Day 4

Detailed paper information

Back to list

Paper title Agricultural spraying drone based on centrifugal nozzles for precision farming applications
Authors
  1. Manuel Vázquez-Arellano Crop Production Robotics / University of Hohenheim Speaker
  2. Fernando David Ramírez-Figueroa
Form of presentation Poster
Topics
  • C4. HAPs/UAVs
    • C4.01 Innovative UAV applications
Abstract text Agricultural spraying drone based on centrifugal nozzles for precision farming applications
Manuel Vázquez-Arellano1* and Fernando D. Ramírez-Figueroa1
1 Crop Production Robotics (start-up), Institute of Agricultural Engineering, University of Hohenheim, Garbenstrasse 9, Stuttgart 70599, Germany
* manuel@vazquez-arellano.com, mvazquez@uni-hohenheim.com

Introduction
Agriculture is facing enormous challenges: it must provide food, feed, fibre and fuel for an increasing population by using the available arable land more efficiently while avoiding the intense use of resources like fuel, water, pesticides and fertilizers. Additionally, it must also act more ecologically than before and to adapt quickly to new conditions such as soil erosion, water supply limitations and environmental protection in times of climate change.
According to Rockström (2009) the rate of biodiversity loss, climate change and human interference with the nitrogen cycle are systems that are already beyond the safe operation boundary. Unfortunately, the current spraying practice doesn’t address those problems. New technologies such as unmanned spraying systems (UASS) coupled with satellite technology, Big Data and Cloud Computing could help to make the spaying applications more precise. Crop Production Robotics has taken the challenge to tackle those problems in the following way: biodiversity loss through precise delivery of pesticides to the identified pest hotspots to perform a sustainable pest management; climate change through low-emission application technology with the use of electric powered UASS; nitrogen cycle disruptions through precise and demand-driven liquid fertilization, with a more homogeneous droplet size spectra, for an adequate deposition.

Crop Production Robotics satisfies farmer’s need to adapt to the previously mentioned urgent issues, that affect their ability to maintain/drive profitability. Those issues are also stipulated and regulated in the Farm to Fork Strategy (European Union, 2020), which is at the centre of the European Green Deal aiming to make food systems fair, healthy and environmentally-friendly through the following targets by 2030:
Reduce the use and risk of chemical and more hazardous pesticides by 50%
Reduce emissions by 55%
Reduce fertilizer use by at least 20%

Methodology
The strategy of Crop Production Robotics is to design a centrifugal nozzle together with the University of Hohenheim. The droplet size spectra measurement will be performed in order to analyse, not just the typical parameters in the agricultural nozzle industry such as draftable fines (V100) and the droplet size distribution percentiles: Dv10, Dv50 – also known as volume median diameter (VMD) – and Dv90; but more important, the relative span (RS), which is an often-underplayed parameter which provides a measure of the droplet size distribution that will be used as feedback for the mechanical design of the centrifugal nozzle. The RS is calculated with the following equation:

RS=(Dv90 - Dv10)/(Dv 50)

Improving spray quality in the agricultural practice is not just about reducing driftable fines (V100), but about producing the appropriate droplet size distribution to maximize efficacy while minimizing drift potential. Therefore, we identified that the generation of homogeneous droplet size spectra by a centrifugal nozzle (as seen in the left image of Figure 1) is a cornerstone for the implementation of a sustainable spraying practice, moreover, the droplet size spectra could be adjusted for different target crops/applications while also allowing the implementation of a variable rate application.

Figure 1: Comparison between a homogeneous droplet size spectra by a centrifugal nozzle (left), and a heterogeneous by a hydraulic nozzle (right).

VMD alone is a poor way to describe a spray pattern since it provides no information about the droplet size distribution. In Figure 2 it is depicted two different spray droplet size distributions that have the same VMD value: 300μm, but the centrifugal nozzle has a smaller RS value compared to the hydraulic nozzle, meaning that the droplet size spectra of the centrifugal nozzle is more homogeneous around the target droplet size, and thus more effective at sticking to the plant, than the hydraulic nozzle.

Figure 2: Spray pattern characterisation of a centrifugal and hydraulic nozzle with same VMD value but different RS value

As previously mentioned, the main problem of hydraulic nozzles used in intensive agriculture is that they generate a wide droplet size spectra where small droplets can evaporate or drift off, and/or large droplets bounce off or roll off from the target leaf and land on the soil without achieving the desired purpose (see Figure 3). This is the reason why the scientific community estimates that between 90-95% of the pesticides land off-target (Blackmore, 2017) causing severe environmental impacts at the cost of the farmers – who pay for the wasted product. It is common practice to incorporate adjuvants in the spray mixture to improve the droplet behaviour once it has left the nozzle, and overcome barriers such as properties of the solution, structure of the target plant, application equipment, environmental conditions, among others. However, research suggests that adjuvants can be even more toxic than the active principle of the pesticides themselves.

Figure 3: Commercial hydraulic nozzles generate a wide droplet size spectra that wastes pesticide (Source: SKW Stickstoffwerke Piesteritz GmbH; Whitford et al., 2014)

The big picture of the solution proposed by Crop Production Robotics is depicted in Figure 4, where the centrifugal nozzle is the component that performs the actuation, in this case a precise insecticide application, and forms part of the UASS that receives a global navigation satellite system (GNSS) signal and a digital map of pest infestation to perform a precise application. The digital map of pest infestation is generated by the farm management information system (FMIS), and the data is acquired by either remote sensing or an unmanned aerial vehicle (UAV). The UASS and the FMIS exchange bidirectional communication with the use of telematics.

Figure 4: Big picture of the project with an example of pest management

Results
A prototype UASS is being designed and developed (see Figure 5) with a strong focus on the use of European space technology (e.g., Galileo GNSS, Copernicus remote sensing and telematics) to provide security and reliability for the navigation and bidirectional communication between the UASS and the FMIS.

Figure 5: UASS prototype by Crop Production Robotics

Applications and future
The UASS will apply pesticides and liquid fertilizer in a precise manner and with the right amount. The target droplet size generated by the centrifugal nozzle can be modified by setting the rotational speed of the peristaltic pump and the centrifugal nozzle to match the adequate droplet size with the target crop/application. Additionally, variable rate applications are also possible either by modifying the flying speed of the UASS or the flow rate of the peristaltic pump.
Since the UASS is only used a couple of months a year during the spraying season, other future applications inside greenhouses such as cooling, pest and humidity control are thinkable. Additionally, livestock applications such as barn cooling are also possible.

Bibliography
Blackmore, S., 2017. Farming with robots.
European Union, 2020. Farm to Fork Strategy, European Commission.
Rockström, J., 2009. A safe operating space for humanity. Nature 461, 472–475.
Whitford, F., Lindner, G., Young, B., Penner, D., Deveau, J., 2014. Adjuvants and the Power Spray. Purdue Ext.