PreprintPDF Available

Swarm Robots in Agriculture

Preprints and early-stage research may not have been peer reviewed yet.

Abstract and Figures

Agricultural mechanization is an area of knowledge that has evolved a lot over the past century, its main actors being agricultural tractors that, in 100 years, have increased their powers by 3,300%. This evolution has resulted in an exponential increase in the field capacity of such machines. However, it has also generated negative results such as excessive consumption of fossil fuel, excessive weight on the soil, very high operating costs, and millionaire acquisition value. This paper aims to present an antiparadigmatic alternative in this area. It is proposing a swarm of small electric robotic tractors that together have the same field capacity as a large tractor with an internal combustion engine. A comparison of costs and field capacity between a 270 kW tractor and a swarm of ten swarm tractors of 24 kW each was carried out. The result demonstrated a wide advantage for the small robot team. It was also proposed the preliminary design of an electric swarm robot tractor. Finally, research challenges were suggested to operationalize such a proposal, calling on the Brazilian Robotics Research Community to elaborate a roadmap for research in the area of swarm robot for mechanized agricultural operations.
Content may be subject to copyright.
Swarm Robots in Mechanized Agricultural Operations: Roadmap for Research
Daniel Albiero, D. Sc.*. Angel Pontin Garcia, D. Sc.*. Claudio Kiyoshi Umezu, D. Sc.*. Rodrigo Leme de Paulo, Eng.*
*School of Agricultural Engineering, University of Campinas, Campinas-Brazil, ZIPCODE: 13083-875
Brazil (Tel: 55-19-3521-1024; e-mail:,,,
Abstract: Agricultural mechanization is an area of knowledge that has evolved a lot over the past century,
its main actors being agricultural tractors that, in 100 years, have increased their powers by 3,300%. This
evolution has resulted in an exponential increase in the field capacity of such machines. However, it has
also generated negative results such as excessive consumption of fossil fuel, excessive weight on the soil,
very high operating costs, and millionaire acquisition value. This paper aims to present an anti-
paradigmatic alternative in this area. It is proposing a swarm of small electric robotic tractors that together
have the same field capacity as a large tractor with an internal combustion engine. A comparison of costs
and field capacity between a 270 kW tractor and a swarm of ten swarm tractors of 24 kW each was carried
out. The result demonstrated a wide advantage for the small robot team. It was also proposed the
preliminary design of an electric swarm robot tractor. Finally, research challenges were suggested to
operationalize such a proposal, calling on the Brazilian Robotics Research Community to elaborate a
roadmap for research in the area of swarm robot for mechanized agricultural operations.
Resumo: A mecanização agrícola é uma área de conhecimento que evoluiu no decorrer do último século,
seus principais atores são os tratores agrícolas que, em um período de 100 anos, aumentaram suas potências
em até 3.300%. Esta evolução se traduziu em um aumento exponencial da capacidade operacional de tais
máquinas, mas também gerou resultados negativos, tais como: consumo excessivo de combustível fóssil,
peso exagerado sobre o solo, custos operacionais muito altos e valor de aquisição milhonário. Este paper
pretende apresentar uma alternativa anti-paradigmática nesta área, propondo um enxame de pequenos
tratores robóticos elétricos que juntos tenham a mesma capacidade operacional que um grande trator
agrícola convencional. Foi realizado um comparativo de custos e de capacidade operacional entre um trator
de 270 kW e um enxame de dez tratores robóticos de 24 kW cada, o resultado demonstrou ampla vantagem
para o time de pequenos robôs. Também foi proposto o ante-projeto de um trator robô elétrico e finalmente
foram sugeridos desafios de pesquisa para operacionalizar tal proposta, conclamando a comunidade
pesquisadora em robótica brasileira em elaborar um roadmap para pesquisa na área de swarm robot para
operações agrícolas mecanizadas.
Keywords: Field capacity; Operational cost; Electric tractor; Artificial Intelligence; Plowing.
Palavras-chaves: Capacidade de campo; Custos; Trator elétrico; Inteligência Artificial; Aração.
Agricultural mechanization is the area of knowledge in
agribusiness wich has the highest energy expenditure and the
highest aggregate cost in agricultural production, reaching
60% of energy consumption according to (D Albiero, 2011).
