Industrial Robot: An International Journal

Moisture measurement in crops using spherical robots

Juan D. Hernández Jorge Barrientos Jaime del Cerro Antonio Barrientos David Sanz
Article information:
To cite this document:
Juan D. Hernández Jorge Barrientos Jaime del Cerro Antonio Barrientos David Sanz, (2013),"Moisture measurement in crops
using spherical robots", Industrial Robot: An International Journal, Vol. 40 Iss 1 pp. 59 - 66
Permanent link to this document:
[Link]
Downloaded on: 31 January 2016, At: 07:53 (PT)
References: this document contains references to 22 other documents.
To copy this document: permissions@[Link]
The fulltext of this document has been downloaded 307 times since 2013*
Users who downloaded this article also downloaded:
Downloaded by New Mexico State University At 07:53 31 January 2016 (PT)

Luis Emmi, Leonel Paredes-Madrid, Angela Ribeiro, Gonzalo Pajares, Pablo Gonzalez-de-Santos, (2013),"Fleets of robots
for precision agriculture: a simulation environment", Industrial Robot: An International Journal, Vol. 40 Iss 1 pp. 41-58 http://
[Link]/10.1108/01439911311294246
Gokhan Bayar, (2013),"Long distance autonomous trajectory tracking for an orchard vehicle", Industrial Robot: An International
Journal, Vol. 40 Iss 1 pp. 27-40 [Link]
W.X. Yan, Z. Fu, Y.Z. Zhao, (2013),"Realization of turn-over-wok movement for cooking robot", Industrial Robot: An International
Journal, Vol. 40 Iss 1 pp. 67-76 [Link]

Access to this document was granted through an Emerald subscription provided by emerald-srm:199044 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for Authors service
information about how to choose which publication to write for and submission guidelines are available for all. Please visit
[Link]/authors for more information.
About Emerald [Link]
Emerald is a global publisher linking research and practice to the benefit of society. The company manages a portfolio of
more than 290 journals and over 2,350 books and book series volumes, as well as providing an extensive range of online
products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee on Publication Ethics
(COPE) and also works with Portico and the LOCKSS initiative for digital archive preservation.

*Related content and download information correct at time of download.

 Research article

Moisture measurement in crops

using spherical robots
Juan D. Hernández, Jorge Barrientos, Jaime del Cerro, Antonio Barrientos and David Sanz
Centre for Automation and Robotics – CAR,
Universidad Politécnica de Madrid-Consejo Superior de Investigaciones Cientı́ficas (UPM-CSIC), Madrid, Spain

Abstract
Purpose – The purpose of this paper is to present a new low-cost system based on a spherical robot for performing moisture monitoring in precision
agriculture.
Design/methodology/approach – The work arose from the necessity of providing farmers with alternative methods for crop monitoring. Thus, after
analysing the main requirements, a spherical robot was chosen as a tentative approach. The presented work summarizes the work carried out in
Downloaded by New Mexico State University At 07:53 31 January 2016 (PT)

selecting the basics to apply in the robot, as well as its mechanical and electronic design. After designing and constructing the robot, several tests have
been performed, in order to validate the robot for performing monitoring task and moving on different types of soil.
Findings – The performed tests reveal that spherical robot is a suitable solution for performing the task.
Research limitations/implications – Some improvements in control should be applied in order to reach a fully autonomous navigation in very
slippery soils. Nevertheless, the performance of the robot in teleoperated mode allows validating of the system.
Practical implications – The robot turned out to be friendly and harmless in its use for this application. The cost of final series will be affordable in
comparison with the cost of other methods. Endurance of the robot can be considered as fair.
Originality/value – The paper presents a new tool for farming based on non-common robot.

Keywords Agriculture, Crops, Moisture measurement, Robots, Spherical robot, Precision agriculture, Moisture monitoring, Irrigation control

