Introducing Evja - "Rugged" Intelligent Support System for precision farming


ACTA IMEKO 
ISSN: 2221-870X 
June 2020, Volume 9, Number 2, 83 - 88 

 

ACTA IMEKO | www.imeko.org June 2020 | Volume 9 | Number 2 | 83 

Introducing EVJA: a ‘rugged’ intelligent support system for 
precision farming 

Niccolò Loret1, Antonio Affinito2, Giuliano Bonanomi3 

1 EVJA R&D, EVJA Neaples, Italy 
2 EVJA CTO, EVJA Neaples, Italy 
3 Department of Agricultural Sciences, Federico II University, Neaples, Italy 

 

 

Section: TECHNICAL NOTE 

Keywords: precision farming; internet of things; predictive models; bremia lactucae; hyaloperonospora parasitica; fusarium head blight. 

Citation: Niccolò Loret, Antonio Affinito, Giuliano Bonanomi, Introducing EVJA: a ‘rugged’ intelligent support system for precision farming, Acta IMEKO, 
vol. 9, no. 2, article 13, June 2020, identifier: IMEKO-ACTA-09 (2020)-02-13 

Editor: Mauro D'Arco, University of Naples Federico II, Italy 

Received February 28, 2020; In final form May 14, 2020; Published June 2020 

Copyright: This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 License, which permits unrestricted use, 
distribution, and reproduction in any medium, provided the original author and source are credited. 

Funding: Some of the research projects described in this article are funded by the iGUESS-MED PRIMA programme, funded under Horizon 2020, the 
European Union’s Framework Programme for Research and Innovation. 

Corresponding author: Niccolò Loret, email: niccolo@evja.eu 
 

1. INTRODUCTION 

Smart farming is revolutionising agriculture, helping farmers 
to increase the quantity and the quality of their products. When 
we refer to smart farming, we are talking about a vast array of 
systems, differing in terms of their technology and scope. 
However, at its core, smart farming consists of sensors managed 
by software that is aimed at making the farmers’ job more 
efficient and effective. Some smart farming products focus on 
robotics, machine automation, location technology, or data 
analysis. When smart farming is based on IoT systems, it is called 
precision farming [1][2][3][4]. Precision farming follows a four-
step cycle that starts with the monitoring of the plants via 
sensors, followed by the diagnostics of the collected data and 
ending either with the decision-making of the farmer or with the 
activation of another system; for example, in automatic irrigation 

systems connected to the precision farming platform. The result 
is a more controlled crop cycle, with plant and weather 
conditions monitored metre by metre, and a more accurate 
intervention by the farmers, with action undertaken only when it 
is really needed. The advantages are significant: less pesticides 
and fertilisers are used; irrigation is more efficient; and the final 
product is healthier and more abundant. Said results are achieved 
with minimum impact on the environment, leading to a win-win 
situation for the farmers, consumers, and the environment. 

We now introduce ‘EVJA: Observe, Prevent, Improve’ (or 
just ‘EVJA’), a precision farming system of our design. EVJA is 
an intelligent support system that helps farmers optimise the 
usage of chemical products and water. By using IoT and artificial 
intelligence, EVJA allows farmers to monitor their fields in real 
time, wherever they are. EVJA gathers data from a network of 
customisable sensor nodes (see Figure 1) connected to servers 

ABSTRACT 
Precision agriculture is a farming system based on the combination of detailed observations, measurement, and rapid response used to 
optimise energetic input to maximise crop production. Precision agriculture uses a Decision Support System (DSS) for optimising farm 
management. In this context, ‘EVJA: Observe, Prevent, Improve’ (or just ‘EVJA’) is an intelligent support system used for precision 
agriculture. A vast set of data (temperature, relative humidity, deficit of vapour pressure, leaf wetness, solar radiation, carbon dioxide 
concentration, and soil moisture) is continuously collected, submitted to a local control unit, and processed through algorithms 
specifically developed for different crops. On the other hand, farmers can access EVJA from their PC and mobile devices, and they may 
monitor complex agronomic data analysis presented in a user-friendly interface. 
In this article, we show how EVJA works and how its output can be used to assess the health status of plants through a specific set of 
functions. Moreover, we show the methodology utilised to develop useful predictive models based on this information. 
Specifically, we describe a predictive algorithm that is capable of predicting the infection risks of downy mildew for baby leaves 
plantations and for Fusarium ear blight of wheat. 

