Application of Internet of Things Technologies in Agriculture

NATALJA M. MATSVEICHUK1, YURI N. SOTSKOV2,*

1Department of Automated Systems of Production Control,

Belarusian State Agrarian Technical University,

99 Nezavisimosti av., Minsk 220012

BELARUS

2United Institute of Informatics Problems,

National Academy of Sciences of Belarus,

6 Surganov Street, Minsk 220012

BELARUS

*Corresponding Author

Abstract: - The development of agriculture in the Russian Federation and the Republic of Belarus includes

implementing «smart systems» in agriculture based on modern wireless, intelligent technologies and the

Internet of Things. This survey presents related works published in the last decade on the use of the Internet of

Things to develop agriculture. The survey is based on publications from the scientific electronic library

eLIBRARY.ru. We categorized the publications according to the areas of agricultural production as follows:

animal husbandry, crop production, greenhouses and weather forecast, water management and irrigation,

machinery management, mapping and geodesy, and digital platforms. The survey shows that in Russia and

Belarus IoT technologies are developing in agriculture intensively as in advanced countries.

Key-Words: - smart agriculture; smart farming; internet of things; animal husbandry; crop production;

machinery management; survey.

Received: August 26, 2023. Revised: November 19, 2023. Accepted: December 15, 2023. Published: December 31, 2023.

1 Introduction

Smart agriculture, or Agriculture 4.0, is such an

approach to agricultural production when

technological operations are performed using

computers, which collect, transmit, and analyze

relevant data from sensors, agricultural devices, and

machines connected to agriculture, either

independently or with a minimal human

participation, [1]. As indicated in [2], Agriculture

4.0 technologies are based on the four stages as

follows: data acquisition, data collection, data

transmission, and data processing. The acquisition

and collection of data, as well as the final execution

of agricultural operations in smart agriculture, are

carried out by the equipment related to the Internet

of Things (IoT for short). The Internet of Things is a

promising family of technologies that can offer

many solutions for the modernization of agriculture,

as evidenced by the fact that the most widely

studied issues in the field of the Internet of Things

are the applications of the Internet of Things in the

agricultural sector, followed by a food industry and

other sectors such as an energy, a healthcare and an

industry, [3]. Advances in the IoT technologies have

revolutionized agriculture by providing systems that

can monitor, control, and visualize various

agricultural operations in real time.

In [4], it is noted that the digitalization of

agriculture using artificial intelligence and the

Internet of Things has left the concept stage and has

reached the implementation stage. Farmers using

IoT seek to reduce costs by minimizing operating

costs such as labor costs, fuel, fertilizer, and

pesticide requirements while achieving better

production outcomes such as higher yields, reduced

livestock losses, and less water consumption, [5].

The main aim of this survey is to present recent

publications about the progress of using the Internet

of Things in agriculture in the Russian Federation.

This article should fill the gap in research regarding

the use of Internet of Things technologies in

agriculture in Russia and Belarus. Although a

sufficient number of articles about the Internet of

Things have been already published, there is no

survey dedicated to publications from these two

countries. As a result of our paper, a comparison

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.41

Natalja M. Matsveichuk, Yuri N. Sotskov

E-ISSN: 2224-2872

357

Volume 22, 2023

will be made of the development of IoT

technologies in Russia and Belarus relative to other

countries of the world.

We categorized publications found about IoT,

according to the areas of the agricultural production

to which they belong. The following areas are

identified: animal husbandry, crop production,

greenhouses and weather forecasts, water

management and irrigations, machinery

management, mapping and geodesy, and digital

platforms. The distribution of articles for each of the

above areas is shown in Figure 1.

Fig. 1: The distribution of the surveyed papers by

the agricultural research area

2 Related Surveys in English

We next present survey papers on developing smart

agriculture and applications of the Internet of

Things. We consider digital technologies, the use of

sensor data in agriculture along with communication

protocols used in smart agriculture. Reviews of

recent publications on new technologies in

Agriculture 4.0 are also presented.

2.1 Smart Agriculture with Internet of

Things

The Internet of Things is a computer network of

Internet-connected physical objects capable of

collecting and exchanging data using built-in

technologies for interacting with each other and the

external environment. The Internet of Things

includes various components and technologies such

as sensors, computer applications, software, and

hardware. The review articles, [6], [7] provide

various definitions of IoT and technologies used in

the Internet of Things. Researchers distinguish a

different number of levels while considering IoT. In

[6], four levels of the Internet of Things are

distinguished. The first component is a sensor that

collects data in real-time, followed by a

communication device that processes the data

transmission (the second component). The third

layer is responsible for analyzing data, while the

service layer is responsible for performing required

actions.

A detailed structure of the Internet of Things,

according to CISCO, consists of the following seven

layers; [8].

 Physical devices and controllers layer. At

this level, there are sensors, microcontrollers,

microprocessors, and actuators (devices that collect

data and transmit it for further processing). This

level guarantees the correctness and high accuracy

of collected data. Their collecting simplifies further

processing.

 Connectivity layer. This layer is responsible

for communication protocols. The communication

between sensor devices and microcontrollers is done

via BLE (Bluetooth Low Energy), NFC (Near Field

Communication), Zigbee, Wi-Fi, LoRa due to using

RFID (Radio Frequency Identification), etc. Sensors

can connect directly with microcontrollers via a

cabled connection. A microcontroller and

microprocessors use protocols such as MQTT

(Message Queuing Telemetry Transport), and CoAP

(Constrained Application Protocols) to transmit data

to the gateway. The gateway uses HTTP (Hypertext

Transfer Protocol), MQTT, or CoAP to further store

data in the cloud or on a server.

 Edge computing layer. The main purpose of

edge computing is to perform raw data processing.

These calculations are performed by the gateway

device that performs low-level data mining to

discard unnecessary data and transform

heterogeneous data into a form such that simplifies

decision-making for machine learning and data

mining algorithms.

 Data accumulation layer. Data from the

gateway devices are collected and stored in the

cloud for further processing.

 Data abstraction layer. Various data mining

algorithms are implemented to get more intelligent

information.

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.41

Natalja M. Matsveichuk, Yuri N. Sotskov

E-ISSN: 2224-2872

358

Volume 22, 2023

 Application layer. This layer is dashboards

of smart applications (such as mobile applications)

receiving data from the cloud are deployed.

 User and business layer. This level deals

with user management and business management

aspects of a fully deployed computer application.

More and more review papers in English are

devoted to the application of the Internet of Things

in the agricultural sector; [2], [3], [4,] [7], [8], [9],

[10], [11], [12], [13], [14], [15]. The agricultural

Internet of Things is used in a wide variety of areas

of agriculture (from irrigation and fertilization to

farm management systems and a blockchain).

Among the applications of the Internet of Things in

Agriculture 4.0, the paper [2] lists soil sampling and

mapping, irrigation, fertilization, disease control,

greenhouse management, vertical farming,

hydroponics, and phenotyping.

The review article [9], is devoted to the

application of smart agricultural technologies in

crop production, animal husbandry, and harvesting.

It described the advantages and challenges of

implementing the proposed solutions for each

considered study. In the review [10], it is noted that

most of the conducted research was devoted to crop

production and, to a lesser extent, animal

husbandry; while the most popular direction is

irrigation. Many types of research were aimed at

improving the productivity of the agricultural sector

by solving technical problems in terms of

mathematical and simulation modeling, [3].

2.2 Smart Agriculture Technologies and

Data

Smart agriculture uses a combination of different

technologies, which can be used depending on the

needs of the agricultural processes, [11]. The ever-

growing interest in the technologies of the Internet

of Things in scientific publications is indicated in

[12].

Various research reviews of scientific

publications consider modern technologies used in

agriculture, such as positioning systems (based on

GPS), remote sensors (including unmanned aerial

vehicles), data analytics, decision support tools,

automation, and robotics, [11]; unmanned aerial and

ground vehicles, image processing, machine

learning, big data, cloud computing and the

Wireless Sensor Networks (WSN for short) [5]; the

WSN cloud computations, big data, embedded

systems, security protocols and architectures,

communication protocols, and Web services [16];

Internet of Things, blockchain, big data, and

artificial intelligence [17]; robotics, drones, remote

sensors, and a computer vision with machine

learning and computing software [6]; Internet of

Things, cloud computing, machine learning, and

artificial intelligence [9]; IoT-based tractors, robots,

and cloud computing [2].

In [7], the following IoT technologies are

considered: special sensors, identification and

recognition, hardware, software with cloud

platforms, communications and networks, software

and algorithms, data analysis and data storage,

positioning, and security. These technologies are

divided into four areas as follows: applications,

middleware, networks, and objects.