Paper type Research paper

Motivation Currently, unmanned aerial vehicles (UAV) are also used

for such purposes with lower cost compared to previous
Since farm management costs, and consequently crop results solutions, but still require a service company to perform the
are directly associated with the environmental moisture, water task. UAV mission consists of taking aerial images of the crop
control optimization becomes a critical factor in performing fields. The images acquired are used for building high-
sustainable agricultural practices (Sutar, 2012). Irrigation resolution maps, not only for hydric stress studies but also for
management in large crops is consequently an essential practice weed detection. An important research effort has been
in precision agriculture (PA). Modern control techniques dedicated to computing optimal trajectories for mini UAVs,
are applied to make this control effective (Capraro et al., 2008; such as quad-rotors (Valente et al., 2011a, b).
Rahangadale and Choudhary, 2011; Quanxing et al., 1996). Nonetheless, ground methods such as single sensors or
These techniques use moisture estimation (i.e. indirect wireless sensor networks (WSN) have proven to be good static
methods such as photography or crop analyses) or solutions (Patil et al., 2011; Ke et al., 2011; Shinghal et al.,
measurement methods (i.e. environmental variables 2010; Kim et al., 2008) but require lots of devices in large
monitoring such as temperature and relative humidity fields with heterogeneous characteristics, being most of the
(Bouttier et al., 1993). times too expensive. Considering static sensor monitoring,
Typically, crop humidity is quantified by using water acquisition systems could be classified according to their
stress index (WSI) that can currently be obtained from range and precision. Taking into account that range (not only
multispectral satellite or plane images. Nevertheless, these individual sensor range, but also coverage) and accuracy are
methods are very costly, scarce and highly dependent on the inversely related, a balance is thereby required.
weather. On the other hand, using mobile sensors, i.e. sensors
shipped on ground mobile robots, would reduce the cost of the
The current issue and full text archive of this journal is available at
[Link]/[Link]
This work has been supported by the Robotics and Cybernetics Research
Group at Centre for Automation and Robotics CAR UPM-CSIC, and
funded under the project “Robot Fleets for Highly Effective Agriculture
and Forestry Management”, (RHEA) sponsored by the European
Industrial Robot: An International Journal
40/1 (2013) 59– 66 Commission’s Seventh Framework Programme (NMP-CP-IP 245986-2
q Emerald Group Publishing Limited [ISSN 0143-991X] RHEA). This paper is revised and updated from an original article
[DOI 10.1108/01439911311294255] published at RHEA-2012, Pisa, Italy, September 19-21, 2012.

59
 Moisture measurement in crops using spherical robots Industrial Robot: An International Journal
Juan D. Hernández et al. Volume 40 · Number 1 · 2013 · 59 –66

entire WSN, but obviously a robot is required. Apart from the kind of robots in the literature, regarding concepts and new
cost of the robot, the impact in the farming process (e.g. as prototypes proposals.
they may damage the cultivation) and the human effort Initially, research activities were focused on validating
required to operate the robot are the main drawbacks of this concepts about physics. In this regard, several authors have
solution. proposed different approaches, where the main objective was
to create a mechanical system that allows shifting the centre of
Purpose of the work mass of a sphere and consequently inducing self-motion. Five
typical methods can be found in literature in order to achieve
This paper presents the work carried out in this field in order this goal (Armour and Vincent, 2006):
to minimize the adverse effects of using mobile robot for sensing 1 Spring central member. A central body that includes a
in farming. Thus, a rolling robot with a spherical shape, named driven wheel in one of its ends and a passive wheel in the
“ROSPHERE” (RObotic SPHERE) has been designed and other, with a spring that guarantees contact between both
tested as a low-cost robot that minimizes the damage to the wheels and the spherical shaped body. The main
crop while performing monitoring task. The main purpose of disadvantage of this design is the loss of energy due to
using this alternative vehicle is to minimize crop alterations, and friction between both wheels and the sphere.
at the same time, to be able to have direct measurements from 2 Car driven. The robot relies on a vehicle inside in order to
plant surroundings. In such manner, the action required is only induce motion. The main drawback is the lack of contact
applied on the affected area. These actions may include water when the robot moves on surfaces with depressions and
irrigation, application of pesticides or fertilizers, etc. This
bumps.
Downloaded by New Mexico State University At 07:53 31 January 2016 (PT)