mailto:niccolo@evja.eu


ACTA IMEKO | www.imeko.org June 2020 | Volume 9 | Number 2 | 84 

where those data are processed. Thanks to its self-calibrating 
agronomic models, EVJA helps farmers prevent plant diseases. 
EVJA is equipped with a simple and intuitive interface (see 
Figure 2), which makes it very easy to use. It is available for every 
kind of crop and offers specific advanced features for 
greenhouse farms. We extensively describe the entire system in 
the following paragraphs, for the data collection process and data 
analysis. 

2. STATE OF THE ART 

IoT has the potential to monitor irrigation and productivity, 
and the data gathered by IoT sensors have the ability to provide 
information about the overall performance of the crops. Existing 
monitoring systems use drones and weather stations, which are 
inappropriate for greenhouse crops. Indeed, EVJA includes the 
first predictive algorithm for horticultural products in the 
European Union, while the main direct competitors 
commercialise solutions that address generally all type of crops, 
without focus and verticalisation on specific crops and weak 
results. On the other hand, greenhouse monitoring systems are 
often designed for fully climate-controlled environments, closer 
in concept to a scientific laboratory than a farmhouse, while 
EVJA’s ‘rugged’ sensor nodes are designed to be handled 
roughly, in any kind of working conditions. 

The main features characterising the EVJA system are: 

• technological: the integration with advanced predictive 
models; 

• product: the bundling of hardware and software in a 
single solution, which allows for a seamless user 
experience; and 

• business: the high scalability of EVJA, which allows for 
targeting agricultural businesses anywhere in the world. 

Indeed, compared to many other systems that are available, 
the EVJA hardware works not only with WiFi or mobile 
coverage but also everywhere else due to the use of an innovative 
communication technology called a Long-Range (LoRa) network 

[5], which can fully operate with radio frequencies. Furthermore, 
EVJA is a software, which is more flexible, more extendable, and 
more user-friendly than that of the majority of its peers. 

The EVJA system is based on a Software as a Service (SaaS) 
model, which offers an array of features, including real-time 
monitoring, forecasting, management, business intelligence, and 
social features like chat and media sharing. 

Farmers can monitor and manage everything, in each field, 
directly from their desktop, tablet, or smartphones. They can 
mark every event, like an above-average harvest, and go through 
the history to see trends and correlations between such events 
and the key factors registered by the sensors. If a worker in the 
field spots a plant affected by a parasite, they can take a picture 
and share it with the agronomist in order to check the type of 
disease and take immediate action. 

The hardware consists of a device with sensors to be installed 
in fields in an intuitive ‘plug and play’ manner and does not 
require maintenance. The sensors collect all the relevant data, 
such as humidity, temperature, light, and soil status. 

The EVJA system is totally wireless. It does not need cables 
for its power supply nor for communication with other devices. 
It includes an electronic package containing a customisable 
processing unit and a wireless communication unit to 
communicate wirelessly with external systems.  
More specifically, EVJA’s hardware includes: 

• the base station (12 × 12 × 8 cm³), which has a 
robust waterproof enclosure and covers a 
homogeneous area of about 3 hectares; 

• the battery, which is recharged using the internal or 
external solar panel (23 × 16 × 2 cm³) – the 
rechargeable battery has a load of 6600 mA h, 
which ensures non-stop working time for a month; 

• sensors, which are used to measure temperature, 
leaf wetness, humidity, and solar radiation; and 

• mobile radio options – WiFi and the LoRa network. 

3. THE EVJA SYSTEM IN DETAIL 

The node’s core is a microcontroller powered by a battery 
charged by a solar panel. Sensor probes can be easily attached to 
the device by simply screwing them into the bottom sockets. In 
the same way, sensor probes may be easily replaced. The basic 
EVJA functions need just temperature, humidity, leaf wetness, 
and solar radiation sensors. However, the entire system is easily 
customisable, and new sensors can be added depending on the 
user’s requests. As we will explain in detail later, the next EVJA 
release will be integrated with soil moisture sensors. In addition 

 

Figure 1. The appearance of an EVJA sensor node. 