It was noted in [3], that many studies emphasize

the integration of IoT with such technologies as a

blockchain, big data, cloud computing, machine

learning, image processing technologies, RFID, and

WSN for their applications in agriculture and food

production. Agricultural cloud-based IoT solutions

for monitoring and managing sensor networks,

drones, autonomous vehicles, robots, agricultural

machinery, greenhouses, and food supply chains are

discussed in [12]. The article [16], provides an

overview of the latest research on the application of

IoT and UAV technologies in agriculture. The basic

principles of the IoT technology are described,

including smart sensors, networks, and protocols,

the IoT applications and solutions used in smart

agriculture. The parameters monitored by sensors in

IoT-based smart agricultural applications are

discussed in [10], where it is indicated that the most

commonly used sensors are temperatures, humidity

of soil, and sunlight intensity sensors.

The problems of collecting, processing, storing,

and accessing data from sensors and other Internet

of Things devices used in smart agriculture are

considered in the following studies: [3], [4], [5],

[10], [13], [14], [17]. Many researchers, [4], [13],

note that data collected by sensors are huge (and this

amount is increasing day by day), distributed (tied

to the location and time), heterogeneous (can be

structured, semi-structured, or unstructured at all).

Often data storage and processing must be realized

in real-time, which requires additional computing

resources, [5].

A large amount of data coming from IoT sensors

requires a high processing speed and real-time

decision-making. For IoT cloud processing in smart

agriculture, Lambda or Kappa architectures are

commonly used. However, they are not specialized

in smart farming. The authors of the article [18],

present an optimized version of the Kappa

architecture, to improve a memory management and

data processing speed for fast and efficient IoT data

management in agriculture. The parameters of the

proposed Kappa architecture were configured to

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.41

Natalja M. Matsveichuk, Yuri N. Sotskov

E-ISSN: 2224-2872

359

Volume 22, 2023

process data from the analysis of animal behavior in

precision animal husbandry. It was shown that the

combination of the Apache Samza computing

system with the Apache Druid database provides a

higher performance. The results of the conducted

research showed the essential effect of adjusting the

parameters on the speed of treatment.

The architecture of the IoT service platform has

been developed in [19]. The proposed platform for

watering and fertilizing plants has been modeled

based on the analysis of leaf image data. Soil

temperature, humidity, and pH sensors were used,

and cameras were used to take pictures of the

condition of the leaves based on Raspberry Pi. If

damage is detected, the system takes into account

the temperature and humidity data and then decides

to start the automatic water and nutrient supply

mechanism.

In [20], a common architectural framework for

modeling food systems based on IoT, combining

technical and business aspects was developed. The

applications of the proposed framework to 19 cases

used in different regions of Europe were described.

It demonstrated the effectiveness of modeling the

IoT-based architectures in a wide range of different

agricultural areas.

As a rule, the developed systems have a multi-

level architecture; Figure 2 (four levels as indicated

in [21] and three levels as indicated in [22], [23]).

Fig. 2: A comparison of the multi-level architectures

in different IoT smart-farming systems.

The three levels in IoT-Agro architecture are as

follows: an agricultural perception, an edge analytic,

and a data analytic, [22]. The agriculture perception

layer consists of IoT-enabled devices (e.g., sensors,

actuators, weather stations, tags, and RFID readers)

that monitor real-time weather conditions and

product health across the entire value chain

(cultivation, harvesting, post-harvest, drying, and

storage). This layer provides data to the analysis

components in the data analytics layer.

The edge layer is close to the endpoints for on-

premises real-time computing and other data

processing to reduce the burden on the data

analytics layer and improve the reliability of

agricultural IoT data. This layer includes

interconnected edge devices, and physical or virtual

objects (e.g., routers, switches, wireless access

points, repeaters, embedded systems, and servers)

geographically distributed across farms to collect all

data from the agricultural perception layer.

The data analytic layer consists of data centers

and traditional cloud servers with virtually unlimited

computing and storage resources. At this level, IoT-

agro services are deployed based on data analysis.

Various approaches to the description of the

architecture of the Internet of Things were discussed

in [14], where the hardware technologies and

communication protocols used, their advantages and

disadvantages were described.

This study was continued in [15], where it was

concluded that there was no universal and unique

architecture for the IoT applications in smart

agriculture that meet all the needs of all use cases.

2.3 Communication Protocols in Smart

Agriculture

The unprecedented data collection and management

capabilities offered by the IoT are based on several

factors of the underlying architecture and

technology of the communication network, one of

the most important of which is the protocol that is

used between IoT nodes, gateways, and application

servers. A comparison of the technologies used and

the protocols of the Internet of Things at the

application, data transmission, the Internet, and the

network interface layers, was carried out in the

review, [7].

The authors of the paper [8], analyze the

scientific literature on IoT communication

technologies in smart agriculture. Wired, wireless,

and hybrid technologies are used to transmit the

collected data from the sensors to the control center.

Wireless communication technologies IoT (Wi-Fi,

ZigBee, LoRa, RFID, mobile communications, and

Bluetooth) used to connect to various devices in

different layers of agricultural production are

analyzed. Different technologies are compared in

terms of technical characteristics and cost. It should

be noted, that only one publication used wired CAN

data transmission technology.

The published results show that each technology

has its advantages and limitations, and different

wireless communication technologies are suitable

for different scenarios.

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.41

Natalja M. Matsveichuk, Yuri N. Sotskov

E-ISSN: 2224-2872

360

Volume 22, 2023

Different technologies of data transmission (such

as mobile communications and 2G/3G/4G/5G

technologies, Zigbee, Bluetooth, and LoRa) were

considered in [2], [3], [13]. The most commonly

used data transmission technologies for collecting

sensor data were Zigbee and Wi-Fi, [13]. Wireless

protocols such as ZigBee and LoRa are

advantageous for agricultural applications compared

to other protocols due to their low power

consumption as well as the desired communication

range, [13].

The survey [24], offers an overview of research

on the Internet of Things protocols focusing on their

main characteristics, performance, and frequency of

use in agricultural applications. Protocols for the

data exchange between Internet of Things devices

MQTT, CoAP, XMPP, AMQP, DDS, REST HTTP,

and Web Socket have been considered. These

protocols were compared in terms of efficiency

indicators (delay, bandwidth, and power

consumption). It was shown that the most popular

communication between the device and the IoT

gateway is the MQTT protocol, which is a leader in

almost all performance indicators.

The MQTT protocol was used in [19], for the

communication between different devices. The data

was transmitted to the fog node via the Wi-Fi

protocol. The MQTT protocol for the IoT was

discussed in [8], with the issues of authentication

and security during data transmission and the

problems of protecting IoT devices from physical

and logical attacks.

In [25], it is noted that the use of the MQTT

communication protocol allows different types of

devices to be used in practical agriculture. The

transmission of messages from IoT to the cloud

requires comprehensive protection, even when using

the Transport Layer Security (TLS) protocol. In the

article [26], a system is proposed that standardizes

the transmission of messages from the device to the

cloud platform and vice versa and provides end-to-

end security. The conducted experiments

demonstrate the effectiveness of the developed

system and guarantee a unique identification of

devices in the domain.

Experimental developments on the use of various

communication protocols for transmitting data from

sensors to the network in IoT systems for various

agricultural applications are described in the article

[27], where an adaptive network mechanism for a

smart farm system using LoRaWAN and IEEE

802.11ac protocols is presented. This system can

configure the protocol depending on the state of the

network. For example, IEEE 802.11ac was suitable

for transmitting data that requires high speeds, such

as images or videos. In contrast, the LoRaWAN

protocol was suitable for sending data that has small

data packets such as sensor read data. An adaptive

mechanism that combines the advantages of both

protocols ensures the reliability of the system when

performing the monitoring task. The proposed

mechanism was implemented at the application

level and tested by collecting data in a greenhouse

in South Korea. The obtained result showed that the

proposed system improves network performance

and provides reliability in terms of average latency

and the total amount of sensor data collected.

It is noted in [8], that when designing a

communication network, one should use the same

protocol from IoT devices to the gateway, from the

gateway to the cloud (the IoT platform), from the

cloud to the end-user device, and vice versa to

control these actuators. This structure is optional

and different protocols may be used in different

parts of the communication network architecture

depending on the requirements of the IoT platform,

the hardware and software requirements of the

gateways, etc. In the context of the IoT protocols

and standards, open-source software and hardware

were preferred, [3], since they can solve

interoperability issues concerning protocols and

devices.

