Modular Drone for Life Science Applications
Agriculture and farming are among the most popular and practical use cases of Unmanned Aerial Vehicles (UAVs). Modern farming is highly precise, where continuous and accurate crop monitoring and analysis are crucial.
Our focus was on developing a drone that could serve as a universal device for inspecting outdoor agricultural areas and collecting information from areas that were thus far only accessible with piloted aircraft employing aerial photography. The process of piloting an aircraft is expensive in both time and resources, and the schedules of nearby airport services often limit these inspections. This results in the delayed acquisition of field data which is crucial in time-critical situations, for example, during floods, extensive surface water logging, soil pollution, or pest infestation, where delays can result in very unfavorable time shifts and negatively influence the measurements. Thus using an autonomous and modular drone can significantly speed up the inspection process and make it possible to record many different sources of information by adding different sensors to the platform.
The Inspection System
In cooperation with Life Science University, we selected real-world applications to utilize the UAV in the fields of monitoring water resources and the protection of plants, animals, and birds. We developed a modular quadcopter UAV with the possibility of attaching a wing module. The wing module would act as a floating device for the Unmanned Aerial System (UAS) when working over water surfaces, and at the same time as a mock-up of a predator silhouette used as a scarecrow. Additionally, the drone was also equipped with a sound generator for frightening birds which was based on the tests conducted with commercially available bird-frightening sound generators.
The drone has the capability to carry different sensors, including multispectral and RGB cameras that can be used for high resolution and 3D mapping of an environment. The drone also carries an antenna prototype which can estimate the distance to RFID tags in the environment. Even with a significantly less powerful receiver and a smaller antenna, the drone achieved an accuracy comparable to the larger antennas used on ground stations. Thus the drone is capable of detecting and actively responding to the RFID tags on the fly.
The key advantage of modularity is that it enables the integration of different sensors, such as thermospectral or RGB cameras and soil and water quality sensors that ensure the collection of a variety of data. Evaluating such a rich data can provide crucial information about the farms and the ecosystem around it. Some of the common use cases of our system are:
- Detection of the occurrence and infestation by pests and weeds in fields and forests
At the beginning of the twentieth century after the so-called Green Revolution, primary agricultural production was intensified in many parts of the world in order to produce sufficient food for an ever-growing population. Along with the intensification and mono-cultural cultivation of agricultural commodities, the share of harmful organisms in cultivated crops increased as well. The development of chemical plant protection and its large-scale application was therefore only a matter of time. The non-targeted and often preventive nature of pesticide application has resulted in increased resistance of plant pests and pathogens. Furthermore, the increased load of pesticide residue in the food chain is of significant concern and has also significantly affected surface and groundwater quality.
Currently, the strategy of banning dangerous pesticides and replacing them with less dangerous methods of protection is being applied. Thanks to the development of pesticide application technology, new strategies in plant protection are being developed, such as targeted space-specific interventions against harmful organisms. This method is not only much more environmentally friendly as the burden of pesticides is reduced, but with targeted intervention at an optimal dose, it also eliminates the risk of developing resistant populations.
However, the local delimitation for this type of protection is tied to a new technology for monitoring the occurrence of harmful organisms in individual fields or in the forest, with high accuracy of localization and determination of the type of infestation. For the purpose of accurate localization of pests, it is suitable to use various detection methods using UAVs. Usage of UAVs is economically and ecologically very advantageous. One example includes the issue of a fungal disease of sugar beets (Cercosporiose of sugar beet) where significantly more expensive aerial photography associated with GPS monitoring allows the for the application of fungicides on only exposed parts of the land to thereby reduce the environmental burden on soil, as well as both surface and groundwater.
Similarly, detection by UAV would be useful in the presence of phytoparasitic nematodes, which in most cases occur locally and are therefore necessary to go through the entire growth and mark the sites for subsequent intervention. Moreover, action against nematodes is very costly and usually affects non-targeted organisms as well.
Another possible application for which this technology is suitable is the monitoring of changes in the metabolism of plants infected with a harmful organism (fungus, virus, insect, nematodes, etc.) by means of external changes in the appearance of cultivated plants by imaging in different spectra. A plant that is infested with a particular type of pest starts to fight it to some extent and exhibits a change in metabolism. These changes are manifested, for example, by changes in the temperature of the plant, a change in the intensity of photosynthesis, etc. As in the previous cases, the location of such affected plants would allow a targeted, effective intervention.
