Industrial ag practices fail---they create regulatory uncertainty, data bottlenecks, and are practically useless.

The advent of precision agriculture has pushed the use of aerial systems in farming even further. Because precision ag focuses on spatial variability in crop production, getting accurate and timely measurements is crucial. Traditional aircraft has aided in these efforts, and modern drones have made this work faster, cheaper, and more accurate.

As a result, the market for agricultural drones is estimated to grow to more than $1 billion by 2024. Pushing this growth has been the increased use of drone technology to perform standard farming tasks, such as spraying crops with pesticides and fertilizer. Applications such as drone-based aerial mapping of growing areas are also pushing the technology forward in the farming world.

More and more farming operations are expected to incorporate uncrewed systems into their work in the coming years, but there are still some companies that are hesitant to go “all in” on drone-related technology. But why?

Some cite prohibitive start-up costs, as getting into the drone space requires investments in vehicles, software, and training. Others state that difficulties in determining drone data workflows and pinpointing appropriate levels of data accuracy can be daunting challenges. For other farming operations, regulatory uncertainty and managing the expectations of company leaders and clients have held them back from incorporating drones.

To learn more about the major barriers to drone adoption in agriculture and to find ways to overcome them, Commercial UAV News spoke with Matthew Johnson, Regional Vice President, Prairies & Director, Education for Volatus Aerospace. A seasoned drone operator and trainer, Johnson has witnessed, first-hand, recent developments of drone hardware and software technologies in the agriculture sector.

We also reached out to Bryan Sanders, President of Homeland Surveillance and Electronics (HSE). In his role, Sanders leads a full-service provider of industrial unmanned aircraft, and he has extensive experience helping agricultural operations adopt new technologies.

Here are their views on the “6 Barriers to Drone Adoption in Precision Agriculture and How to Overcome Them.”

1. Getting started – Adopting new systems requires investments in hardware, software, and training. Do agricultural operations have the resources, time, and knowledge to stand up effective drone programs?

Johnson: There is quite a spectrum of applications for drones in precision agriculture and trying to “do it all” at once is going to present major problems, especially for someone who has no experience with the technology. It is best, in most cases, for producers wishing to adopt drone technology to start with a crawl-walk-run approach; starting with a small microdrone—that is, a drone that weighs less than 0.55lbs (250g). These small, maneuverable, and capable systems have come a long way over the past few years, and can be used primarily for crop scouting purposes, to improve one’s situational awareness about what is going on in their fields. They are particularly valuable to assess damage from storms, wildlife, and insects, by starting with a high-altitude (~400ft) overview to get a picture of growing patterns and lowering down to specific areas to get a closer look. The cameras on some of the platforms that are available these days have improved by an order of magnitude from even just a few years ago, and the obstacle avoidance technology is also continually improving, making it easier and easier to fly without any issues.

Sanders: Although most of our customers achieve a Return on Investment (ROI) within the first year on their spraying drone, there are several factors that impact that. Setting up an effective (and efficient) drone spraying (or seeding) program is complex. From field-charging, tank refilling and safety protocols, it's a smart investment for you to learn industry best practices from people who have mastered this. Training, training, training! It will pay tremendous dividends as you'll be able to 'hit the ground running' on your first job. Additionally, new operators should get guidance on the best hardware for their specific needs. This helps to ensure your platform has the spraying components, autonomous programming, rapid battery charging, and field-serviceability that helps optimize efficiencies. Our approach at HSE-UAV is unique. We regard this hardware first as spray application systems, and then secondly, they are drones (or UGVs). The quality of the application outweighs the 'bells & whistles' because no matter how great the features—if your application isn't efficacious, you aren't making money.

Johnson: Anybody who wishes to use a drone should, at the very least, read the regulations—it only takes a few minutes! There are several courses out there that offer quality online training programs that can provide you with everything you need to know to get started within just a few hours. It is worth the investment and your time!

2. Drone data workflows – How do we create drone and data collection operations tailored to specific agriculture projects?

