Autonomous agricultural spraying with unmanned flying vehicles is being refined for practical prairie applications
CAMROSE, Alta. — Spot spraying weeds in a field, killing a tough patch of tansy in a pasture, beating back brush along oil and gas well sites are all potential locations that would be ideal for drone spraying.
But there is plenty of research required before spraying with drones gets beyond a good idea. In a small Camrose farm shop, Markus Weber of LandView Drones, is spending the summer gathering data to help prove spraying with drones is safe and effective.
“The purpose of the research is to open the door with PMRA (Pest Management Regulatory Agency) and secondly to just figure out what we don’t know in the drone spraying world,” said Weber.
“Right now PMRA does not allow the application of chemical pesticides by drone except for research purposes. We’re doing research to open the door for many products to ultimately be applied.”
In Canada, the PMRA is the federal agency responsible for regulating pesticides. Applying chemicals with a drone is only allowed through a research permit.
This summer, Weber and his staff are developing a workflow to map Canada thistle patches with the drones and then test the efficacy of chemicals on the thistles. The ultimate goal, likely a few years away, is to have drone application on chemical labels, as it is now with ground and aerial application.
“In Canada, our research, and many other projects are looking at many chemicals and documenting them to see if that application by drone is safe.”
Weber is testing two large spray drones: the Chinese manufactured DJI Agras T20 that can carry 20 litres of fluid and has eight nozzles. DJI manufactures 70 to 80 percent of all the personal and commercial drones in the world.
They are also testing the American manufactured Hylio Ag 122, which can carry 22 litres of liquid and also has eight nozzles
“They are comparable models from two different manufacturers. We want to know the difference in the models. They are different brands and have slightly different flight patterns and spray systems. We want to see how realistic these would be in a field situation,” he said.
“We don’t think the models we are using today are the ones we will be using next year let alone five years from now. We really want to figure out what needs to be improved on the spraying systems and the flight systems.”
Helping with the calibration and monitoring is sprayer consultant and scientist Tom Wolf of Saskatoon.
Wolf said PMRA wants to identify a few key elements before authorizing the use of drones for spraying chemical. The agency wants to know if drone-applied pesticides behave the same as they do when applied in conventional ways. It wants to know if drone-applied chemicals have the same crop tolerance and efficacy and it wants to know if bystanders are more at risk.
“They also want to know if it drifts off target, how much does it drift, how do you mitigate it and if there are buffer zones, how long do the buffer zones have to be. These are all normal questions, it’s just they have solved them for the other sprayer methods. They haven’t got the answers for drones,” said Wolf.
“Even in the short time I have looked at spray drones they have improved and become more sophisticated. This one has a pulse width monitor. It is a sophisticated metering system,” said Wolf, pointing to the DJI Agras T20.
“I think it is incredible to have that technology on a drone.”
The pulse width monitor allows a change in the flow rate of the nozzle without adjusting pressure.
The possibility for drone spraying is still being discovered, said Wolf, who believes drone spraying would be ideal for touch ups and misses in a field after a large sprayer comes through, or for targeting weeds in areas that are difficult to access.
“Maybe a drone style will better be able to look after those areas in rangelands and pasture lands. It’s easier to see and scout for weeds with a drone and it is easier to spray them as well.”
One of the biggest areas of research will be on spray drift. During a test of one of the spray drones, spray from the two-metre-wide drone was measured as far away as 10 metres because of the air movement by the propellers, said Weber.
“We still have a lot of work to do to document exactly how wide that is. We initially laid out 10 metre blocks because we were only flying three metres above the ground with the drone and a couple metres wide. We were getting some spray on the entire 10 metre width. It’s a lot wider than we had initially imagined. The trade-off is we need to do a fair bit of research on how far the drift carries. Nice wide spray is great, but we have to be concerned about the drift.”
Wolf said a wide spray pattern is not uncommon in aircraft because of the speed of aircraft and closeness to the field. The same spray width is not unexpected in drones, but future research will focus on understanding the drift in drones.
“Spray drift hasn’t gone away. It is the key thing that can happen on a farm that the public at large takes note of.”
With drones, drift can be reduced by droplet size, but with only a small amount of water carried with each drone, finer sprays, or droplets are more efficient.
“When you have a small amount of water, you want to use that as efficiently as you can. You don’t want to make the droplets too big, but at the same time you don’t want them to drift. We are trying to find that exact sweet spot. We will likely start off using spray volumes exactly the same as the current aerial business is using and then move around from there.”
While drone spraying in Canada is still in the research phase, other countries have allowed spraying with drones. Some countries have allowed the same chemicals to be applied with drones as with aircraft. In China, about one-third of the chemical is applied using drones.
Wolf said many of the regulators, including PMRA, are working with the Organization for Economic Cooperation and Development to gather as much information from existing drone spraying. The OECD has commissioned a literature review on everything that is known about drift from drones.
“Once we see the review, we will see what kind of information is still required. Every country is different. They also have different risk criteria. We are risk averse in Canada. We do not want drift to damage anything off target and we will do whatever is required to ensure that. Those risk strategies are different in each countries,” said Wolf
“I am very optimistic we can figure this out.”