2.4 Some Projects Effectively Realized in

Agriculture

Reviews of research papers demonstrating the use of

new technologies in Agriculture 4.0 are presented in

[2], [4], [5], [8]. In [28], it is shown that many

technical leaders (such as Dell, IBM, Microsoft,

CISCO, Google, Intel, Qualcomm, etc.) are making

a lot of efforts toward the potential use of the

Internet of Things in agriculture. In particular, 17

academic and 17 commercial developments of the

Internet of Underground Things (IoUT) systems

were described. Eight platforms for storing,

analyzing, and processing data in precision

agriculture are presented. A list of companies and

manufacturers of drones and sensors, providers of

services for processing drone data, the use of drones

for processing fields, and platforms for managing

them was given. Note that the above companies are

located in the United States, Canada, Australia,

Switzerland, Hong Kong, the United Kingdom, and

South Africa.

Brief characteristics of the developed

applications using the Internet of Things, UAVs,

and IoUT for agriculture (soil sampling and

mapping, irrigation, fertilization, disease control,

greenhouse management, vertical farms,

hydroponics, and phenotyping) are given. In the

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.41

Natalja M. Matsveichuk, Yuri N. Sotskov

E-ISSN: 2224-2872

361

Volume 22, 2023

most developed countries, tractors with a built-in

navigation system and sensors that track all

elements in the field are already common on farms.

Tractor manufacturers (such as John Deere and Case

IH) have begun offering autonomous tractors to

farmers, [4].

A review of 94 scientific publications on

communication technologies of the Internet of

Things in smart agriculture from three databases

(Science Direct, IEEE Xplore, and Scopus) is

presented in [10]. The geographical distribution of

the selected articles shows that the most productive

are the authors from India with 25 study examples.

It is followed by China (15 articles); the USA and

Korea (five articles and four articles, respectively);

Mexico, Spain, Italy, Pakistan, Viet Nam and

Malaysia (three articles each); United Kingdom,

Tunisia, Indonesia, Brazil and Turkey (two articles

each); and Portugal, South Africa, Australia,

Ireland, Macedonia, Greece, Thailand, Egypt,

Nigeria, Norway, Colombia, Algeria, Saudi Arabia,

Romania, Russia, Kuwait and Bangladesh (one

article each).

The authors of the survey [9], discuss the

application of smart farming to crop production,

animal husbandry, and post-harvest processing. This

survey examines research on the identification of

diseases on cucumber leaves in the Cyprus region,

on the development of a 3-D method for the visual

detection of sweet pepper peduncles in Australia, on

the use of smart sensors in poultry houses and farms

in Europe, on solving specific problems in fish

farming (biomass monitoring, a feed delivery

control, parasite monitoring, and a crowding

management) in Norway, and on determining the

best time to harvest coconut for aromatic coconut

producers in Thailand.

In [29], it was proposed a MooCare model

designed to help producers manage dairy cattle to

increase productivity. Using IoT devices, MooCare

automates and personalizes animal feeding. The

proposed model applies for on-premises deployment

to a farm or cloud compute resource that is

accessible from the Internet.

This model contains the following functions:

obtaining data on cow's productivity (based on the

animal identifier and milking sensor), predicting the

milk production of the animal (using the ARIMA

model to determine the predicted value), supplying

the concentrate individually to each cow depending

on its milk production (regulated by the actuator)

and sending warning notifications to the producer

using threshold values. The estimates include

modeling based on cow lactation data from a real

farm located in the southern region of Brazil. The

computations showed that the model can provide an

adequate forecast of milk production, the reliability

of the forecast was 94.3%.

The authors of the article [5], provide an

overview of European projects in the field of smart

farming. The first part of this paper presents the

results of research by scientists, who apply

innovative technologies to grow various crops in

Europe, classified depending on the European

country, the technologies used, the type of field

work, and the type of crops.

Realized projects tested in the fields or

greenhouses in Europe were considered, including

one study, [30], on the deployment of IoT in a

tomato greenhouse in Russia, using wireless

sensors, cloud computing, and artificial intelligence

to monitor and control plants and conditions in the

greenhouse, and predict the growth rate of tomatoes.

In the second part of the paper [30], an analysis of

the 18 most significant projects in the field of

intelligent agriculture, funded in Europe, is carried

out.

A scalable IoT-based monitoring system with

forecasting capabilities for agriculture has been

developed in [21]. The proposed IoT system was

designed and experimentally tested by monitoring

the temperature and humidity in a commercial

tomato greenhouse in Mexico for six months.

Predictive modeling of the greenhouse microclimate

based on data using an artificial neural network was

implemented.

The obtained results showed that the ANN model

can be successfully used to predict a temperature for

24 hours with a simple three-layer ANN of 8

neurons in a hidden layer. The temperature forecasts

were accurately fulfilled 24 hours in advance with

an error of 1°C. The results obtained confirm that

the proposed IoT framework can make it easier for

farmers to monitor their crops and increase the

production of crops.

Disease damage to crops poses a problem to

agricultural production by reducing productivity.

The solution to this problem may be early diagnosis

of diseases. In the paper [31], a deep learning

method for classifying and predicting guava leaf

diseases was developed. This method was tested on

a data set of 1834 leaf specimens and was shown to

be rather effective compared to traditional visual

diagnostic methods.

The paper [32], discusses the effect of radiation

on food images and proposes metaheuristic

optimization algorithms for detecting images of

irradiated fruits and vegetables. The image

segmentation was based on three different

metaheuristic algorithms used to detect the

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.41

Natalja M. Matsveichuk, Yuri N. Sotskov

E-ISSN: 2224-2872

362

Volume 22, 2023

difference between food images before and after

irradiation. These algorithms were able to detect the

effects of the radiation on a green apple, cucumber,

and orange, even if it was not visually recognizable.

A flexible platform for soilless cultivation in

fully re-circulated greenhouses using moderately

salty water was proposed in [23]. The system was

implemented in a real prototype in a greenhouse in

southeastern Spain as a part of the EU Drain-Use

project. Two cycles of tomato cultivation were used.

The first is to test the correctness of the architecture,

and the second is to analyze the improvements of

the system compared to the harvest in the open field.

Savings of more than 30% on water consumption,

and up to 80% on nutrients were obtained.

The field irrigation management system receives

soil moisture data from sensors installed at several

points in the sowing field. The paper [33], presents

an intelligent irrigation system predicting soil

moisture data using weather, yield, and irrigation

data. Computational experiments were conducted to

calculate irrigation forecasts using data from

experimental coconut and cashew fields in Paraiba

(Brazil). Soil moisture data from sensors and

weather data from a public weather station were

used. The computational results showed that

predictive models were quite effective and

contributed to saving irrigation water.

An irrigation strategy is proposed in [34], based

on zonal irrigation, fuzzy logic, wireless

communication, and the IoT to monitor irrigation

and maintain soil moisture in ideal conditions for

plant growth while consuming minimal water and

energy. The developed zonal irrigation system was

applied to irrigate tomato plants in a greenhouse in

Algeria. The experiment lasted for eight days. The

area of six square meters in a question was divided

into two zones. In each zone, there was a wireless

unit with a solenoid valve a soil moisture sensor,

and a sensor unit for measuring the ambient

temperature. The Raspberry Pi was used and served

as the HMI server and host. The system sends

sensor data to the server via a radio frequency

communication. A fuzzy logic controller processes

this data and decides to control irrigation. The

developed system can monitor and control

greenhouse irrigation from anywhere and at any

time using a human-man interface (HMI) developed

as a part of IBM's Node-RED. The conducted

experiments have shown that combining a fuzzy

logic controller with a zoning strategy is superior to

other tested algorithms in terms of minimizing water

and energy consumption.

Vertical Gardens (VG for short) is a method of

growing plants in vertically stacked layers in closed,

including high-rise, buildings. The vertical

cultivation concept uses indoor cultivation systems

in a controlled environment, where each individual

environmental factor can be monitored and

permanently controlled. In [13], 30 implemented

projects (in the period from 2014 to 2018) on the

vertical cultivation of plants using IoT technologies

are considered. It was found that VG using the

Internet of Things is most common in the United

States (41.2% of the projects considered) and in

China (23.5%, respectively). Due to their high level

of technological readiness, VGs are popular in the

USA and Europe. Interest in VG adoption is

expected to increase in the near future in Turkey,

Singapore, Japan, South Korea, and Malaysia.