This fact occurs due to the specificities of farming operations
that requires a lot of energy in the mechanical form (Goering
& Hanson, 2004) referring to the different phases of
agricultural production: soil preparation, seeding, planting,
crop management, harvesting and conditioning of crop
This energy is from power sources known as agricultural
tractors (Goering et al., 2003), which enables the operation of
plows, harrows, seeders, harvesters, sprayers, brush cutters,
chisels, subsoilers, crushers, conditioners, rakes, terriers,
planters, cutters, etc. (Srivastava et al., 2006). Since the
appearance of the agricultural tractor at the end of the 19th
century and the beginning of the 20st century, the power and
weight of these machines have tended to increase, due to the
need of improving their field capacity in the area (Goering &
Hanson, 2004). For comparison, at the beginning of the 20th
century, the largest tractors had approximately 15 kW of
power (Renius, 2020). Today, at the beginning of the 21st
century, we have reached a point of scale between the power
and field capacity of agricultural tractors in the 500 kW range
(Deere, 2020). There is a consensus in contemporary literature
that the power growth curve of these machines is stabilizing
and reaching an asymptotic limit. This trend is approaching a
technological limit for parameters that represent three crucial
problems. The first is the excessive energy consumption of
large tractors that consume a lot of fossil diesel fuel (up to 150
liters per hour) (ASABE, 2013); the second refers to the weight
DOI: 10.48011/asba.v2i1.1144
of these machines, which increased from about 1300 kg in
1902 to 25000 kg in 2019 (Renius, 2020). This weight increase
is necessary for the traction generated by the machine to be
used, since a light tractor with great power would skate, sliding
on the agricultural soil, which represents loss of efficiency
(Macmillan, 2002).
Today there are tractors weighing more than 25 tons and this
fact generates a very significant degradation of the soil in
physical-mechanical terms which is translated into soil
compaction, a phenomenon that reduces the infiltration of
water in the soil, increases the force necessary for seedlings to
emerge and it does not allow plant roots to go deeper (Kiehl,
1979), all of these facts represent losses in food production;
and the third not least is the investment cost of these machines,
which in this category (500 kW), reach values of US $
550,000.00 (TractorHouse, 2020)
In this context, an exciting hypothesis is to change the current
paradigm in agricultural mechanization to increase the field
capacity of tractors by increasing their power and weight. This
paper proposes a roadmap for research in the opposite
direction: to decrease the power and weight of the tractors and
to increase their number, optimizing the agricultural
operations in terms of logistics, operational geometry, and
energy efficiency: instead of using a gigantic machine of 500
kW, use 20 small tractor 25 kW.
However the problem with this anti-paradigmatic approach.
He comes up against the current socio-economic situation of
agricultural fields in western nations (Brazil among them) (D.
Albiero et al., 2019; D. Albiero et al., 2015; D. Albiero, 2019):
Tractor operators are scarce, and their costs (wages, charges,
taxes, training, and insurance) are relatively high. Thus, a very
suitable solution offered by the science of robotics is the
operationalization of these small tractors as multiple robots
operating in a swarm configuration. To economically justify
such a solution and present the literature concerning swarm
robots for agriculture.
The main objective of this paper is to propose that researchers
of the Brazilian Robotics Research Community develop a
roadmap for future research aiming to operate the use in the
agricultural field of swarm robots be developed in a
commercial, concrete and practical way focused on
mechanized agricultural operations.
The contribution of this paper is to foster discussion about the
commercial implementation of swarm robots for mechanized
agricultural operations, generating discussions and perhaps
initiating fruitful interactions and transdisciplinary
partnerships between the robotic research community and
researchers in the field of agricultural mechanization.
Robots are not new in agriculture; there is much research being
developed, some of them very advanced, and already with real
applications in the field, agricultural robotics is an
overwhelming trend (D. Albiero, 2019). Wolfert et al. (2017)
describe these advances in Agriculture 4.0, which in farming
is called Smart Farming. They explain that smart machines and
crop sensors on farms have obtained large amounts of
agricultural data and that the quantity, quality, and scope have
grown enormously, which makes data available to improve
processes. In this context, innovations in the field are
developing at an accelerated rate. (Bechar & Vigneault, 2016).
There are robots for the application of phytosanitary products;
for sowing; for diagnosis of soil, plants, water; with computer
vision systems; for harvest; with remote steering control
systems; with transplant systems; for weed control; for
monitoring diseases and pests; for pruning (Bechar &
Vigneault, 2017).
An exciting innovation in the Smart Farms concept was a robot
for irrigating pots in agricultural greenhouses. It uses sensors
for humidity, position, and computer vision to assess how
much each plant, individually, needs water and then performs
the necessary water slide for each plant. This system makes it
possible to save water and substantially improve irrigation
efficiency (Araújo Batista et al., 2017). Xaud et al. (2018)
developed an interesting robot for use in bioenergetic crops,
De Lemos et al. (2018) present a uni-sensor strategy for
navigation between rows of crops for robots and Oliveira et al.
(2018) presented a methodology to adapt conventional
commercial systems to autonomous robotic systems.
Davis (2012) described a family of agricultural vehicles that
has collective sensing and computational infrastructure. There
is an exciting European research program that deeply studies
applications of swarm robotics concepts with UAVs used to
obtain information on the productivity of beet fields and to
generate data on weeds, diseases, and nematodes (Toorn,
2020). Albani et al. (2019) use UAVs swarm robots to monitor
and map weeds in agricultural fields. Albani et al. (2017)
presented an exciting roadmap for future studies on swarm
robotics for applications in farm monitoring and mapping.
Mukherjee et al. (2020) studied the challenges in
operationalizing the use of UAVs in swarm robotics
configuration to operate decentrally and heterogeneously in
the agricultural environment, which has very variable and non-
trivial control edges.