precise action has consequently an effect on the economical

3 Ballast mass with fixed axis system. This system uses an
and environmental cost minimization and, at the same time, it
inner pendulum with two rotational degrees of freedom.
maximizes revenues.
The first DoF rotates around a fixed transverse axis and
Nevertheless, the aim of the proposed system is not to
the second around a longitudinal axis (Michaud and
completely replace what aerial systems do, but complement
Caron, 2002; Kayacan et al., 2011). This configuration
them in their daily work. Thus, aerial units can only be
required to perform their task in specific moments, which was chosen in order to design the ROBOSPHERE robot.
4 Ballast mass with moving axis. It also has an inner
thereby reduces the final cost of the process.
pendulum, but in this case with an additional DoF that
allows the main axis to move.
Approach
Due to the method chosen for designing the robot, the
As previously mentioned, the main objective of PA is to ROSPHERE prototype is a non-holonomic robot. Therefore,
minimize environmental impact while maximizing the use of the vehicle requires forward or backward displacements so as
non-renewable resources (e.g. water). Local measurements of to rotate. Nevertheless, several alternatives to create a
control variables such as temperature, pesticide concentration, holonomic vehicle can be found based on mobile masses
luminosity, humidity, etc. are useful indicators in order to system (Amir and Mojabi, 2002; Shengju et al., 2010). Thus,
determine if an action is required on the crop. the prototypes that use this concept take advantage of the
Based on the mentioned above, the main goal is to be able to movement of masses along radial axes in order to modify the
evaluate the real status of the crop, not only from a global point
position of the centre of mass.
of view but also locally. Therefore, the solution must be able to
Several robotics spheres have been developed for quite
assess the state of the crop without affecting the involved plants.
different areas of applications. Perhaps the most cited and
Thus, the capability of the robot to navigate in crop rows in
ambitious application of robotic spheres has been the one
order to monitor variables is the foremost requirement to be
proposed by Zhan et al. (2008). They propose to explore
considered. The analysis of this requirement links two more
unstructured and unknown environments by exploiting
aspects such as size and weight. Taking into account the previous
the robustness and versatility of spherical robots. Meanwhile
considerations, a robotic sphere is proposed as a solution.
Robotic spheres are systems in which movements are induced Bruhn et al. (2008) or Michaud et al. (2001) have also proposed
by instability. Besides, considering their regular shape, the robot their use for planetary exploration. Other fields of application
may recover easily from collisions. Regardless the direction of can take advantage of the mentioned characteristics, such as
the impact, the robot tends to fall into a recoverable security, surveillance and inspection (Seeman et al., 2006),
configuration. In comparison to other robots, i.e. wheeled where robotic spheres are endowed with sensors and cameras in
robots with similar moving capabilities and equivalent size order to allow the teleoperation of the robot. However, no
(e.g. Mobilerobots Pioneer 3-AT), ROSPHERE is drastically references regarding the use of spherical robot for agricultural
lighter. Thus, the previously mentioned wheeled robot weights purposes have been found.
about 12 kg while ROSPHERE weights 2.5 kg. On the other hand, since one of the primary requirements
In addition to this, the furrows help the spherical robot to move for service robots is the capability for harmless interaction
along the row, minimizing in this way the damage to the plants. with humans and the environment, robotic spheres have also
At the same time, the lighter the robot, the lower consumption. been applied as service robot in direct interaction with
Consequently, not only more endurance for the robot but also humans, validating this feature. Thus, Michaud et al. (2005)
cheaper drivers and motors. used a robotic sphere equipped with some special control
routines and sensors in order to monitor child development
Spherical robots: concepts and first prototypes while it was being used as a toy.
Although robotic spheres are not widely used as mobile platforms, Finally, more academic contributions presenting
it is possible to find quite significant references focused on this robotics spheres to study kinematics, dynamics and control

60
 Moisture measurement in crops using spherical robots Industrial Robot: An International Journal
Juan D. Hernández et al. Volume 40 · Number 1 · 2013 · 59 –66

of non-holonomic systems can also be found in literature Figure 2 ROSPHERE ABS model
(Bicchi et al., 1997).

System description
ROSPHERE is a spherical shaped robot with the capability to
WIFI Antennas
self-induce non-holonomic movements. In order to make that
possible, the robot has an inner two-degree-of-freedom Internal Control Unit
pendulum. The robot includes:
.
a spherical shaped body (30 cm of diameter);
.
a fixed main axis;
.
a central unit or internal control unit (ICU); and
.
the ballast or hanging mass.
The first DoF rotates the ICU (consequently the hanging mass)
about the fixed axis, while the second has a mechanically
limited range of rotation, and rotates about a perpendicular
axis. Current version of the robot (Figures 1 and 2) was Main
designed to get the centre of mass as far as possible from Axis
the geometrical centre in order to induce the movements
Downloaded by New Mexico State University At 07:53 31 January 2016 (PT)

more easily.
In order to allow a comprehensive visual inspection of how
the pendulum works during the operation of the robot, the
ROSPHERE prototype has been made by using a commercial Battery
transparent ball, as shown in Figure 3. Although good results Ballast
have been obtained by using this commercial ball, a more
suitable material should be used in its final version in order to
increase friction and consequently, the efficiency of its Figure 3 ROSPHERE prototype
traction. The grooves of the case have been sealed using
special rubber bands in order to keep the robot protected
from dust and humidity, as shown in Figure 9. Different
sensors can be placed on both of its sides, out of the tread
contour, in order to avoid shocks. Nevertheless, if a heavy
sensor (e.g. a camera) is used, a counterweight should be
added to the other side of the sphere in order to maintain the
centre of gravity in the same position.