 

Figure 2. EVJA’s simple and intuitive interface 



ACTA IMEKO | www.imeko.org June 2020 | Volume 9 | Number 2 | 85 

to the agronomic observables, the system gathers data about the 
state of the system, such as GPS, accelerometer data, the status 
of the battery, and GSM signal power. The data is collected by 
the sensors at every time interval (fixed by the system manager). 
At the end of the process, an antenna sends both agronomic and 
diagnostic data to the servers. The chosen data transmission 
protocol is very flexible, making use of 4G and WiFi protocols. 
In the case of poor coverage, a customised radiofrequency 
network is deployed ad hoc using Gateway to deploy a ‘star’ 
network or a mesh network of sensor nodes. In the case of a 
temporary lack of signal, the measurements are stored in a 
memory inside the control unit and are then communicated to 
the server when the signal is restored. The data stored in the 
servers are available to the users, who will always be able to 
access the required information through a simple and responsive 
interface (in Figure 2) on their smartphone or laptop. 
Logging into the interface, the user can see historical data in the 
form of simple graphics (as shown in Figure 2), select the 
required time intervals, and check the reports about thresholds 
exceeded, parasites, and other useful information on plant health. 

4. EVJA ALGORITHMS 

EVJA is equipped with several generic functions that are 
useful for defining plant status and needs, and it also has custom-
designed predictive algorithms specifically designed to keep 
parasites under control. We give below some examples to better 
explain how the system works and how farmers can use it to 
rationalise their activities and spearhead resources. 

4.1. Functions 

Using the EVJA interface, a farmer can check the conditions 
of a crop in real time. The system allows to fix thresholds for 
temperature, humidity, leaf wetness, solar radiation and other 
customisable observables (depending on the kind of sensors 
mounted on the device). In case those thresholds are exceeded, 
the system sends a warning email to the farmer. EVJA, however, 
does not just visualise those data. It also processes it in order to 
calculate the functions that are fundamental for depicting a clear 
picture of the plants’ health status, such as dew point, Vapour 
Pressure Deficit (VPD), Growing Degree Days (GDDs), and 
evapotranspiration. The dew point is the temperature T DP, to 
which air must be cooled to become saturated with water vapour. 
Dew point can be calculated through the formula: 

 (1) 

where b = 18.678 and c = 257.14 °C. When the air temperature 
is close to the dew point, the air is saturated with moisture, plant 
perspiration will not evaporate, and droplets will form on leaves. 
Dew point can also be called frost point if TDP < 0 °; in this case, 
the system warns the user about the possible damage to the 
crops. 

GDDs is a parameter with the dimensions of a temperature, 
expressing the heat accumulation by plants and insects during 
their development. Ambient temperature influences plants and 
pests’ rate of growth; therefore, when specific predictive models 

 
1 One should bear in mind that GDDs are defined by taking into account 

average daily temperatures, while the EVJA system can perform and store 
hundreds of measures each day. 

calculating some harmful insect outbreak probability or plant 
productivity are lacking, calculating the GDDs can be a simple 
preliminary step in setting an indicator [6] [7]. A simple 
estimation of the GDDs can be obtained by the formula 

 

. 

(2) 

where t is the time, N is the number of measures performed in 
one day,1 n is the total number of measures, and Tm and TM are 
respectively the minimum and maximum temperatures defining 
the survival range of a living being. Many calculation tricks, from 
trapezoid to Monte Carlo formulae for the numerical calculation 
of integrals, could make the approximation in Equation (2) more 
reliable. However, since phenomenological tables are compiled 
by taking into account the rough approximation of the 
rectangles, it is always better to stick with the simpler GDDs 
definition. 