The IoT system with a new nitrogen,

phosphorus, and potassium sensor (denoted as NPK

sensor) with a light-dependent resistor and light-

emitting diodes was developed by the authors of the

paper [35]. Data collected by the developed NPK

sensor from selected agricultural fields is sent to

Google's cloud database to support fast data

retrieval. A fuzzy logic is used to detect nutrient

deficiencies from the sensed data. A warning SMS

message is sent to the farmer about the amount of

fertilizer to be used at regular intervals. A hardware

prototype of the sensor and Python software for the

Raspberry Pi-3 microcontroller were developed.

This model was tested in India on three soil samples

(red soil, mountain soil, and desert soil). The

analysis of the developed NPK sensor in terms of

throughput, end-to-end latency, and jitter was

performed using the Qualnet simulator. The

experiments showed that the developed IoT system

can increase crop yields.

In [28], it is described a technological equipment

and different approaches used in precision farming

and the IoT. These researchers considered case

studies from different countries (Italy, Greece,

France, the USA, Japan, Argentina, and Tanzania).

An example of a Mediterranean farm (a commercial

winery) in France is presented, in which digital and

precision farming tools are used for winemakers and

wine consultants. A study on precision farming in

perennial crops in Greece highlights the use of

remote sensing and near-range sensing in a variety

of settings. The application of variable-rate nitrogen

fertilizers based on prescription maps and sensors

“on the go” is an example of corn cultivation in

Northern Italy. Smart irrigation is an important

theme in the United States, [28]. This case study

highlights technological advances in cotton

irrigation to optimize yield and sustainability.

In Japan, the production of rice based on

proximal sensors and IoT is also described in [28].

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.41

Natalja M. Matsveichuk, Yuri N. Sotskov

E-ISSN: 2224-2872

363

Volume 22, 2023

Some other case studies (in Argentina and

Tanzania) discuss an overview of the

implementation of smart farming technologies and

techniques used in these countries. The results of the

research [28] show the potential of precision

farming and the economic profitability of the latest

technologies, as well as improving environmental

sustainability. It is emphasized that in some tested

countries, there is a lack of technology (for example,

new machine systems and knowledge in the field of

data analysis). The transition to an IoT system will

require significant investments.

Applications, software, and hardware play a

crucial role in ensuring the intellectualization of the

agricultural systems. The developed applications are

responsible for collecting data for further analysis,

such as Nutrient ROI calculator, Sirrus, FieldAgent,

OpenIoT, Farmbot, SmartFarmNet, iSOYLscout,

AgVault 2.0, AgriSync, FARMapper, are

considered in [28]. There is also an overview of

agricultural projects developed in Italy, Spain,

Austria, the USA, India, Pakistan, Brazil, and other

Asian countries.

The survey paper [3], reviewed 30 articles with a

hardware implementation and six articles with real

applications of the Internet of Things in agriculture,

such as automated irrigation, monitoring of soil

parameters, and product traceability systems. For

example, the IoT Agriculture (AIoT) pilot project in

China, which uses IoT technologies to ensure food

safety, is considered.

Note that most of the publications reviewed

related to China and India, followed by Spain, Italy,

France, the Netherlands, the USA, and South Korea,

as well as some European countries, while none of

the survey articles described Russian developments

in agriculture.

3 Developments of the Internet of

Things in Agriculture of the Russian

Federation

It should be noted that there are no reviews in

English that include a description of the use of the

Internet of Things in agricultural production in the

Russian Federation, while Russia is one of the world

leaders in the production and export of several

agricultural products. We next are going to fill this

gap and offer a reader a review of the works on the

digitalization of agriculture in the Russian

Federation.

Nowadays, in agriculture in the Russian

Federation, elements of the “precise” farming

system (parallel driving systems, fuel consumption

accounting, differentiated application of fertilizers

and plant protection products) are developed, and a

lot of projects are implemented to digitalize animal

husbandry (herd management systems, automated

animal feeding, traceability of animals and

products). The importance of work on information

support for monitoring the resource potential of

fields and precision farming using GIS is increasing,

which is due to the geographically distributed

structure of production and the economic need for

more accurate agricultural production.

Research using advanced information

technologies (neural networks, genetic algorithms,

artificial intelligence methods, cloud technologies)

is promising. The agricultural sector is fully capable

of using technologies such as the Internet of Things,

cloud services (agro-scouting, accounting,

management of an agricultural enterprise through

mobile devices), big data, blockchain, artificial

intelligence, ERP systems (integration of disparate

data in a single system).

The researchers examine the strategic directions

of digitalization, adapted to the activities of an

agricultural enterprise, which are possessed by

scientific achievements in the fields of robotics,

automated control systems, precision farming, and

remote sensing of the earth and satellite

cartography. It has been substantiated that the use of

innovative technologies in agricultural production

has undeniable prospects that will allow obtaining

positive dynamics in the production and sale of

products, reduce operating costs, storage, and

transportation costs of agricultural products, and

increase the innovative component in the added

value of the product.

It is listed different types of digital technologies

used in agriculture. Computational decision-making

tools, cloud technologies, various types of

surveillance equipment, micro-robots, digital

communications (mobile, broadband, LPWAN),

geo-location (GPS and RTK), GIS, yield monitors,

UAVs, automatic control and guidance, variable

speed technology, on-board computers, radio

frequency identifiers, automated milking, feeding

and monitoring systems are listed.

The following technologies are especially in

demand in agriculture: GIS technologies

(geographic information systems and technologies

for remote sensing of the Earth); precision

agriculture technologies; big data technologies;

Internet of Things technologies; artificial

intelligence technologies (digital twins), etc.

We next present a survey of recent achievements

on using the Internet of Things and digital

technologies in agriculture in the Russian Federation

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.41

Natalja M. Matsveichuk, Yuri N. Sotskov

E-ISSN: 2224-2872

364

Volume 22, 2023

respecting the following areas: an animal

husbandry, a crop production, greenhouses and a

weather forecast, a water management and

irrigation, machinery management, mapping and

geodesy, and digital platforms.

3.1 Animal Husbandry

Systems for monitoring animals, their behavior,

physiological parameters, and productivity are

widely introduced in animal husbandry, [36].

An automated veterinary information system is

being developed. An animal identification and

tracking module has already been developed for this

system. This module allows recording and tracing

animals based on visual methods (using ear

identification numbers), as well as radio frequency

identification methods (by implanting a

subcutaneous microchip); carrying out the

movement of animals between livestock points and

owners; generating reports on transfer, input,

disposal of animals, etc. The developed software is

widely used in the Stavropol State Veterinary

Service. The database includes 827.1 thousand

heads of small ruminants and 209.6 thousand heads

of cattle, including 7.5 thousand heads identified

using radio frequency technologies, placed on 23.2

thousand livestock facilities in the Stavropol

Region. The use of the animal identification module

made it possible to increase the speed of data

processing, improve the quality of information, its

suitability for analytical processing, and strengthen

control over the movement of animals, which

creates the prerequisites for the digital

transformation of the management of the state

veterinary service of the Stavropol Region.

The problems of managing dairy farms and

remote control of the milk quality are considered in

literature and a four-layer IoT network structure for

managing a dairy farm is proposed. This network

includes milk analyzers, a gateway, a cloud

platform, and mobile applications for farmers and

operators. The selection of an appropriate network

protocol was carried out according to the following

four network indicators: transmission speed,

distance, frequency, and security.

The 4th generation of the LTE network using

СIoT-LTE-M technology was chosen as a network

for transmitting information from dairy farms to the

cloud environment. A generalized algorithm for

farm milk quality control has been developed, which

includes receiving information from analyzers,

transmitting it through a gateway to a cloud

platform for storage and intelligent processing, and

displaying the results through operator applications.

In Belarus, for the mechanization and automation

of technological processes in pig breeding, a wide

range of equipment has been developed for

automated preparation and normalized distribution

of liquid feed mixtures and dry feed, an automated

station for individual feeding of sows, and a set of

equipment for multiple feeding of animals by bio-

phases, [36]. All the above equipment operates in

automatic mode with the possibility of remote

control via the Internet.

3.2 Crop Production

A method for building expert Decision Support

Systems (DSS) was developed in the number of

articles for solving the following three problems: the

formation of strategies for applying mineral

fertilizers and long-acting ameliorants in crop

rotations of various types; managing the state of

spring wheat by forming a sequence of

technological operations in one growing season;

choosing the optimal date for harvesting fodder

from perennial grasses; [36].