Shamshiri et al. (2018) have made an extensive literature
review on agricultural robotics; its challenges and special
attention is given to multi-robots and swarm methods.
(Blender et al., 2016) introduced Mobile Agricultural Robot
Swarms (MARS) is an approach for autonomous farming
operations by a coordinated group of robots and describes an
application in seeding. Trianni et al. (2016) described the
concept of a set of swarm robots for weed control and define a
roadmap for the execution of such a project. Minßen et al.
(2017) presented conceptual studies for agricultural care in
plants considering several swarm robots.
The aforementioned papers present the current state of the art
on the theme of swarm robots for agricultural applications,
when analyzing the issues of problems studied and solved by
the authors it is noticed that there is still much to be
accomplished, many important solutions for the commercial
operationalization of a swarm of robots operating in the field
has by no means been resolved, in this context there is a huge
opportunity for developments, which is very motivating.
To carry out the operational and cost comparison between the
configuration of small power swarm robots and a large tractor,
one of the most power demanding agricultural operations was
chosen: deep plowing in loose clayey soil with a moldboard
plow. According to the ASABE D497 standard (ASABE,
2013), the request for the tractive force for such an operation
is given by (1):
 = . + . + . . .  (1)
Where: D is the tensile strength of the implement (N);
A, B, C are dimensionless coefficients tabulated by
the D497 standard;
S is the travel speed (km s-1);
W is the cutting width of the implement (m);
T is the cutting depth of the implement (cm).
In this paper, a plow of 5 molds by Marchesan, model ARR2,
was chosen for the large tractor, in double composition by a
tandem header, making ten molds, cutting width of 4 m and
cutting depth of 0.35 m. For the small robot, the same plow
with one moldboard was chosen, which configures a cutting
width of 0.40 m and depth of cut of 0.35 m.
According to the D497 standard, the typical displacement
speed of this plow is 5 km/s, the dimensionless coefficients are
A = 652; B = 0; C = 5.1. For soil with a clay texture, the Fi
factor is 1. The following data are obtained:
D10aivecas=109,130.00 N; D1aiveca=10,913.00 N.
Considering loose soil, the large tractor with the power source
from the internal combustion diesel engine coupled to a
mechanical transmission system in the MFWD system,
according to the D497 standard, has an overall efficiency in
the transfer of tractor/plow power of 53.9%. In this work,
agricultural swarm robots will be considered as a result of
research by Melo et al. (2019) and Vogt (2018) that designed
and dimensioned a small electric tractor powered by
electrochemical power from batteries and electric engines
direct drive in tracks on a rubber mat. In this configuration, the
overall efficiency in the transfer of tractor/plow power is
The equation (2) gives the nominal power required in tractor
engines (conventional for the large tractor and electric tractor
for the small swarm robot tractor):
 =.
Where: D is the tensile strength (N);
S is the travel speed (m s-1)
η is the overall tractor/implement efficiency
So, we have the following data:
Pnom (Large Tractor) = 281.20 kW;
Pnom (Swarm Tractor) =19.83 kW;
The John Deere 8730R large tractor was selected ((Deere,
2020). With a nominal power of 272 kW and a maximum
power of 300 kW. Maximum weight with ballast of 19.805 kg.
For the Swarm Electric Robot Tractor (TRSE)
recommendations of (Melo et al., 2019; Vogt, 2018; Vogt et
al., 2018), the drivetrain being sized with two 10 kW electric
motors each coupled to the wheels, the power source comes
from a quick replacement pack consisting of 4 stationary lead-
acid batteries 12V/220 Ah per battery, 2.5 hours autonomy for
each pack and total weight of the 700 kg electric swarm robot
Considering the operational efficiency of the tractor/plow set
of 70% (ASABE, 2013), the field capacity of the set can be
calculated by (3):
 = ..
 (3)
Where: Cc is the field capacity (ha h-1);
S is the travel speed (km h-1);
W is the cutting width (m);
ef is the operational efficiency (decimal).
We have the following data:
Cc (Large Tractor) = 1.4 ha h-1;
Cc (Swarm Tractor) = 0.14 ha h-1
In this work, the operating cost for the large tractor will be
considered only the composition between the cost of diesel
fuel and the cost of the operator (salary, charges, insurance,
and training), maintenance, depreciation, financial, and
investment costs will not be considered.
The fuel consumption cost can be obtained according to the
D497 standard considering the diesel consumption obtained by
 = 2.64 $ + 3.91 − 0.203 738$ + 173 (4)
Where: Cf is the diesel consumption (L kW h-1);
X is the ratio between the power in the Power Take
Off (PTO) equivalent for the agricultural operation and the
total power of the PTO (decimal).
