all parts of a farm. Drones may be the least expensive and most versatile way to accomplish that. Outfitted with cameras and sensors that can give high-resolution readings of soil moisture, plant evapotranspiration rates, pest invasions, and other crop conditions, the drones can help track both the temporal and spatial differences of various portions of a farm so that farmers can deploy resources only when and where they are really needed.
“Every farm should have a few data drones to tell them precisely what the stress level of their crops is, when they need to fertilize, water, and harvest,” says Chen. With the right resolution of data, these practices can be customized for all parts of the farm, saving expense and time over human observers.
To meet that need, Chen and colleagues are proposing to create CIDER (the California Institute of Drone Engineering Research) a UC Merced 2020 Strategic Academic Focusing Initiative proposal.
Meanwhile, it is easy to imagine these data drones facilitating the jobs of researchers in CITRIS labs on all four campuses. To keep the barrier to entry low, both to farmers and environmental researchers of all types, Chen plans to offer his drones at an affordable price. The eDNA-drone prototype his team is developing with Miller will probably cost about $1,000, despite its specialized ability to land on and remain stable in moving water. Simpler drones for precision agriculture may cost as little as $600, Chen says.
“They need to be cheap enough so that if you lose one you do not cry too much,” he says.