When solving the first problem, an algorithm for

minimizing the risk of crop losses and overspending

of mineral fertilizers and ameliorants was

developed. For different variants of the initial values

of the parameters of the chemical state of the soil

and different climatic conditions for each type of

crop rotation, the optimal strategies for applying

mineral fertilizers and ameliorants were determined.

To form a cloud knowledge base for managing the

state of spring wheat, several algorithms were

developed for the formation of optimal programs

that minimize the risks of losses in the spring wheat

crop. To make decisions about the optimal dates for

harvesting fodder from perennial grasses, two

variants of algorithms were used in the local DSS

and tested.

The developments of the Federal Scientific,

Agricultural and Engineering Center VIM in

horticulture and crop production are devoted to the

implementation of management decisions using

robotic technologies. Intelligent robotic systems for

regulating microclimate and nutrition parameters,

controlling fertilizer application and protective

equipment plants, picking berries, etc., are

developed.

An automated control system for agricultural

technologies in industrial horticulture is described

with the possibility of conducting ground

inspections using a mobile application. This system

provides real-time processing of information flows

reflecting the characteristics of the growth and

condition of plants in critical phases of

development. It is shown that the system

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.41

Natalja M. Matsveichuk, Yuri N. Sotskov

E-ISSN: 2224-2872

365

Volume 22, 2023

automatically optimizes machine technologies for

cultivating horticultural crops according to the

biological criterion (implementation of the potential

biological productivity of crops) and the economic

criterion (improving the efficiency of the use of

production resources). For agriculture, a promising

direction is the use of GIS and GPS for both ground-

based and aviation and satellite sensing.

Different possibilities of remote ground and

airborne sensing methods using controlled manned

and unmanned aerial vehicles and satellites to

improve the performance of phytosanitary

monitoring of agro-ecosystems are investigated. It is

shown that monitoring using a recognition of the

phytosanitary state of agro-ecosystems, together

with cartographic information obtained based on

GIS and GPS, makes it possible the short-term,

long-term, and long-term forecasts for assessing the

distribution of harmful organisms and the volume of

plant protection from pests and pathogens in agro-

ecosystems. These possibilities are presented as

applications «Agronomist's Diary» for smartphones

and tablets.

Unmanned aerial vehicles (UAVs) make it

possible to cultivate land plots of a complex

configuration, apply fertilizers differentially

according to the given program, and automate the

processing of plants with minimal human contact

with pesticides, and the work at night without

compromising the quality of work. At the

Belarusian enterprises OJSC Govyady-Agro and the

Novitsky state farm, agricultural drones were used

to spray plants. The advantage of this treatment is a

deep penetration of the fertilizer into the plant mass

due to the airflow from the UAV propellers [36].

3.3 Greenhouses and a Weather Forecast

Greenhouses are sites where IoT technologies are

most actively applied, and some of their

developments are presented in the survey, [36]. To

support such technology, “Ruselectronics” holding

has developed “Smart Greenhouse” software, which

is a constructor for the accelerated deployment and

implementation of IoT devices and networks.

For a practical implementation of the developed

system for smart greenhouse, it is necessary to use

modern wireless and IoT technologies. When

solving the problem of adapting the communication

standard to the conditions of agricultural production

to ensure wireless data transmission over long

distances, the LoRa standard at a frequency of 433

MHz was justifiably chosen as a standard for data

transmission from bots and sensors. The selected

standard was proposed to be used for data

transmission in the intelligent control system of the

plant hydro-melioration robot in artificial

ecosystems “Hydrobot 1.0”.

The following sequence for using digital

technologies in artificial ecosystems is considered:

collecting data on ecosystem parameters (a sensor

network) → transfer to databases (cloud storage) →

data processing and decision making (control signal

generation). Based on the Arduino platform, a

“registrar” device has been developed that allows

real-time recording of the object indicators, and

external factors and saves them in a cloud database.

It is described the registrar design, which is based

on a programmable controller with an ATMega

processor.

A greenhouse microclimate control device

based on the Arduino Uno was developed. The

functional requirements for the developed device are

determined, the block diagram of the device is

given, and the user interface developed in the

IoControl cloud service is presented. The developed

device allows one to take readings from devices in

the greenhouse, transfer them to a personal

computer or phone via the Internet, and also control

the actuators inside the greenhouse online using a

cloud service. A user can view data and remotely

control the actuators, turning on and off the heating,

irrigation, lighting, and window opening systems.

The intelligent robotic complex that regulates

microclimate and nutrition parameters to control

plant growth in closed artificial ecosystems is

developed.

Neural networks could be used to build a short-

term forecast of air temperature. As a basic neural

network, a non-linear autoregressive model with

external input data (NARX) was used. The training

was carried out on the data obtained with an interval

of 10 min and an observation time of 72 hours. The

Neural Time Series utility of the MATLAB package

was used.

A check of the predictive properties of the

resulting neural network, carried out on a sample of

data not used in training, showed that in most cases

the correlation coefficient of the input data and the

forecast was more than 0.96, and the root-mean-

square error did not exceed 1.5 °C. Despite the high

accuracy of forecasting, it is noted that to build a

functioning temperature forecasting system, it is

necessary to periodically retrain the network to take

into account the variability of parameters depending

on the season. Such a retraining scheme can be

easily implemented using cloud services and the

Internet of Things.

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.41

Natalja M. Matsveichuk, Yuri N. Sotskov

E-ISSN: 2224-2872

366

Volume 22, 2023

3.4 Water Management and Irrigation

The management of the reclamation regime of

agroecosystems is considered in many papers; [36],

[37]. The requirements for the use of digital

technologies, such as IoT, have been formulated,

and a list of priority tasks for automating the

technological processes of reclamation agriculture

has been developed.

An automation of agricultural production

management on reclaimed lands ensures the

implementation of the established sequence of

technological procedures with maximum speed and

accuracy. The implementation of management

decisions in an automatic mode using intelligent

algorithms will provide energy and resource-saving

by identifying patterns of controlled processes based

on the use of innovative data processing algorithms.

In the work [37], promising areas are shown for

improving the digital development of agricultural

production on reclaimed lands: cloud technologies

and big data, software products based on neural

networks and artificial intelligence, software-

controlled complexes that provide the user with the

resulting information for corrective actions; Internet

of things and other innovative developments in the

management automation, providing data collection,

support and implementation of management

decisions.

The requirements for the functional structure and

architecture of modern automated production

process control systems have been determined.

These systems monitor and record the ameliorative

state of agro-ecosystems, intelligent data processing,

the formation of management decisions and their

implementation automatically without human

intervention.

The commercial automated process control

systems that ensure the regulation of irrigated crop

production operations were analyzed. It was showed

a significant lag of domestic products from the best

foreign samples. In [37], systems for operational

monitoring of soil and weather conditions in the

practice of agricultural production was described,

which help not only to track changes in conditions

and remotely control irrigation systems, but also

generate effective management decisions. It is noted

that the processing of the primary information

should be carried out online and used for the

operational management, adaptation, and

development of the control system by setting the

parameters of mathematical models and for solving

tasks of higher levels of the management hierarchy.

Studies of water bodies were carried, and a series

of interactive hydrographical maps of the city of

Brest (the Republic of Belarus) was developed, with

a visualization of data on the content of micro-

plastic particles in 25 water bodies.

3.5 Machinery Management

Precision farming includes not only crop production

technologies but the use of the latest robotic

machines in an optimal way. It is required to

optimize not only the monitoring and management

of the agricultural equipment but also the compound

of the machine and tractor park, as well as the

content and order of technological operations. The

problem of an optimal selection of agricultural

machinery is considered in [36].

The development of software for a choice of

technologies and machinery in crop production is

considered and the requirements for the developed

software, its main components, their functions, and

rules of communication, using cloud technologies

are given.

The general structural scheme for choosing of

technologies and the rational compound of the

machine and tractor fleet is proposed in [38]. The

scheme provides for taking into account the main

restrictions imposed by the agro-climatic and

production conditions of the agricultural producer

(the scope of the work and their timing,

phytosanitary conditions, relief, and a contour of

fields).

In the article [38], a temporal data model has

been developed that makes it possible to draw up

daily work plans for the selected equipment and to

calculate the economic indicators of mechanized

tillage. The developed data model is integrated with

the geo-database and with the database of

agricultural machinery of the farm.

The scientific and practical center of the

National Academy of Sciences of Belarus for

agricultural mechanization has developed the

equipment and software for a remote monitoring

system for machine and tractor units, including a

telemetry module, an identification module, fuel

sensors, a server, and user software. The system is

designed to determine the coordinates, direction,

and speed of the machine-tractor unit. The system

allows one to determine the composition of the unit,

the cultivated area, and fuel consumption.