Hardware architecture
ROSPHERE is equipped with all necessary resources in order
to behave as an autonomous vehicle. Thus, the equipment
includes a novel embedded computing system composed by a
Robovero and an Overo Fire embedded computer.

Figure 1 ROSPHERE CAD design

Considering communication capabilities, ROSPHERE relies

on Wi-Fi, Bluetooth and Xbee modules as communication
Ballast Main alternatives, not only with the base station but also with
Axis external sensors.
Thus, real time irrigation control could be performed by a
centralized system by using the information provided by the
robot on-line or by downloading it after finishing the mission.
Regarding sensing capabilities, the system includes the
equipment required for guidance, navigation and control
(GNC) systems, such as a low-cost inertial measurement unit
Battery (IMU) in order to measure angular velocities and

61
 Moisture measurement in crops using spherical robots Industrial Robot: An International Journal
Juan D. Hernández et al. Volume 40 · Number 1 · 2013 · 59 –66

accelerations, a magnetometer with pan-tilt correction using a wide spread of algorithm and already developed tools
capability and a single GPS. Additionally, an optical and their connection with robotic simulators (e.g. Gazebo or
encoder is required (still under development) in order to V-REP).
measure the rotational speed of the sphere and compare it The Overo communicates with Robovero by using a USB hub.
with the real speed of the robot. This will allow developing Robovero is responsible for low-level control and sensing. Thus,
more robust control algorithms that are useful when the it reads all the sensors and performs basic filtering in order to
traction is very low (e.g. very wet surfaces). create complete messages for the high-level controller. An
External sensors such as humidity and temperature are important development has been required in order to optimize
currently connected by using an analogic-digital (A/D) Robovero firmware so as to improve update frequencies in
converter. Figure 4 shows the hardware architecture of the sensors readings and processing.
system. Figure 5 shows the basic software modules of each part.
Considering power system endurance, the robot currently
operates by using a 2,200 mAh battery that weights 120 g. It
allows the robot to move from 20 to 45 min, depending on the
Tests and results
surface. The more irregular surface the higher consumption.
Additionally, next to the mentioned battery, the robot has two Preliminary tests
counterweights with a weight equivalent to seven batteries. Multiple tests have been carried out in order to assess the
Therefore, if these counterweights were replaced by similar performance of ROBOSPHERE in different terrains. The
batteries, the endurance of the robot would be longer than two movement capability on different surfaces was observed by
Downloaded by New Mexico State University At 07:53 31 January 2016 (PT)