Johnson: This is where the industry has been getting very interesting over the last few years. Initially, almost a decade ago when drones were beginning to hit the scenes in agriculture, producers were told that they would be able to “do it all!” with the technology, using NDVI to find out which plants were healthy and which were not. However, many soon discovered that there is a major data bottleneck, and for a long time there wasn’t really a very good solution to deal with that data effectively. It can take anywhere from minutes to hours to collect the data, and then there is the challenge of either processing it locally on a very high-powered computer with high-end, expensive software, or uploading it to the cloud through an ultrafast internet connection for some data analytics provider to process and analyze as a service. Both present very real challenges. But there are a mountain of other challenges that can be faced based on what exactly you want to do with the data.

First of all, what is your objective? Research? Field scouting? Something more advanced, such as seeding or spraying crop inputs? Research activities require much higher resolution, and therefore more data. Field scouting, as mentioned in the previous section, can be done relatively quickly with a small and inexpensive system. If the desire is to create a workflow for a specific objective, such as identifying weeds in a soybean field, then we are starting to rely more on machine learning algorithms that must be purpose-built around the specific task of identifying desirable (or undesirable) constituents within a field, such as a weed, disease, or a pest. These algorithms will take a long time to develop, requiring a monumental amount of source data, and will likely need to be trained for specific crops and/or pests, and considering geography as well. This is a mega project!

3. Defining and capturing the accuracy you need – A single error in final accuracy might delay a project by a whole day or more, but what is the right trade-off between simplicity and accuracy? What happens when farming operations attempt to capture more data than they really need? How much accuracy is too much?

Johnson: What are you trying to accomplish? This is a critical question that needs to be answered. You don’t buy the hammer and start hammering nails into boards before you decide that you are building a house. In the same way, you must identify the purpose, and the process for your actions when using drones in precision agriculture. Research applications are probably the number one reason to use drones in ag, aside from simple crop scouting. Drones can capture higher resolution than satellites and planes—and higher resolution is key in research when using specialized tools such as multispectral and hyperspectral sensors to obtain imagery. Researchers don’t want to see pixels the size of plants, or even leaves, they want resolutions around half-an-inch or less! Some activities require resolutions that are less than 1mm!

It is important to consider using RTK or PPK when capturing data that is going to be used for historical tracking, research purposes, or precision treatment programs. Drone pilot’s need to be familiar with using ground control targets and GNSS stations. It is often a good idea to partner with a data processing company if you are unsure of data processing in any sort of situation that would require these levels of accuracy. You don’t want to be providing bad data—that is the worst possible outcome. Worse even than no data.

4. Regulations – Unlike other industries that use drones, such as construction and public safety, farming operations largely take place over wide stretches of land with few interactions with people or buildings. Still, government regulations around drone use must be adhered to. How do we face the ongoing challenges of complying with local, state, and federal regulations and adapting to changes in the regulatory environment?

Johnson: Regulations are slowly advancing with regard to drone spraying, but they are fairly entrenched by now when it comes to flying standard-sized drones less than 55lbs (25kg). People need to realize that the amount of work required to get certified as a pilot within your jurisdiction is not that big of an ask when you compare the value you can glean from using drones in your operations. There is often sentiment that “nobody will know!” which is unfortunate because accidents can happen, and insurance companies don’t like covering incidents that take place through illegal means—i.e., operating outside of the regulatory requirements. The first step is to take a few minutes to explore the regulations—there are not that many of them! Understand what is required and go from there.

5. Training and personnel – To move forward, companies must get staff trained and properly licensed to operate drones for agricultural operations. What’s the best way to do it?

Sanders: The regulatory framework can be confusing, and, if you're new to this, downright intimidating! The FAA states that the majority of application delays are caused by errors from the applicant! We recommend people hire an expert to help them. Our company and others offer a guidance or consulting service which helps make sense of it all while ensuring operators have the necessary permissions for your specific business. Additionally, HSE-UAV and others host some great online resources, videos and blogs about the process, too! Applicators also must comply with your local state pesticide licensing. Generally, a Google search will provide helpful information about each state.