A prototype of an onboard computer for

tractors Belarus 3022/3522 with a navigation

module was developed to determine the current

coordinates while moving with an accuracy of up to

10 cm. The computer allows one to control more

than 15 tractor operating parameters and

automatically guide the unit along a given trajectory

with an accuracy of one cm. Studies have shown

that the optimization of the operating modes of

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.41

Natalja M. Matsveichuk, Yuri N. Sotskov

E-ISSN: 2224-2872

367

Volume 22, 2023

high-performance units will increase their

productivity by 5-10% and reduce specific fuel

consumption by up to 10%.

An automated system for analyzing and

monitoring the status indicators of heavy vehicles

using machine learning has been proposed. The

signals from the sensors are sent via the CAN bus to

the onboard computer and are wirelessly transmitted

to the server to monitor the parameters and

determine the transition to a critical state.

A scheme for a digital control system for

agricultural machinery based on the IoT, cloud, big

data, and AI technologies is proposed in [39]. The

authors presented an algorithm by which the control

system for each actuator of an agricultural machine

operates; Figure 3.

Fig. 3: The schematic diagram of controlling

technical means in crop production, [39]

The concept of a new generation of IoT (called

IntellIoT) has been proposed. The IntellIoT project

aims to develop a framework for managing smart

IoT systems and their applications. An example of

the use of a smart IoT in agriculture for autonomous

management of a fleet of agricultural machines

based on IntellIoT is given.

We next list the developments of Russian

companies operating in the Internet of Things

technology market for monitoring and managing

technical means. Tibbo Systems, a leading Russian

developer of software for control and monitoring

systems, has developed a unified aggregate IoT

platform that provides monitoring of vehicles and

agricultural equipment (cars, tractors, and

combines), management of sorting, storage, and

processing of raw materials (the automatic

recognition forklift trajectories to determine work

intervals and calculate the weight of the supplied

raw materials), monitoring the storage conditions of

raw materials. The mobile operator MTS specializes

in transport monitoring. Its developments may be

useful for tracking agricultural machinery and

commercial vehicles in the logistics of agricultural

products. The mobile operator MegaFon has

launched NB-IoT technology, which is expected to

be widely used in the agro-industrial complex.

3.6 Mapping and Geodesy

The studies of land resources of the regions are

considered in many papers on the example of the

Volgograd region, and the southern seas of Russia,

the Brest region in Belarus. Using story map

templates from the ArcGIS online cloud mapping

platform, the information and analytical system

"Land Fund of the Brest Region" and the "Atlas of

the State of the lands of the Brest Region" were

developed. These systems contain structured

information about land types, analysis and

assessment of their current state, dynamics of

nature-forming land types in the Brest region, and a

comprehensive geo-ecological assessment of the

region's land resources.

The experience of using GIS technologies to

visualize data on the content of micro-plastic

particles in water reservoirs of Brest is analyzed.

The result of this study is a series of interactive

hydrographic maps of the city. These maps are

freely available on the Internet, can be viewed by

many users, and can be used to create similar maps

and map schemes using an ArcGIS Online account.

To assess agricultural land, spatial databases

based on the object-functional approach were

developed. The necessity of a practical

implementation of agronomic geo-databases (Agro-

GIS) with a hierarchical structure based on

databases of local and regional levels is shown. The

filling of the geo-database at the local level is

provided by the inclusion of objects associated with

agricultural workers, land plots, agricultural

implements, soils, technological maps, and tractors.

The main components of the proposed geo-

database are separate sets of spatial classes (climate,

relief, soils, vegetation, hydrographs, agro

landscapes). Three different ways of user interaction

with the Agro-GIS database have been developed.

The integration of geo-information databases with

agricultural machinery databases for making

decisions on the optimal choice of equipment, the

choice of technological operations, solving logistics,

and other practical problems is studied.

Within the framework of the GIS project, ready-

made solutions for the cloud database and GIS

applications of the local information system

"Ecological Study of the Southern Seas of Russia"

are presented. When generating information for

entering into the geo-database, topographic maps at

a scale of 1:1000000, satellite images prepared by

Monitoring systems (Satellite imagery

of UAV’s, meteorological stations)

Database

Tractor

Cultivator

Seeder

Sprayer

Control

system (AI)

Vehicle

Adapters

(header, pick-up

unit, digger)

Self-propelled

unit (harvester)

Data

Control signal (instruction)

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.41

Natalja M. Matsveichuk, Yuri N. Sotskov

E-ISSN: 2224-2872

368

Volume 22, 2023

Landsat-7 ETM+, Sentinel-2, and statistical

calculation data were used.

The researches in the field of organic agriculture

were fulfilled in the Republic of Belarus; [36]. The

features of the production and circulation of organic

products, the current state of development of the

industry, as well as the prospects for the

development of geo-information products as active

means of electronic inventory of individuals and

legal entities engaged in economic activities are

considered. Geo-information products have been

created and developed in the form of web maps,

web-passports, electronic databases, and web-

catalogs.

3.7 Digital Platforms

A digital transformation of the economy requires

replacing or upgrading production equipment to

digital ones. This process is quite complex and very

expensive. For effective management of production

processes in agriculture, it is necessary to have

objective and reliable information about the

characteristics, parameters, and state of

technological processes. It is required to constantly

monitor the parameters and physical quantities of

soil and climatic resources, cultivated plants, farm

animals, machines, and the environment.

The unified IoT Platform Aggregate has been

developed to automate many aspects of agricultural

enterprises to improve efficiency and financial

performance. This platform provides monitoring of

vehicles and agricultural machinery (cars, tractors,

and combines), management of sorting, storage, and

processing of raw materials (automatic recognition

of forklift trajectories to determine operating

intervals and calculations of the weight of the

supplied raw materials), monitoring of storage

conditions of raw materials.

Promising is the use of artificial intelligence

methods in precision agriculture, and the issues of

integrating these methods into a single digital

platform; [36]. An approach based on AIoT

(Artificial Intelligence of Things) allows one to

automate the full cycle of the agricultural work

related to crop and livestock production. In this

case, hardware elements play an important role:

sensors, communication channels, and AIoT. In

some cases, the AIoT platform and its application

are a single entity. The most common area of this

application in the agricultural sector is precision

farming. The concept of a new generation of

Internet of Things using AI has been developed.

The adaptation of the IBM Watson Decision

Platform for agriculture to improve the efficiency of

agriculture in Russia was developed. The main

element of this platform is the PAIRS Geo-scope

system from IBM research, which quickly processes

massive complex sets of geospatial and temporal

data collected by satellites, drones, millions of IoT

sensors, and weather models. This platform allows

the integrating of the weather company data, remote

sensing data (such as a satellite), and IoT data from

tractors. It can be used to analyze hyper-local

weather forecasts for real-time recommendations

based on specific agricultural fields or crops.

The integrated cloud service called ANT, which

was created on the Geo-Look platform and intended

for agricultural enterprises, is considered. The ANT

is a tool kit for precision farming. The basis of the

service is the electronic contours of the agricultural

fields. Each user can create them in the service or

upload existing ones in the database. By adding data

from agrochemical measurements, the information

system obtains accurate maps of the distribution of

elements in the soil, identifies heterogeneous soil

areas, and determines their need for fertilizers.

Agricultural units of differentiated fertilizer

applications receive individual tasks from the

system. Meteorological data and the serviceability

of the equipment are also monitored to minimize the

human factor and errors in the preparation and

conduct of agricultural operations. The system

allows one to keep records, optimize the plan of

work, predict yields, use interactive dashboards to

monitor the progress of sowing and harvesting in

real-time, track deviations from the plan, and see the

causes of deviations and factors affecting the final

results.

Since the use of digital technologies, the Internet

of things, and artificial intelligence is becoming a

significant condition for competitive agricultural

production, the main direction of state support is to

create 2030 a unified digital platform for making

operational decisions, as well as forecasting and

modeling the development of the agro-industrial

complex.

Concluding this section, we note that in animal

farming and greenhouses, research is carried out

mainly on creating an IoT management system.

Most of the researches deal with dairy farms. In

crop production, water management and irrigation,

more attention is paid to decision-making systems,

and corresponding algorithms and models being

developed. In machinery management, researchers

focused on equipment and software development.

The applicability of GIS systems for mapping

agricultural land is explored. When considering the

problem of connecting disparate control systems for

various enterprises and various technological

processes in agriculture, approaches are proposed

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.41

Natalja M. Matsveichuk, Yuri N. Sotskov

E-ISSN: 2224-2872

369

Volume 22, 2023

for creating unified digital decision-making

platforms.