For many species of insects, the time interval in which T < Tm 
is called diapause; in that period, their development process 
basically stops, only to resume when weather conditions are 
more favourable. Climate changes have recently affected the 
reliability of this indicator for pest control, because given the 
rising average temperatures in winter, many insects do not go 
through a diapause as they used to a few decades ago. Moreover, 
a crop can be infested by many different pests, like aphids or 
non-diapausing insects. That is why EVJA’s GDDs numbers 
should be used for pest control purposes only, with the careful 
assistance of an entomological specialist. Much simpler is the use 
of the GDDs indicator to predict the time left for a plant to 
develop and flourish: using weather forecast data, the system can 
estimate future heat accumulation and predict the period in 
which it is more likely that the plant will reach the GDDs value 
related to their full development (by confronting the system with 
well-known phenomenological tables). In this way, farmers are 
assisted in planning their future activities, such as harvesting or 
determining the right amount of irrigation needed by their crops 
in the near future. The system will then help farmers to save 
water through other parameters. 

VPD is the difference between the vapour pressure within the 
leaves and that of the atmosphere: 

 (3) 

where Tl is the leaves’ temperature, and (Ta, Ua) are the 
temperature and humidity measured by the sensors. VPD is an 
indicator of the water flow within the plants, providing 
information on the efficiency of the inner steam pump in the 
plant’s stem. Knowing both VPD and evapotranspiration ET, 
the user can determine the actual plants’ need for water. 
Evapotranspiration is estimated in EVJA through a simplified 
version of the Penman-Monteith equation [8][9]: 

 (4) 

where S is a function characterising the role of the solar radiation, 
and Kc is the crop coefficient (which depends on the plants’ 
growth stage). 

( )








−








−


















100
ln1

100
ln

U

T+c

bT
b

T+c

bT
+

U
c

=UT,T
DP

( ) ( ) dtTtT=T,TGDD
mMm

−

Mm
T

mi

T<T<
TT

N








−

1

( ) ( ) ( )
aaals

U,TPTP=UT,VPD −

( ) ( )
s0cscc

PS,ETK=PS,,KET



ACTA IMEKO | www.imeko.org June 2020 | Volume 9 | Number 2 | 86 

A clearer indication about whether is appropriate to irrigate a 
crop can be given by integrating some information about the 
actual soil moisture into the system. 

4.2. Soil moisture measurement and water saving 

Water wastage in agriculture and excessive fertilisation are 
two important issues in present-day agriculture. Problems related 
to the excessive and non-rational use of nitrogen fertilisers are 
related both to the accumulation of nitrates and nitrites in soil 
and plants as well as to the leaching of these nutrients to ground 
and surface water. While nitrates and nitrites in food are 
precursors of carcinogenic substances to humans, from an 
environmental point of view, a high concentration of these ions 
in water sources favours the phenomenon of eutrophication. 
Management of water and nitrogen fertilisers in agriculture are 
strongly interconnected practices: the optimal absorption of 
fertilisers by plants depends mainly on temperature and soil 
moisture [10]. 

EVJA is being upgraded with a dynamic forecasting model 
that simulates the mineral nitrogen content in the soil within an 
integrated sensor-based irrigation system that provides data on 
atmospheric climatic conditions, integrated with soil moisture, 
soil temperature data, and weather forecasts. This will potentially 
be a key tool for high-tech agriculture aiming to reduce the 
adverse environmental impacts thereof. The system eliminates 
the difficulties in reading and interpreting data, facilitating the 
involvement of farmers in the field, who will receive real-time 
updates on soil water content and crop water needs directly on 
their mobile device, allowing for effective and efficient 
interventions. 

The system will acquire data from wireless soil moisture 
sensors (Figure 3) to run computer simulations [11], which are 
validated through a chemical analysis of the soil in order to 
determine the actual nitrogen content in its different forms (total, 
organic, nitrate, and ammonium, which may be quantified 
through more advanced sensors such as those in [12][13]). 

4.3. Downy mildew predictive model 

Since during the first years of EVJA’s life, our activity focused 
mostly on baby leaf crops, we studied a very reliable predictive 
model for downy mildew of lettuce (Bremia lactucae) and rocket 
(Hyaloperonospora parasitica). Those oomycetes have a complex 

biological cycle [14][15] divided into many phases: a sporangia 
releases a large number of spores in the atmosphere, which 
eventually land on the plant’s leaves, where they develop (with 
the right climatic conditions) into zoospores. 