Last but not least, the mentioned technology can be used for monitoring the micro-climate in a cultivated crop. The environment in the growth can be suitable for the epidemic development of fungal diseases. In this way, not only the scope, but also the exact date of intervention can be optimized.
This technology can also be used in forest growths for detecting symptoms of infestation by wood-destroying insects or fungal diseases, especially in terms of detecting and locating the source of the disease and its spread over time. It will be possible to intervene quickly, accurately, and to a limited extent in the area of the initial infection, which will save considerable funds and contribute to better protection of the forest and water resources in the locality.
- Measuring weeds in a field of cultivated mono-cultures
Modern worldwide crop production is focused on the cultivation of crop mono-cultures in order to achieve the optimal quality of the cultivated crop to produce a good economic result. As such, the presence of weeds in crops is undesirable and an effort is made to keep it to a minimum. This can be achieved through agro-technical interventions and by applications of selective herbicides. The herbicides in particular form a group of substances that are usually very environmentally unfriendly. They can contaminate the soil, and thus also affect surface and groundwater. Therefore, their optimization (and minimization to the necessary extent) of use is highly recommended. For plant protection, there is currently an effort to apply herbicides locally. The use of UAVs will be possible particularly for weed detection and even local intervention.
- Water-flowers monitoring on water bodies
The aquatic flower is a massive macroscopically visible overgrowth of aquatic cyanobacteria or algae in stagnant or sometimes flowing waters during the summer months. It is a recurring state every year.
The presence of cyanobacteria in the water is currently detected by taking a water sample in a plastic bottle, which is left to stand for about 30 minutes. Cyanobacteria accumulate in the upper part of the bottle at the neck due to their relative buoyancy. Cyanobacteria are the most common producer of water flowers. The problem with this method is the time required and the need to travel by boat through the entire water tank. Another problem with this method is that in the water column of the tank, cyanobacteria migrate vertically thanks to the gas sacs that regulate their buoyancy. They vegetate in a cycle, moving between the surface and the deeper layers depending on whether the sacs are filled with gas or are empty due to photosynthesis. They then sink to a depth where they pump more nutrients and replenish the gas.
The occurrence of water blooms is a growing problem in our reservoirs. In recent years, the occurrence of water blooms was detected in 80% of monitored water reservoirs. There have already been several cases of mass poisoning of the population after drinking water with cyanobacteria in Brazil and China. This has created the need for monitoring occurrences of water blooms both locally in some places of the reservoir, as well as in the spread of the water bloom “cloud” over time and area, the rate of development, and the process of occupying the water layer in the reservoir. It is possible to use UAV with sensors measuring reflection, fluorescence, and other characteristics together with corresponding geodetic data.
- Position determination of strayed cattle in free pastures
To identify cattle, a RFID sensor is placed on the ear or nose stamp of the cattle. RFID chips are small, waterproof, shockproof, and have a long battery life. They periodically broadcast a unique ID to a short distance. To determine the position of cattle, currently triangular towers or persons with mobile antenna systems are primarily used to search for the RFID chip.
Unfortunately, the technologies using active RFID have only a limited range. Active chip searches involve the construction of dense infrastructure or the use of mobile towers. The use of drones as mobile towers would bring still unimaginable mobility.
In addition, RFID chips are used in other industries, especially the transport and construction sectors. In the construction industry, the application is suitable for monitoring construction machinery, as well as the movement and occurrence of small tools.
- Ultrasound and infrasound “Scarecrowing” of birds
A long-term problem in viticulture and arboriculture throughout Europe are the significant losses in final production caused by the invasion of birds in production sites and the subsequent destruction of parts of the crop. The presence of birds in the airspace at airports also causes considerable problems in endangering air safety. As such, the interests of farmers and transportation conflict with the interests of wildlife protection.
One way to reduce the activity of birds (which mostly concerns flocks of birds) without damaging or disposing of wild bird populations is to effectively scare them off. Nowadays, stationary “scarers” that are largely acoustic based (e.g. gas cannons, explosives) or moving structures (types of so-called “scarecrows” affected by the wind) are used to deter birds in mostly vineyards and orchards.
However, birds soon get used to these stationary scarecrows and their effectiveness drops quickly, devaluing the investment. Acoustic deterrents also create an excessive noise burden and negatively affect other wild and domestic animals, as well the human population.
An ultrasonic scarecrow already exists on the market, however its focus is mainly on stationary repulsion of stinging insects in rooms or rodents in parked cars. It also has only one frequency and sound intensity.