4 Promising Research Directions

The potential value of modern Internet of Things

technologies for farmers is combined with

management problems in their application.

Management of technological processes of

agricultural production must be based on the

analysis of large data sets (such as data on

production volumes, data from weather stations,

agro-ecological surveys, field passports, data on

agricultural field contours, a crop rotation, acreage,

and crop, data on the state of the herd, a veterinary

condition, product traceability, telemetry data on the

state of agricultural machinery, agrochemical

surveys and product quality control parameters).

As noted in [16], the integration of

heterogeneous data from different sensors used in

smart farming systems is essentially difficult due to

software and hardware compatibility issues.

The Internet of Things, being a network of small

and remotely located objects, needs very limited

resources in terms of data processing and storage.

The quality and cost of devices and sensors, and the

reliability of the system, are the main challenges for

smallholder farmers seeking to implement advanced

technologies, [4].

In addition, since IoT devices are

heterogeneous, there is a problem of device

compatibility and synchronization for better

performance, since primary analysis or pre-

processing of data may not be sufficient to store

data from different sources, [13].

There is an urgent need to upgrade IoT devices

to improve their reliability, endurance,

intellectualization, etc. while reducing costs and

operational difficulties. The most common factors

hindering the widespread implementation of

information and communication systems and agro-

technologies of reclaimed agriculture in agricultural

production are the lack of proper development of the

Internet in rural areas and the low motivation of

agricultural producers to use digital solutions, [37].

Potential applications of the Internet of Things

in smart agriculture include the development of

smart agricultural machines, irrigation systems,

weed, and pest control, fertilization, the use of crop

protection unmanned aerial vehicles (UAVs), crop

health monitoring, etc., as well as questions data

security and safety, [4], [6].

An agricultural production is carried out under

the influence of many uncertain factors that cannot

be predicted and which a person cannot influence

[36]. It is necessary to take into account the

uncertainty inherent in agriculture, both when

setting tasks for planning agricultural production,

and when searching for effective solutions to

management problems that arise in the process of

agricultural production. Note that the use of

stochastic approaches or fuzzy logic approaches

may be unjustified, for example, due to the

unknown distribution laws of the stochastic

parameters.

It would be appropriate in this case to use the

stability approach, [40], [41], [42], which allows

one to determine the range of changes in the given

initial data that does not lead to a change in the

optimal solution.

In future research, one can apply the stability

approach to combat the uncertainty that often arises

in agricultural production. This will mean, e.g.,

determining the optimal list and order of agricultural

operations, which will remain unchanged, despite

the uncertainty of the agricultural job durations. At

the same time, the schedule for the execution of

works will vary, depending on the weather, sensor

data, and other uncertain factors. Combining this

approach with the Internet of Things and cloud

computing will improve the quality and quantity of

smart agricultural production.

5 Conclusions

In smart agriculture centralized integrated

processing of information coming from sensors

must be carried out online and must be used for an

operational management, adaptation, and evolution

of the control system with a subsequent correction

of the parameters of mathematical models for

solving problems of higher levels of management.

This will increase production productivity,

reduce the shortage of skilled labour, simplify the

delivery of the final product to the buyer, provide

manufacturers with the required information, predict

natural disasters, and take into account climate

change, which will make it possible to minimize

possible risks and losses.

Precision farming and the Internet of Things are

not only technologies that can help increase yields,

but they are mainly dedicated to optimizing

resources and the sustainability of the agro-

ecosystems. Using these technologies in closed

agro-ecosystems, including vertical farms, off-soil

plant growing, and robotic crop complexes, in the

future will allow reaching self-sufficiency in food in

areas such as megalopolis, settlements in the Far

North, and in deserts.

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.41

Natalja M. Matsveichuk, Yuri N. Sotskov

E-ISSN: 2224-2872

370

Volume 22, 2023

Since agricultural production is carried out

under the influence of many uncertain factors, it is

necessary to take these uncertain factors into

account both when setting tasks for planning

agricultural production and when searching for

effective solutions to management problems that

arise in the process of agricultural production.

Important tasks of smart agriculture are

associated with the need to ensure sustainable

agricultural production, improve mathematical

modeling of agricultural production, and forecast

economic indicators of the agricultural production.

Using agricultural technologies 4.0 will allow

one to move from managing technological processes

and installations to managing the profitability of the

agricultural enterprise as a whole, which, in addition

to the economic effect, will improve the working

conditions and prestige of agricultural production

specialists.

The above study shows that agriculture in

Russia and Belarus is rapidly developing towards

digitalization, Internet of Things and other

innovative technologies are spreading. The main

sectors of agricultural production in which this

development is most intensive are listed, and the

difficulties encountered in the digitalization of

agriculture in these countries are identified.

This survey shows the development of Internet

of Things technologies in Russia and Belarus

compared to other countries of the world and may

be useful for further research on the dissemination

of innovative digital technologies in global

agriculture.

In the future, it will be promising to apply

stability approach, [40], [41], [42] to combat the

uncertainty that usually arises in agricultural

production. Combining the stability approach with

the Internet of Things and cloud computing may

improve the quality and quantity of the agricultural

production.

Acknowledgement:

This research was funded by the Belarusian

Republican Foundation for Fundamental Research,

grant number Φ23PHФ-017.

References:

[1] Korotchenya, V.M., Lichman, G.I., Smirnov,

I.G., Digitalization of technological processes

of crop production in Russia, Agricultural

Machinery and Technologies, Vol. 13, No. 1,

2019, pp. 14–20.

[2] Raj, M., Gupta, S., Chamola, V., Elhence, A.,

Garg, T., Atiquzzaman, M., Niyato, D., A

survey on the role of Internet of Things for

adopting and promoting Agriculture 4.0,

Journal of Network and Computer

Applications, No. 187, 2021, 103107.

[3] Maroli, A., Narwane, V.S., Gardas, B.B.,

Applications of IoT for achieving

sustainability in agricultural sector: A

comprehensive review, Journal of

Environmental Management, No. 298, 2021,

113488.

[4] Subeesh, A., Mehta, C.R., Automation and

digitization of agriculture using artificial

intelligence and internet of things, Artificial

Intelligence in Agriculture, No. 5, 2021, pp.

278–291.

[5] Moysiadis, V., Sarigiannidis, P., Vitsas, V.,

Khelifi, A., Smart farming in Europe,

Computer Science Review, No. 39, 2021,

100345.

[6] Sinha, B.B., Dhanalakshmi, R., Recent

advancements and challenges of Internet of

Things in smart agriculture: A survey, Future

Generation Computer Systems, No. 126,

2022, pp. 169–184.

[7] Čolaković, A., Hadžialić, M., Internet of

Things (IoT): A review of enabling

technologies, challenges, and open research

issues, Computer Networks, No. 144, 2018,

pp. 17–39.

[8] Patel, C., Doshi, N., A novel MQTT security

framework in generic IoT model, Procedia

Computer Science, No. 171, 2020, pp. 1399–

1408.

[9] Idoje, G., Dagiuklas, T., Iqbal, M., Survey for

smart farming technologies: Challenges and

issues, Computers and Electrical Engineering,

No. 92, 2021, 107104.

[10] Tao, W., Zhao, L., Wang, G., Liang, R.,

Review of the internet of things

communication technologies in smart

agriculture and challenges, Computers and

Electronics in Agriculture, No. 189, 2021,

106352.

[11] O'Grady, M.J., Langton, D., O'Hare, G.M.P.,

Edge computing: A tractable model for smart

agriculture, Artificial Intelligence in

Agriculture, No. 3, 2019, pp. 42–51.

[12] Tzounis, A., Katsoulas, N., Bartzanas, T.,

Kittas, C., Internet of Things in agriculture,

recent advances and future challenges,

Biosystems Engineering, No. 164, 2017, pp.

31–48.

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.41

Natalja M. Matsveichuk, Yuri N. Sotskov

E-ISSN: 2224-2872

371

Volume 22, 2023

[13] Halgamuge, M.N., Bojovschi, A., Fisher,

P.M.J., Le, T.C., Adeloju, S., Murphy, S.,

Internet of Things and autonomous control for

vertical cultivation walls towards smart food

growing: A review, Urban Forestry & Urban

Greening, No. 61, 2021, 127094.

[14] Debauche, O., Trani, J.-P., Mahmoudi, S.,

Manneback, P., Bindelle, J., Mahmoudi, S.A.,

Guttadauria, A., Lebeau, F., Data

management and internet of things: A

methodological review in smart farming,

Internet of Things, No. 14, 2021, 100378.