In deep plowing operation, the ratio X is equal to 1. Therefore,
the estimated fuel consumption of the John Deere 8730R
tractor, according to the D497 standard (ASABE, 2013) is:
Cf = 0.42 L kW h-1
Considering a working period of 1 hour at the tractor's nominal
power, there is a consumption of 114.24 liters of diesel. With
an exchange value for the dollar in January 2020 of R$ 4.16
per US $, the amount of one liter of diesel in January 2020 was
US $ 0.77. Therefore, the hourly fuel cost for the large tractor
is US $ 87.96. The same dollar quote was used to convert all
About the operator's monthly cost, in a 40-hour workday,
according to data from (BRASIL, 2020; CNA, 2020), the
median salary of a tractor operator in Brazil is US $ 361.53 in
the exchange rate. January 2020. In addition to this amount,
there are labor charges, approximately 70% of the salary
(Fernandes, 2020) US $ 252, the cost of life insurance US $
13.90 (CNA, 2020) and costs with training US $ 6.27
(SENAR, 2020). In total, there is an operator cost value of US
$ 634.40 per month. Per hour the value is US $ 15.86.
Therefore, the operating cost of the large tractor is US $ 103.82
per hour.
Regarding TRSE, there is no cost for the operator. Only the
one related to the electrical system and the charging of energy
from the grid. This cost is around US $ 2.72 per hour (Vogt,
2018). It doubles if the second pack of 4 batteries is considered
to increase the system's autonomy to 5 hours.
A 272 kW tractor with operating capacity ten times greater
than a small 20 kW TRSE has a much higher cost. However,
when considering a set of 10 multi-robots operating according
to swarm methods, the field capacity is equalized. In this case,
the total cost of the ten multi-robots would be around US $
27.20 per hour, considering 5 hours of autonomy. With the
second battery pack, it’s cost is US $ 54.40 per hour, half the
hourly cost of a large tractor.
The purchase price of a John Deere 8730R tractor
(TractorHouse, 2020) is US$ 355,400.00. For cost estimation
this paper proposes a multifunctional agricultural swarm robot
(TRSE) through the integration of robotic technologies with a
new version of the electric tractor developed by (Vogt,
2018).The TRSE value composition is shown in Table 1.
Figure 1 and 2.
Table 1. TRSE value composition.
Component UN.
(US$) Ref.
Electric Motor HPEVS
AC23-96V/650A 2 3,800.00 7,600.00 (HPEVS, 2020)
Curtis1238-96V 1 2,150.00 2,150.00 (Cu rtis, 2020)
for tract or
access/power systems
Advantech APAX5620KW
1 1,130.00 1.130,00 (Advantech,
Processor A
Intelligence Intel Core-i7
1 673.00 673.00 (Terabyte,
systems module Advantech
CANbus communication
4 172.00 688.00 (Advantech,
Communication CAN
module Curtis 1351 for
Trimble EZ and Controller
Curtis 1238
1 198.00 198.00
Controllers |
Automatic Pilot
Trimble EZ 1 7,250.00 7,250.00 (Trimble, 2020)
Tractor Chassis/Tires
Hydraulic systems
Drivetrain/tracks - - 500.00 -
Power systems (PTO,
hydraulic arm, drawbar) - - 500.00 -
Moura 12MS234
12V/220Ah 4 352.00 1,425.00
Summarizing this comparison, we have that a 10 TRSE swarm
has the same field capacity in hectares per hour in plowing as
the John Deere 8730R tractor (1.4 ha/h). However, the cost of
purchasing a JD8730R is US $ 102,240.00, more than the sum
of the value of 10 TRSE. In the comparison of operating cost,
for the same field capacity, the cost of the JD8730R is 3.2
times higher than the operating cost per hour of the TRSEs
swarm. And the weight of the large tractor is 2.8 times greater
than the total weight of the 10 TRSE, but a caveat is necessary.
In terms of soil mechanics, the value to be considered in the
operation of the TRSE is 7000 N, as it’s the request that the
soil receives in compression from an electric robot swarm
tractor. In the region where a TRSE passes, no other will pass,
since the operation was carried out. There is no need for traffic
on the ground. On the other hand, in the region where a
JD8730R passes, the soil undergoes a compression equivalent
to 198.050 N. The effects soil compaction will be completely
different, and favorable to TRSE.
Figure 1. Tractor Robot Swarm Electric (TRSE) simulating
the use of a moldboard plow.
Fig. 2. Tractor/Implements Power Transfer Systems
A critical issue in this context is the impact of the widespread
use of robotic systems in the agricultural field in terms of
personnel with technical knowledge to operationalize this type
of vehicle and perform the maintenance of equipment. This
challenge may be as difficult as developing robot swarms.
However, the authors believe that if Brazil wants to continue
to be a major player in global agribusiness, it has no alternative
but to invest in education and training for its workforce
(notably operators conventional tractors) is transformed
according to the new world trend of agriculture 4.0, which has
one of its fulcrums in robotization.