Again, given the right climatic conditions, the surviving 
zoospores will then infect the plant. It can take some days for 
downy mildew to properly develop, but after some time, which 
can be estimated by taking into account the temperature, it can 
affect the entire crop, with devastating effects. 

We managed to parametrise this process, identifying the key 
conditions allowing each stage of the aforementioned cycle to 
take place. When the system identifies the ideal conditions for 
the zoospores to survive and penetrate the plants’ leaves, it 
calculates the approximate date of the manifestation of the 
symptoms (see Figure 4) and sends a warning to the farmers. The 
warning can be visualised through an easy-to-read calendar (see 
Figure 2), where the user can also add annotations, which are 
useful to us as feedback. That feedback is actually fundamental 
to us, since it allows us to validate the algorithm with real data, 
adjusting the parameters and the functions in close relationships 
with the farmers. 

5. INTEGRATION WITH SATELLITE DATA 

As we have seen in the previous sections, EVJA algorithms 
work based on a wide variety of data through which one can 
sharply determine the health status of the plants from their 
photosynthetic metabolism on the leaves to the water draining 
efficiency of roots, but they also reliably predict the onset of 
diseases and future growth rate. Those predictions tend to be 
correct for plants near to the control unit and less reliable the 
more we step away from it. The plants are constantly monitored 
by the system, with a measurement rate of around fifteen 
minutes, and alarms are triggered any time an observable (like 
temperature) exceeds a certain threshold. Satellite information is, 
of course, inherently different from ground-gathered data: it 
allows us to monitor the crop in its entirety with a time interval 
between two measures of a few days, instead having localised 
data within minutes. Satellites and sensors give, then, two 
different kinds of information on the crop status. The first one 
is very well localised and dense over time, while the second is 
available only at given moments and concerns the crop in its 
entirety with a certain rate of precision. The best solution is 
therefore not to rely on a single source of information; rather, 
they should be integrated using custom-made algorithms. 

 

Figure 3. Wireless soil moisture sensors to be integrated in the EVJA system. 

 

Figure 4. Ideal representation of the relationship between the NDVI and crop 
coefficient for a generic plantation. In this graphic, we modelized the 
behaviour of the crop coefficient over time for a fictitious plantation with 
Kcmax = 1.1. 



ACTA IMEKO | www.imeko.org June 2020 | Volume 9 | Number 2 | 87 

Satellite images provide information on soil moisture only for 
the first 5 cm in depth, while soil moisture sensors planted in the 
ground provide data until 30 cm in depth. However, the number 
of sensors that one can plant into a crop is limited, and their data 
should be interpolated in order to approximate the water content 
of the soil in the points between the sensors. Finding out the 
correlation between the soil water content observed by the 
satellite in the first 5 cm with the one measured by the sensors in 
the spots where they are present gives a good starting point for 
adjusting the interpolation algorithms in order to better 
generalise the measures of the sensors to the entire field, with a 
good rate of reliability. Once that one gets a reliable grid that 
models the soil water content, it is possible to determine the crop 
water requirement, crossing over this information with 
evapotranspiration and VPD. In fact, while the reconstruction of 
the soil moisture in the first 30 cm of depth gives information 
about the roots’ efficiency and health, the VPD provides indirect 
knowledge of the dispersion of water from the leaves to the 
atmosphere. 

Satellite data enhances the estimation of the crop’s complex 
evapotranspiration in a significant way. On the one hand, 
satellites record the heat flux from the ground, which is 
incorporated in the ET0 term of Equation (4). On the other hand, 
taking into account the Normalised Difference Vegetation Index 
(NDVI) [14] solves the problem of the identification of the 
correct value to assign to the Kc coefficient. NDVI is defined as 
the reflection difference in the near-infrared (NIR) INIR and red 
spectrum IRed divided by its total: 

𝑁𝐷𝑉𝐼 =
𝐼NIR − 𝐼Red
𝐼NIR + 𝐼Red

 (5) 