and half hours in the worst case. The maximum speed of the using teleoperation mode. Thus, the robot was tested on a
robot also depends on the surface. Thus, velocities up to 2 m/s tarmac, cement, grass, sand, and gravel.
have been reached on flat surfaces. Nevertheless, speeds under In order to validate the capability of ROSPHERE for
0.1 m/s are required in order to perform a precise monitoring monitoring environmental variables, some tests have been
considering thermal inertias, reducing in this way the energy performed at a large extension in a public park in Madrid
consumption drastically. (Retiro Park), as shown in Figure 6.
During the test, not only was the capability for monitoring
Software architecture tested, but also the friendly relationship with people.
System architecture can be split into two parts. The first part The temperature and humidity were successfully recorded
is responsible for the high-level and the second for the low- and geo-referred as Figures 7 and 8
level (sensors and actuators level).
The first part is in charge of interpreting remote commands
from the base station and performing basic GNC tasks by Final tests
using processed information from the sensorial system. Thus, ROSPHERE was finally tested in two different crops; narrow
this layer transfers teleoperation commands to low-level rows crop such as winter cereal (Figure 9) and wide row crop
controller or calculates references to controllers in order to such as corn, as shows.
perform basic maneuvers. This part is executed on Overo In order to test the capability of ROSPHERE to move
board under Ubuntu Linux v. 11.0.4. Additionally, the ROS- across crops lines, the robot was teleoperated while collecting
Core has also been included in order to implement a standard the above-mentioned variables. The chosen field for winter
robot operating system (ROS) architecture. The project aims cereal had an inter-row distance smaller than expected
to deploy a complete ROS compliance robot, which will allow (farmers decision); nevertheless the robot was able to perform
its mission (Figure 10).
Figure 4 Hardware architecture of ROSPHERE Temperature and humidity evolution were registered and are
shown in Figures 11 and 12. As expected, both moisture and
temperature remained stable along the crop. Nevertheless,
a slight increment (lower than 3 per cent) could be appreciated
in the humidity, near the irrigated area.
Sampling rate was set to one sample per second during the
tests. Hardware and software capabilities allows the robot to
monitor the variables at 10 Hz but frequency was reduced
taking into account the speed of the robot and the characteristic
of the monitored variables.
On the other hand, it can be seen that three different plots
compose the temperature-monitoring chart. The black and red
ones correspond to the raw data of both sensors (LM31 and
SHT71, respectively). The blue line corresponds to the
compensated data, where LM31 provides the accuracy and
the SHT71 contributes with a higher stability.
Finally, some test were carried out in order to validate a
complete field monitoring. Figure 13 shows in red the GPS
position estimation of the trajectories (the aerial picture does not
reflect the actual state of the crop during the tests). Figures 14
and 15 show the measurements that were taken during the test.

62
 Moisture measurement in crops using spherical robots Industrial Robot: An International Journal
Juan D. Hernández et al. Volume 40 · Number 1 · 2013 · 59 –66

Figure 5 Software architecture of ROSPHERE

Downloaded by New Mexico State University At 07:53 31 January 2016 (PT)

Figure 6 Trajectory performed at Retiro Park

Figure 8 Temperature registration

Temperature registered by ROSPHERE [°C]

40.419

40.4185 32

40.418 30

40.4175 28
Longitude

40.417
26
40.4165
24
40.416
22
40.4155
20
40.415
–3.686 –3.685 –3.684 –3.683 –3.682 –3.681 –3.68
Figure 7 Humidity map
Latitude
Humidity registered by ROSPHERE [%] Note: Differences due to sun or shade are appreciated
40.419
36
40.4185
34
40.418 32

40.4175 30
Figure 9 ROSPHERE working on winter cereal field
Longitude

28
40.417
26
40.4165
24
40.416
22
40.4155 20

40.415 18
–3.686 –3.685 –3.684 –3.683 –3.682 –3.681 –3.68
Latitude

63
 Moisture measurement in crops using spherical robots Industrial Robot: An International Journal
Juan D. Hernández et al. Volume 40 · Number 1 · 2013 · 59 –66

Figure 10 ROSPHERE working on maize crop field Figure 13 Aerial view of inspected field

Figure 14 Humidity measurement during field inspection

Downloaded by New Mexico State University At 07:53 31 January 2016 (PT)

Figure 11 Humidity registration during tests

Humidity registered by ROSPHERE [%]
Humidity registered by ROSPHERE sensors 40.3145
30 40

40.3144 35

Humidity [%] 40.3143 30

25 25
Humidity [%]

40.3142
Longitude

20
40.3141
15
20 40.314
10

40.3139 5

0
40.3138
15
50 100 150 200 250 300 350 400 450 500 –5
Sample [-] 40.3137
–3.485 –3.4848 –3.4846 –3.4844 –3.4842 –3.484 –3.4838
Latitude

Figure 15 Temperature measurement during field inspection

Figure 12 Temperature registration during the tests
Temperature registered by ROSPHERE [°C]
Temperature registered by ROSPHERE sensors 40.3145
21 25
40.3144
LM35
20 SHT71 20
Average temperature 40.3143
19
15
Temperature [C]

40.3142
18
Longitude

10
40.3141
17
5
16 40.314

0
15 40.3139

–5
14 40.3138

13 –10
0 100 200 300 400 500 40.3137
–3.485 –3.4848 –3.4846 –3.4844 –3.4842 –3.484 –3.4838
Sample [-] Latitude

64
 Moisture measurement in crops using spherical robots Industrial Robot: An International Journal
Juan D. Hernández et al. Volume 40 · Number 1 · 2013 · 59 –66