Institutions like National Rural Learning Service (SENAR,
2020) must pay attention to this need and train frontline
workers, Federal Institutes (Instituto Federal, 2020) need to
train the necessary technicians and universities to develop
research and train engineers able to implement them on the
This draft of roadmap can generate very fruitful
interinstitutional and transdisciplinary partnerships that will be
able to implement innovative and essential research so that this
proposal goes off the record and becomes one carried out in
the agricultural world. In this context, it is important to
emphasize that there is a huge field for studies with great
challenges concerning the operationalization of each of the
agricultural operations (Minßen et al., 2017) within the
universe of robotics, specifically in the area of multi-robots
working in swarm methods. The ASABE D497 standard
defines 48 different types of agricultural operations with
different characteristics and parameters (ASABE, 2013). For
each of them, several challenges for the operationalization of
a swarm robot system are presented and are suggested below:
4.1 Behavior –base systems
The great challenge of this line of research is to develop
control architectures based on behavior that can be adapted to
the unstructured agricultural environment, new methods of
incremental learning must be developed and adapted of robots
based on war environments or catastrophe environments. In
this context, reinforcement learning is an excellent
methodology to optimize agricultural robotic systems based on
behavior, because through the "decomposition" of the
behavior in small sub-behaviors it is possible to reduce the size
of the phase space effectively. This is another major challenge
in this line research to find the best learning network through
which the modularization of learning "policies" results in
accelerated and more robust learning.
4.2 AI reasoning methods
The essential question for robotics in the elaboration of the
application of artificial intelligence methods is to define which
are the suitable formats for KR, and from this definition find
the state function that refers to generation and maintenance, in
real-time, of a symbolic description of the robot's environment,
based on a recent situational condition of the environmental
information obtained by sensors and communication with
other agents involved in such a way that decision-making is
correct and optimized for solving a problem or overcoming a
4.3 Collective-level behavior
The behavior at the collective level of the swarm is what
defines the conclusion of the global mission about each
specific objective of the agricultural operation, the challenge
here is to develop swarm robotics techniques that enable the
emergence of emergent group behaviors, such as self-
organization, flexibility in joint operations, and scalability in
terms of common objectives.
4.4 Operational strategy
The division of the agricultural field can be configured in cells,
lines, bands, blocks, in short, several sets with different
topologies, which in the real agricultural environment take
very complex forms due to the specificities of the relief,
contour lines, shape of the fields, planting configuration of
cultures. All these topological parameters lead to complex
logistical problems for the optimization of the movement
strategy of the elements of the swarm, reaching questions
related to differential geometry.
4.5 Stochastic process computing
A significant challenge for the operationalization of a robotic
tractor swarm in the field is the extreme unpredictability that
the agricultural environment has. Even in a homogeneous
culture sown with a uniform pattern, it has a high variability of
shapes, geometries, positions, and scales. This fact occurs due
to the treatment of living elements, which interact with the
climate, soil and other living beings (micro and
macroscopically), in this concept it is necessary to enter into
the area of stochastic processes so that there is a better
understanding of the operational strategies as well as an
adaptation of the decision-making algorithms against random
4.6 Multiple decision-making
In agricultural swarm robotics its necessary the definition of
the algorithms that can be used, because of the processing
capacity of the machines about the extraordinarily complex
and multiple decision-making problems required in
unstructured agricultural environments and which has objects
very fragile and variable (living beings).
4.7 On-board systems
Computer vision systems are a universe, from the development
of specific hardware to the elaboration of suitable firmware.
When thinking about the immense range of sensors, receivers,
and transducers necessary to make a swarm robot operational
in the field, it is essential to divide this field of studies into the
proprioception, exteroception, and guidance system. The
challenges are the sensors optimization of the sensors and
actuators, necessary to enable the internal and external robot’s
operation, depending on the specific agricultural process. In
terms of guidance, the agricultural environment offers
immense obstacles, from varying light conditions, such as
irregular reliefs, to complex and often discontinuous contour
4.8 Hardware enhancement
According to (D. Albiero, 2019) the main obstacle is the
development of adapted to the agricultural conditions systems,
because there are many very good elements of automation and
robotics used in industry and smart cities, but when they are
part of the agricultural world (with high susceptibility of the
agricultural products in the spoiling of the most varied forms),
problems occur. There is the urgent necessity of developments
in robotics technologies for agricultural reality.
4.9 Networked Swarm robots
This line of research is very challenging because, through the
connection between the members of the swarm, it is possible,
through distributive computing, to imitate the behavior of
decentralized animals through simple behaviors, capable of
generating complex responses at the collective level, the so-
called emergent behaviors.
4.10 Distributive computing
In particular, the concept of parallelism has become essential;
we have seen the development of multicore CPUs. With a
swarm of robots in the agricultural field, the distributed
computing system functionality is immense, both in terms of
capturing data and generating useful information to optimize
the specific function performed. The challenge is to integrate
all this processing through wireless communication networks
in the field.
4.11 Particle Swarm optimization
According to (Nedjah & Macedo Mourelle, 2006), particle
swarm optimization is a mathematical optimization method
that mimics the behavior of insect swarm, the challenge is to
find the optimal path or solution for the entire swarm and
implement communication between the swarm robots network
optimizing the answer ahead of the swarm before the common
goal. This line of research has deep interfaces with lines 5.1,
5.3, 5.9, and 5.10.