Chlorophyll strongly absorbs visible light, while the cell structure 
of leaves strongly reflects near-infrared light. The layer along the 
bottom of a plant causes these reflections. When a plant becomes 
dehydrated, sick, affected by disease, etc. this layer deteriorates, 
and the plant absorbs more of that near-infrared light rather than 
reflecting it. Conversely, when near-infrared light hits a leaf on a 
healthy plant, it is reflected back. So, considering how NIR varies 
compared to red light provides an accurate indication of 
chlorophyll, which correlates to plant health. NDVI is also 
related to the average leaf area of vegetables in a crop; therefore, 
studying its evolution gives direct insight into the plants’ growth 
stage. This information is of paramount importance to the EVJA 
system, since it allows us to determine the value of Kc as 

 
(6) 

avoiding being forced to rely only on farmers’ estimates. 
Satellite data are useful also for algorithms on plant disease 
outbreaks. In fact, those algorithms are highly reliable for plants 
that are close to the sensor node (for example, inside the same 
greenhouse), but they do not sufficient information about plants 
that may lie in a different area. It may happen, for instance, that 
while the climatic conditions all over the field are ideal for the 
development of a disease (e.g. Peronospora Bremia), but close to the 
control unit, the plants are just not wet enough or the 
temperature is too high for the zoospores to survive. In this case, 
the support system would not raise the alarm, and even if this is 
correct in theory, it would be ineffective. Of course, most of the 
time, the data gathered by the sensor node represent the 
conditions all over the field quite well, but such issues are not 
uncommon. As for the water requirement model, through 
satellite data, the system would be able to split a crop in smaller 

sections and find out the relationships connecting the climatic 
conditions in different areas, using the sensor node as a landmark 
for extrapolating the other data. In this way, it may be possible 
to run the predictive algorithms in each cluster, obtaining 
indications about plant disease all over the crop, thereby 
enhancing the algorithms’ predictive power. 

6. CONCLUSIONS 

The EVJA system is currently working with top Italian 
farmers and monitoring more than 500 hectares. Using EVJA, 
farmers have been able to substantially reduce the number of 
chemical treatments required to hold off parasites and to save a 
large amount of water. Moreover, such an intelligent 
management of chemicals and water saves important economic 
resources. 

In the near future, we plan to add many useful functions and 
algorithms to EVJA in order to improve the service quality we 
provide to the users: multi-spectral and hyper-spectral analysis to 
directly monitor plants’ health; intelligent insect traps to keep 
track of many dangerous species; and a novel predictive model 
for the Fusarium graminearum fungus for adapting the EVJA 
system to work on outdoor crops (specifically cereals). The 
Fusarium head blight is a harmful parasite, since it releases 
deoxynivalenol within kernels, which is toxic for humans [17]. 
We managed to refine an algorithm whose skeleton is similar to 
the downy mildew one, while also taking into account rainfall 
(whose intensity should be measured with state-of-the-art 
sensors [18][19]) and dishomogeneities in T and U. We are now 
in the testing stage, and we are obtaining some encouraging 
results in confronting the algorithm’s output with historical data. 

In case the chemical treatment of such disease has not been 
possible, one can act a posteriori, isolating the infected areas and 
preventing the infection from spreading to the unaffected wheat 
inside the warehouse where the product is stored. We are 
improving the EVJA system to calculate the percentage of 
infected ears in the different zones of the crop and to report on 
a map the areas in which the infected plants have exceeded the 
concentration threshold considered acceptable. Using a 
geolocation system, EVJA should be able to suggest a smart path 
through the field in order to avoid the infected spots, telling the 
farmer to turn left or right, just like a common navigation system 
on a cell phone. Both in the case of Fusarium and downy mildew, 
as well as for every other predictive algorithm that will be 
implemented on the system, the most important thing for 
EVJA’s operation will be to endlessly evolve accordingly to the 
farmer’s feedback. This will allow us to adjust the functions’ and 
algorithms’ parameters in order to flexibly describe the situation 
on the fields and provide more precise suggestions. Ultimately, 
this progress will make agricultural work easier and more 
efficient. 

ACKNOWLEDGEMENT 

EVJA acknowledges a fruitful collaboration with Libelium 
Comunicaciones Distribuidas S.L. for the sensor node hardware 
supply and with Sentek Technologies for the soil moisture 
sensors. 