[15] Debauche, O., Mahmoudi, S., Manneback, P.,

Lebeau, F., Cloud and distributed

architectures for data management in

agriculture 4.0: Review and future trends,

Journal of King Saud University – Computer

and Information Sciences, Vol. 34, No. 9,

2022, pp. 7494–7514.

[16] Boursianis, A.D., Papadopoulou, M.S.,

Diamantoulakis, P., Liopa-Tsakalidi, A.,

Barouchas, P., Salahas, G., Karagiannidis, G.,

Wan, S., Goudos, S.K., Internet of Things

(IoT) and agricultural unmanned Aerial

Vehicles (UAVs) in smart farming: A

comprehensive review, Internet of Things, No.

18, 2022, 100187.

[17] Lezoche, M., Hernandez, J.E., del Mar Eva

Alemany Díaz, M., Panetto, H., Kacprzyk, J.,

Agri-food 4.0: A survey of the supply chains

and technologies for the future agriculture,

Computers in Industry, No. 117, 2020,

103187.

[18] Nkamla Penka, J, B., Mahmoudi, S.,

Debauche, O., A new kappa architecture for

IoT data management in smart farming,

Procedia Computer Science, No. 191, 2021,

pp. 17–24.

[19] Hsu, T.-C., Yang, H., Chung, Y.-C., Hsu, C.-

H., A Creative IoT agriculture platform for

cloud fog computing, Sustainable Computing:

Informatics and Systems, No. 28, 2020,

100285.

[20] Verdouw, C., Sundmaeker, H., Tekinerdogan,

B., Conzon, D., Montanaro, T., Architecture

framework of IoT-based food and farm

systems: A multiple case study, Computers

and Electronics in Agriculture, No. 165, 2019,

104939.

[21] Hernández-Morales, C.A., Luna-Rivera, J.M.,

Perez-Jimenez, R., Design and deployment of

a practical IoT-based monitoring system for

protected cultivations, Computer

Communications, No. 186, 2022, pp. 51–64.

[22] Rodríguez, J.P., Montoya-Munoz, A.I.,

Rodriguez-Pabon, C., Hoyos, J., Corrales,

J.C., IoT-Agro: A smart farming system to

Colombian coffee farms, Computers and

Electronics in Agriculture, No. 191, 2021,

106442.

[23] Zamora-Izquierdo, M.A., Santa, J., Martínez,

J.A., Martínez, V., Skarmeta, A.F., Smart

farming IoT platform based on edge and cloud

computing, Biosystems Engineering, No. 177,

2019, pp. 4–17.

[24] Glaroudis, D., Iossifides, A., Chatzimisios, P.,

Survey, comparison and research challenges

of IoT application protocols for smart

farming, Computer Networks, No. 168, 2020,

107037.

[25] Ferraz Jr., N., Silva, A.A.A., Guelfi, A.E.,

Kofuji, S.T., Privacy-preserving cloud-

connected IoT data using context-aware and

end-to-end secure messages, Procedia

Computer Science, No. 191, 2021, pp. 25–32.

[26] Freitas Bezerra, D., de, Medeiros, V.W.C., de,

Gonҫalves, G.E., Towards a control-as-a-

service architecture for smart environments,

Simulation Modelling Practice and Theory,

No. 107, 2021, 102194.

[27] Ramli, M.R., Daely, P.T., Kim, D.-S., Lee,

J.M., IoT-based adaptive network mechanism

for reliable smart farm system, Computers

and Electronics in Agriculture, No. 170, 2020,

105287.

[28] Fastellini, G., Schillaci, C., Precision farming

and IoT case studies across the world, In:

Agricultural Internet of Things and Decision

Support for Precision Smart Farming,

Castrignano, A., Buttafuoco, G., Khosla, R.,

Mouazen, A., Moshou, D., Naud, O., Eds.,

Academic Press, 2020, pp. 331–415.

[29] Da Rosa Righi, R., Goldschmidt, G., Kunst,

R., Deon, C., da Costa, C.A., Towards

combining data prediction and internet of

things to manage milk production on dairy

cows, Computers and Electronics in

Agriculture, No. 169, 2020, 105156.

[30] Somov, A., Shadrin, D., Fastovets, I., Nikitin,

A., Matveev, S., Seledets, I., Hrinchuk, O.,

Pervasive agriculture: IoT-enabled

greenhouse for plant growth control, IEEE

Pervasive Comput, Vol. 17, No. 4, 2018, pp.

65–75.

[31] Doutoum, A. S., Eryigit, R., Tugrul, B.,

Classification of Guava Leaf Disease using

Deep Learning, WSEAS Transactions on

Information Science and Applications, vol. 20,

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.41

Natalja M. Matsveichuk, Yuri N. Sotskov

E-ISSN: 2224-2872

372

Volume 22, 2023

2023, pp. 356-363,

DOI:10.37394/23209.2023. 20.38.

[32] Elaraby, W. S., Madian, A. H., Meta-heuristic

Optimization Algorithms for Irradiated Fruits

and Vegetable Image Detection, WSEAS

Transactions on Computers, Vol. 21, 2022,

pp. 118-130,

https://doi.org/10.37394/23205.2022.21.17.

[33] Cordeiro, M., Markert, C., Araújo, S.S.,

Campos, N, G.S., Gondim, R.S., Silva, T, L,

C, Da, Rocha, A, R, Da, Towards smart

farming: Fog-enabled intelligent irrigation

system using deep neural networks, Future

Generation Computer Systems, No. 129,

2022, pp. 115–124.

[34] Benyezza, H., Bouhedda, M., Rebouh, M.,

Zoning irrigation smart system based on fuzzy

control technology and IoT for water and

energy saving, Journal of Cleaner

Production, No. 302, 2021, 127001.

[35] Lavanya, G., Rani, C., Ganeshkumar, P., An

automated low cost IoT based fertilizer

intimation system for smart agriculture,

Sustainable Computing: Informatics and

Systems, No. 28, 2020, 100300.

[36] Matsveichuk, N.M., Sotskov, Yu.N. Using

digital technologies for the development of

agriculture in Russia and Belarus. Economics,

modeling, forecasting, Vol. 17, 2023, pp. 94-

108, (in Russian).

[37] Yurchenko, I.F., Assessment of the current

state of the industry of digitalization of land

reclamation, Nature Management, No. 2,

2022, pp. 6–12, (in Russian).

[38] Efremov, A.A., Sotskov, Y.N., Belotzkaya,

Y.S., Optimization of selection and use of a

machine and tractor fleet in agricultural

enterprises: A case study, Algorithms, No. 16,

2023, No. 311, pp. 1–22.

[39] Starostin, I.A., Belyshkina, M.E.,

Chilingaryan, N.O., Alipichev, A.Yu., Digital

technologies in agricultural production:

implementation background, current state and

development trends, Agricultural

Engineering, Vol. 3, No. 103, 2021, pp. 4–10.

[40] Sotskov, Y.N., Egorova, N.G., Single

machine scheduling problem with interval

processing times and total completion time

objective, Algorithms, Vol. 11(5), No. 66,

2018, pp. 1–30.

[41] Sotskov, Y.N., Matsveichuk, N.M., Hatsura,

V.D., Two-machine job-shop scheduling

problem to minimize the makespan with

uncertain job durations, Algorithms, Vol.

13(5), No. 4, 2020, pp. 1–45.

[42] Sotskov, Y.N., Lai, T.-C., Werner, F.,

Measures of problem uncertainty for

scheduling with interval processing times, OR

Spectrum, Vol. 35, No. 3, 2013, pp. 659–689.

Contribution of Individual Authors to the

Creation of a Scientific Article (Ghostwriting

Policy)

The authors equally contributed to the present

research, at all stages from the formulation of the

problem to the final findings and solution.

Sources of Funding for Research Presented in a

Scientific Article or Scientific Article Itself

This work was partially supported by the Belarusian

Republican Foundation for Fundamental Research

(project No. Φ23PHФ-017).

Conflict of Interest

The authors have no conflicts of interest to declare.

Creative Commons Attribution License 4.0

(Attribution 4.0 International, CC BY 4.0)

This article is published under the terms of the

Creative Commons Attribution License 4.0

https://creativecommons.org/licenses/by/4.0/deed.en

_US

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.41

Natalja M. Matsveichuk, Yuri N. Sotskov

E-ISSN: 2224-2872

373

Volume 22, 2023