The comparison of costs and field capacity between a set of
Swarm Electric Robot Tractors (TRSE) and a large tractor was
performed, demonstrating the feasibility of the swarm
configuration to replace the large tractor for deep plowing. The
constructive preliminary design of a new autonomous 24 kW
electric robot tractor within the swarm operating
methodologies have been described. Challenges for future
research focused on the implementation in the agricultural
field of swarm robots aiming at mechanized operations are
suggested. This initiative can constitute a roadmap for
interinstitutional and transdisciplinary research bringing
together the scientific community dedicated to robotics in
The authors thanks the productivity grant of CNPq and the
physical and financial resources made available by FUNCAP,
Advantech. (2020a). APAX-5490-IP4AE Advantech | Mouser
Advantech. (2020b). APAX-5620KW-AE Advantech | Mouser
Albani, D., Haken, R., & Trianni, V. (2017). Monitoring and
Mapping with Robot Swarms for Agricultural Applications.
Albani, D., Manoni, T., Arik, A., Nardi, D., & Trianni, V.
(2019). Field coverage for weed mapping: toward
experiments with a UAV swarm.
Albiero, D., Xavier, R. S., Garcia, A. P., Marques, A. R., &
Rodrigues, R. L. (2019). The technological level of
agricultural mechanization in the state of ceará, brazil.
Engenharia Agrícola, 39(1), 133–138.
Albiero, D. (2011). Utilização de energia na agricultura –
Parte II - Jornal Dia de Campo. Jornal Dia de Campo.
Albiero, D. (2019). Agricultural Robotics: A Promising
Challenge. Current Agriculture Research Journal, 7(1), 01–
Albiero, D, Cajado, D., Fernandes, I., Monteiro, L. A., &
Esmeraldo, G. (2015). Tecnologias Agroecológicas para o
Semiárido (Daniel Albiero (Ed.)). UFC.
Araújo Batista, A. V., Albiero, D., de Araújo Viana, T. V., de
Almeida Monteiro, L., Chioderoli, C. A., de Sousa, I. R. S.,
& Azevedo, B. M. (2017). Multifunctional Robot at low cost
for small farms. Ciencia Rural, 47(7).
ASABE. (2013). Standard D497. 2011, 5.
Bechar, A., & Vigneault, C. (2016). Agricultural robots for
field operations: Concepts and components. Biosystems
Engineering, 149, 94–111.
Bechar, A., & Vigneault, C. (2017). Agricultural robots for
field operations. Part 2: Operations and systems. Biosystems
Engineering, 153, 110–128.
Blender, T., Buchner, T., Fernandez, B., Pichlmaier, B., &
Schlegel, C. (2016). Managing a Mobile Agricultural Robot
Swarm for a seeding task. IECON Proceedings (Industrial
Electronics Conference), 6879–6886.
BRASIL. (2020). Ministério da Economia — Português
CNA. (2020). Home | Confederação da Agricultura e
Pecuária do Brasil (CNA).
Curtis. (2020). Curtis 1238-7601 HPEVS AC-50 Brushless
AC Motor Kit - 96 Volt, EV West - Electric Vehicle Parts,
Components, EVSE Charging Stations, Electric Car
Conversion Kits.
Davis, B. (2012). CMU-led automation program put robots in
the field, AUVSI’s unmanned systems. Mission Critical, 2.
De Lemos, R. A., De, L. A. C., Nogueira, O., Ribeiro, A. M.,
Mirisola, L. G. B., Koyama, M. F., De Paiva, E. C., & Bueno,
S. S. (2018). Unisensory intra-row navigation strategy for
orchards. CBA.
Deere, J. (2020). Tratores agrícolas Grandes | Comprar
Trator | John Deere BR.érie-7j-grande-200cv-
Fernandes, D. P. (2020). Encargos trabalhistas: quanto custa
um funcionário? [Tabela].
Goering, C. E., & Hanson, A. C. (2004). Engine and tractor
power. American Society of Agricultural Engineers.
Goering, C. E., Stone, M. L., Smith, D. W., & Turnquist, P.
K. (2003). Off-Road Vehicle Engineering Principles.
HPEVS. (2020). HPEVS AC Electric Motor AC-50 Power
Instituto Federal - Instituto Federal. (2020).
Kiehl, E. J. (1979). Manual de edafologia: relação solo
planta - E. J. Kiehl - Google Books. Ceres.
Macmillan, R. H. (2002). The Mechanics of Tractor.
University of Melbourne.
Melo, R. R., Antunes, F. L. M., Daher, S., Vogt, H. H.,
Albiero, D., & Tofoli, F. L. (2019). Conception of an electric
propulsion system for a 9 kW electric tractor suitable for
family farming. IET Electric Power Applications, 13(12),
Minßen, T. F., Schattenberg, J., Cord, C. G. M., Urso, M.,
Hanke, M. S., & Frerichs, L. (2017). Robots for Plant-
Specific Care Operations in Arable Farming. Montpellier
Motor Controllers | Curtis Instruments. (n.d.). Retrieved
April 29, 2020, from
Mukherjee, A., Misra, S., Sukrutha, A., & Raghuwanshi, N.