REFERENCES 

[1] W. Aulbur, R. Henske, G. Morris, G. Schelfi, Farming 4.0: How 
precision agriculture might save the world, Roland Berger Focus 
(February 2019) [Online]. Available:  

Kc=Kc
max
NDVI



ACTA IMEKO | www.imeko.org June 2020 | Volume 9 | Number 2 | 88 

https://alumni.rolandberger.com/pages/publications/2019/med
ia/roland_berger_digitalisierung_als_chance.pdf 

[2] K. Nowakowski, Farming: There’s an App for That, National 
Geographic, (5 June 2018). [Online]. Available:  
https://www.nationalgeographic.com/environment/future-of-
food/food-future-precision-agriculture/ 

[3] V. Arenella, P. Gabriele, F. Leccese, M. Cagnetti, L. Maiolo, 
A. Pecora, E. De Francesco, R. Đurović-Pejčev, Procedure for the 
space certification of a controller for soilless cultivation, Proc. of 
the 2016 IEEE Metrology for Aerospace (MetroAeroSpace), 
2016, Florence, Italy, pp. 359-364.   
https://doi.org/10.1109/MetroAeroSpace.2016.7573241 

[4] S. Kalidas, S. Bharadwaj, C. M. Vidhyapathi, B. Karthikeyan, 
Wireless node-base automatic irrigation control system, ARPN 
Journal of Engineering and Applied Sciences 12(19) (2017), pp. 
5379-5383. [Online]. Available:  
http://www.arpnjournals.org/jeas/research_papers/rp_2017/jea
s_1017_6364.pdf 

[5] C. Swedberg, LoRa Solution Offers Wireless View Into 
SoilHealth, RFID Journal (Nov.  9 2018). [Online]. Available: 
https://www.rfidjournal.com/lora-solution-offers-wireless-view-
into-soil-health 

[6] S. S. Liu, X. D. Meng, Modelling development time of Lipaphis 
erysimi (Hemiptera: Aphididae) at constant and variable 
temperatures, Bulletin of Entomological Research 90 (2000). 
https://doi.org/10.1017/S0007485300000468 

[7] R. A. Jaramillo, M. O. Guzman, Relationship between 
temperature and growth in Coffea arabica L. var. Caturra, Cenicafé 
(Colombia) 35(3) (1984) pp. 57-65. [Online]. Available: 
http://hdl.handle.net/10778/708 

[8] J. L. Monteith, Evaporation and environment, Symposia of the 
Society for Experimental Biology 19 (1965).  
https://doi.org/10.4236/ojce.2016.63029 

[9] C. H. Priestley, R. J. Taylor, On the Assessment of SurfaceHeat 
Flux and Evaporation Using Large-Scale Parameters, Mon. Wea. 
Rev. 100 (1972) pp. 81-9. [Online]. Available: 
https://doi.org/10.1175/1520-
0493(1972)100<0081:OTAOSH>2.3.CO;2 

[10] G. Bonanomi, G. B. Chirico, M. Palladino, S. A. Gaglione, 
D. G. Crispo, U. Lazzaro, B. Sica, G. Cesarano, F. Ippolito, 
T. C. Sarker, M. Rippa, F. Scala, Combined application of photo-
selective mulching films and beneficial microbes affects crop yield 
and irrigation water productivity in intensive farming systems, 
Agricultural Water Management 184 (2017) pp. 104-113. [Online]. 
Available: 
https://doi.org/10.1016/j.agwat.2017.01.011 

[11] G. Incerti, G. Bonanomi, F. Giannino, F. Carten, R. Spaccini, 
P. Mazzei, A. Piccolo, S. Mazzoleni, OMDY: a new model of 
organic matter decomposition based on biomolecular content as 
assessed by 13C-CPMAS-NMR, Plant and Soil 411(1-2) (2017) pp. 
377–394. [Online]. Available:  
https://doi.org/10.1007/s11104-016-3039-2 

[12] V. Pasquali, G. D'Alessandro, R. Gualtieri, F. Leccese, A new data 
logger based on Raspberry-Pi for Arctic Notostraca locomotion 
investigations, Measurement 110 (2017) pp. 249-256. [Online]. 
Available:  
https://doi.org/10.1016/j.measurement.2017.07.004 