S. (2020). Distributed aerial processing for IoT-based edge
UAV swarms in smart farming. Computer Networks, 167,
Nedjah, N., & Macedo Mourelle, L. de. (2006). Swarm
intelligent systems. Springer-Verlag.
Renius, K. T. (2020). Fundamentals of Tractor Design.
SENAR. (2020). Portal Senar EAD.
Shamshiri, R. R., Weltzien, C., Hameed, I. A., Yule, I. J.,
Grift, T. E., Balasundram, S. K., Pitonakova, L., Ahmad, D.,
& Chowdhary, G. (2018). Research and development in
agricultural robotics: A perspective of digital farming. Int J
Agric & Biol Eng, 11(4), 1–14.
Srivastava, A. K., Goering, C. E., Rohrbach, R. P., &
Buckmaster, D. R. (2006). Engineering Principles of
Agricultural Machines, Second Edition. ASABE.
Terabyte. (2020). Processador Intel Core i3, i5, i7, i9, 8a e 9a
Geração Terabyteshop.
Toorn, J. aan den. (2020). SAGA - Swarm Robotics for
Agricultural Applications - The European Coordination Hub
for Open Robotics Development.
TractorHouse. (2020). JOHN DEERE 8370R For Sale - 337
Listings | - Page 1 of 14.
Trianni, V., Ijsselmuiden, J., & Haken, R. (2016). The SAGA
concept: Swarm Robotics for Agricultural Applications.
Trimble. (2020). Trimble Agriculture.
Vogt, H. H. (2018). Electric Tractor System Propelled by
solar energy.
Vogt, H. H., Albiero, D., & Schmuelling, B. (2018). Electric
tractor propelled by renewable energy for small-scale family
farming. 2018 13th International Conference on Ecological
Vehicles and Renewable Energies, EVER 2018, 1–4.
WinnerShop. (2020). Bateria Solar Moura 12Ms234
Estacionaria Energia Solar 12V 220Ah | WinnerShop.
Wolfert, S., Ge, L., Verdouw, C., & Bogaardt, M. J. (2017).
Big Data in Smart Farming – A review. Agricultural Systems,
153, 69–80.
Xaud, M. F. S., Leite, A. C., Barbosa, E. S., Faria, H. D.,
Loureiro, G. S. M., & From, P. J. (2018). Robotic tankette for
intelligent bioenergy agriculture. CBA.
ResearchGate has not been able to resolve any citations for this publication.
Full-text available
This work addresses the challenges of a decentralized and heterogeneous Unmanned Aerial Vehicle (UAV) swarm deployment – some fitted with multimedia sensors, while others armed with scalar sensors – in resource-constrained and challenging environments, typically associated with farming. Subsequently, we also address the resulting problem of sensing and processing resource-intensive data aerially within the Edge swarm in the fastest and most efficient manner possible. The heterogeneous nature of the Edge swarm results in under-utilization of the available computation resources due to unequal data generation within its members. To address this, we propose a Nash bargaining-based weighted intra-Edge processing offload scheme to mitigate the problem of heavy processing in some of the swarm members. We do this by distributing the data to be processed to all the swarm members. Real-life hardware tuned simulation of a large UAV swarm shows that by increasing the number of UAVs in the swarm, our scheme achieves better scalability and reduced processing delays for intensive processing tasks. Additionally, in comparison to regular star and mesh topologies, our scheme achieves an increase in collective available network processing speeds by 100% for only 25% of the number of UAVs in a star topology.
Full-text available
This study presents an updated review of the application of electric tractors. A customised drive system for the conception of a novel low‐power electric tractor suitable for family farms is also introduced and discussed. The introduced system comprises several aspects regarding energy generation, transmission, conversion, storage, utilisation, conservation, and management, as well as sustainability issues. A 9 kW prototype composed of two three‐phase induction motors, two independent inverters, and a lead–acid battery bank is presented. Flexible and safe operation is ensured by using an electronic control unit specifically designed for this project, as a dedicated control algorithm is also developed to provide greater versatility under common rural activities. Also, a supervisory system is proposed for data storage and performance analysis. To verify the proper performance of the electric tractor, the methodology used for conducting drawbar tests has been based on document CODE 2 by Organisation for Economic Co‐operation and Development (OECD). Experimental results are presented and discussed; thus, demonstrating that the proposed electric tractor is technically feasible in terms of performance when compared with a similar internal combustion engine one.