[13] F. Leccese, M. Cagnetti, S. Giarnetti, E. Petritoli, S. Tuti, 
I. Luisetto, A. Pecora, L. Maiolo, R. Durovic-Pejcev, T. Dordevic, 
A. Tomasevic, V. Bursic, V. Arenella, P. Gabriele, E. De 
Francesco, Electronic nose: a first sensors array optimization for 
pesticides detection based on Wilks' A-statistic, Proc. of the 5th 
IEEE International Workshop on Metrology for AeroSpace 
(MetroAeroSpace), 2018, Rome, Italy, pp. 440-445.  
https://doi.org/10.1109/MetroAeroSpace.2018.8453631 

[14] H. Scherm, A. H. C. Bruggen, Concurrent spore release and 
infection of lettuce by Bremia lactucae during marnings with 
prolonged leaf wetness, Phytopatology 85(5) (1995).  
https://doi.org/10.1094/Phyto-85-552  

[15] H. Su, A. van Bruggen, V. K. Subbarao, H. Scherm, Sporulation 
of Bremia lactucae affected by temperature, relative humidity, and 
wind in controlled conditions, Phytopathology 94 (2004) pp. 396-
401.  
https://doi.org/10.1094/PHYTO.2004.94.4.396  

[16] J. W. Rouse, R. H. Haas, J. A. Scheel, D. W. Deering, ‘Monitoring 
vegetation systems in the Great Plains with ERTS’, Proc. of the 
3rd Earth Resource Technology Satellite (ERTS) Symposium 1 
(1974) pp. 48-62.  
https://ntrs.nasa.gov/search.jsp?R=19740022614  

[17] B. G. Shaner, Management and resistance in wheat and barley to 
Fusarium head blight, Annual Review of Phytopathology 42 
(2004) pp. 135-161.  
https://doi.org/10.1146/annurev.phyto.42.040803.140340  

[18] S. R. Barkunana, V. Bhanumathib, V. Balakrishnanb, Automatic 
irrigation system with rain fall detection in agricultural field, 
Measurement 156 (2020) p. 107552.  
https://doi.org/10.1016/j.measurement.2020.107552 

[19] X. A. Mary, L. Rose, K. Rajasekaran, Continuous and remote 
monitoring of ground water level measurement in a well, 
International Journal of Water 12(4) (2018), pp. 356-369. 
https://doi.org/10.1504/IJW.2018.095397 

 
 

 

https://alumni.rolandberger.com/pages/publications/2019/media/roland_berger_digitalisierung_als_chance.pdf
https://alumni.rolandberger.com/pages/publications/2019/media/roland_berger_digitalisierung_als_chance.pdf
https://www.nationalgeographic.com/environment/future-of-food/food-future-precision-agriculture/
https://www.nationalgeographic.com/environment/future-of-food/food-future-precision-agriculture/
https://doi.org/10.1109/MetroAeroSpace.2016.7573241
http://www.arpnjournals.org/jeas/research_papers/rp_2017/jeas_1017_6364.pdf
http://www.arpnjournals.org/jeas/research_papers/rp_2017/jeas_1017_6364.pdf
https://www.rfidjournal.com/lora-solution-offers-wireless-view-into-soil-health
https://www.rfidjournal.com/lora-solution-offers-wireless-view-into-soil-health
https://doi.org/10.1017/S0007485300000468
http://hdl.handle.net/10778/708
https://doi.org/10.4236/ojce.2016.63029
https://doi.org/10.1175/1520-0493(1972)100%3c0081:OTAOSH%3e2.3.CO;2
https://doi.org/10.1175/1520-0493(1972)100%3c0081:OTAOSH%3e2.3.CO;2
https://doi.org/10.1016/j.agwat.2017.01.011
https://doi.org/10.1007/s11104-016-3039-2
https://doi.org/10.1016/j.measurement.2017.07.004
https://doi.org/10.1109/MetroAeroSpace.2018.8453631
https://doi.org/10.1094/Phyto-85-552
https://doi.org/10.1094/PHYTO.2004.94.4.396
https://ntrs.nasa.gov/search.jsp?R=19740022614
https://doi.org/10.1146/annurev.phyto.42.040803.140340
https://doi.org/10.1016/j.measurement.2020.107552
https://doi.org/10.1504/IJW.2018.095397