Conference Paper
Full-text available
This work presents an unisensory autonomous navigation strategy for orchards environment using information provided by a single Light Detection and Ranging (LIDAR) sensor. A reference path is obtained with the use of the Random Samples and Consensus (RANSAC) method and an improvement is achieved using the Extended Kalman Filter (EKF). This path is used as the reference for a trajectory controller to guide the platform between the plantation corridor. The control strategy considers a proportional-integral controller actuating upon a given "look ahead error" composed of a lateral distance error plus a predict error which is an estimation of the lateral error evolution. Implementation codes are made available and a validation experiment is performed.
Full-text available
Agricultural mechanization is a significant factor of agricultural modernization, but there are few data on the technological levels of this mechanization in the State of Ceará, Brazil. To study this topic, we designed a structured questionnaire, and interviews were conducted across the state. The survey data made it possible to calculate an index of agricultural mechanization technology and study hypotheses specific to farms of Ceará. The results indicate that the technology levels are different between farms of different sizes: large and medium farms are level I-M5 (automation), while small farms ranged from level IV-M1/M2 (primary/animal) to III-M3 (preliminary). We conclude that public policy for training should be directed at small farms. There is a need to show new production alternatives to small farmers, which leads to savings, and increases the profitability of the activity.
Conference Paper
Full-text available
In recent years, the use of robots in agriculture has been increasing mainly due to the high demand of productivity, precision and efficiency, which follow the climate change effects and world population growth. Unlike conventional agriculture, sugarcane farms are usually regions with dense vegetation, gigantic areas, and subjected to extreme weather conditions, such as intense heat, moisture and rain. TIBA - Tankette for Intelligent BioEnergy Agriculture -is the first result of an R&D project which strives to develop an autonomous mobile robotic system for carrying out a number of agricultural tasks in sugarcane fields. The proposed concept consists of a semi-autonomous, low-cost, dust and waterproof tankette-type vehicle, capable of infiltrating dense vegetation in plantation tunnels and carry several sensing systems, in order to perform mapping of hard-to-access areas and collecting samples. This paper presents an overview of the robot mechanical design, the embedded electronics and software architecture, and the construction of a first prototype. Preliminary results obtained in field tests validate the proposed conceptual design and bring about several challenges and potential applications for robot autonomous navigation, as well as to build a new prototype with additional functionality.
Full-text available
Digital farming is the practice of modern technologies such as sensors, robotics, and data analysis for shifting from tedious operations to continuously automated processes. This paper reviews some of the latest achievements in agricultural robotics, specifically those that are used for autonomous weed control, field scouting, and harvesting. Object identification, task planning algorithms, digitalization and optimization of sensors are highlighted as some of the facing challenges in the context of digital farming. The concepts of multi-robots, human-robot collaboration, and environment reconstruction from aerial images and ground-based sensors for the creation of virtual farms were highlighted as some of the gateways of digital farming. It was shown that one of the trends and research focuses in agricultural field robotics is towards building a swarm of small scale robots and drones that collaborate together to optimize farming inputs and reveal denied or concealed information. For the case of robotic harvesting, an autonomous framework with several simple axis manipulators can be faster and more efficient than the currently adapted professional expensive manipulators. While robots are becoming the inseparable parts of the modern farms, our conclusion is that it is not realistic to expect an entirely automated farming system in the future. © 2018, Chinese Society of Agricultural Engineering. All rights reserved.
This textbook offers a comprehensive review of tractor design fundamentals. Discussing more than hundred problems and including about six hundred international references, it offers a unique resource to advanced undergraduate and graduate students, researchers and also practical engineers, managers, test engineers, consultants and even old-timer fans. Tractors are the most important pieces of agricultural mechanization, hence a key factor of feeding the world. In order to address the educational needs of both less and more developed countries, the author included fundamentals of simple but proved designs for tractors with moderate technical levels, along with extensive information concerning modern, premium tractors. The broad technical content has been structured according to five technology levels, addressing all components. Relevant ISO standards are considered in all chapters. The book covers historical highlights, tractor project management (including cost management), traction mechanics, tires (including inflation control), belt ground drives, and ride dynamics. Further topics are: chassis design, diesel engines (with emission limits and installation instructions), all important types of transmissions, topics in machine element design, and human factors (health, safety, comfort). Moreover, the content covers tractor-implement management systems, in particular ISOBUS automation and hydraulic systems. Cumulative damage fundamentals and tractor load spectra are described and implemented for dimensioning and design verification. Fundamentals of energy efficiency are discussed for single tractor components and solutions to reduce the tractor CO2 footprint are suggested.
Precision agriculture represents a very promising domain for swarm robotics, as it deals with expansive fields and tasks that can be parallelised and executed with a collaborative approach. Weed monitoring and mapping is one such problem, and solutions have been proposed that exploit swarms of unmanned aerial vehicles (UAVs). With this paper, we move one step forward towards the deployment of UAV swarms in the field. We present the implementation of a collective behaviour for weed monitoring and mapping, which takes into account all the processes to be run onboard, including machine vision and collision avoidance. We present simulation results to evaluate the efficiency of the proposed system once that such processes are considered, and we also run hardware-in-the-loop simulations which provide a precise profiling of all the system components, a necessary step before final deployment in the field.