Conclusion and future work Kim, Y., Evans, R. and Iversen, W. (2008), “Remote sensing
and control of an irrigation system using a distributed
The present paper summarizes the work carried out in order wireless sensor network”, IEEE Transactions on
to create a small, light and low-cost robot for moisture Instrumentation and Measurement, Vol. 57 No. 7,
monitoring in crop fields. The performed tests have validated
pp. 1379-87.
ROSPHERE’s capability to move along wide crop rows while Michaud, F. and Caron, S. (2002), “Roball, the rolling
obtaining and geo-referring environmental data. It has been
robot”, Autonomous Robots, Vol. 12 No. 2, pp. 211-22.
concluded that a smaller robot has to be designed for working
Michaud, F., Lafontaine, J.d. and Caron, S.A. (2001),
in narrow crop rows so as to guarantee the integrity of the
“Spherical robot for planetary surface exploration”,
plants.
Proceeding of the 6th International Symposium on Artificial
However, some important issues have to be addressed in order
Intelligence and Robotics & Automation in Space.
to fulfill fully autonomous mission requirements. Thus, regarding
Michaud, F., Laplante, J.F., Larouche, H., Duquette, A.,
autonomous guidance, the preliminary control algorithms only
Caron, S., Létourneau, D. and Masson, P. (2005),
require the magnetometer information in order to follow the row
“Autonomous spherical mobile robot for child-
(considering it as a straight line). The bearing is calculated on the
development studies”, IEEE Transactions on Systems, Man
basis of the initial and final positions of the rows, which are
and Cybernetics, Part A: Systems and Humans, Vol. 35,
obtained by using GPS. A short-range laser sensor will have to
pp. 471-80.
be added in order to determine the exact position of the row so
Patil, P., Vidya, H., Patil, S. and Kulkarni, U. (2011),
as to avoid any damage to the plants.
“Wireless sensor network for precision agriculture”,
Downloaded by New Mexico State University At 07:53 31 January 2016 (PT)

Considering its control features, an additional sensor that

measures the rotation speed of the sphere is required in order International Conference on Computational Intelligence and
to develop advanced controllers for extreme slippery surfaces. Communication Networks (CICN), 7-9 October, pp. 763-6.
Furthermore, a wireless link for external sensors will enhance Quanxing, Z., Wu, Chwan-Hwa and Tilt, K. (1996),
data accuracy and make connectivity easier. New methods “Application of fuzzy logic in an irrigation control
that allow the robot to measure moisture by using probes are system”, Proceedings of the IEEE International Conference on
still in early stages of development. Industrial Technology, (ICIT ’96 ) 2-6 December, pp. 593-7.
Rahangadale, V.S. and Choudhary, D.S. (2011), “On fuzzy
logic based model for irrigation controller using Penman-
References Monteith equation”, IJCA Proceedings on 2nd National
Amir Homayoun Javadi, A. and Mojabi, P. (2002), Conference on Information and Communication Technology
“Introducing August: a novel strategy for an (NCICT ) 22-25 November, Vol. 4.
omnidirectional spherical rolling robot”, Proceedings of Seeman, M., Broxvall, M., Saffiotti, A. and Wide, P. (2006),
IEEE International Conference on Robotics and Automation, “An autonomous spherical robot for security tasks”,
Vol. 4, pp. 3527-33. Proceedings of the IEEE International Conference on
Armour, R.H. and Vincent, J.F.V. (2006), “Rolling in nature Computational Intelligence for Homeland Security and
and robotics: a review”, Journal of Bionic Engineering, Vol. 3 Personal Safety, pp. 51-5.
No. 4, pp. 195-208. Shengju, S., Jichao, Z., Hao, W., Shoujun, C. and Qi, A.
Bicchi, A., Balluchi, A., Prattichizzo, D. and Gorelli, A. (2010), “Modeling and simulation of a spherical mobile
(1997), “Introducing the “SPHERICLE”: an experimental robot”, Computer Science and Information Systems/ComSIS,
testbed for research and teaching in nonholonomy”, Vol. 7, pp. 51-62.
Proceedings of IEEE International Conference on Robotics and Shinghal, K., Noor, A., Srivastava, N. and Singh, R. (2010),
Automation, 20-25 April, Vol. 3, pp. 2620-5. “Wireless sensor networks in agriculture: for potato
Bouttier, F., Mahfouf, J.F. and Noilhan, J. (1993), farming”, International Journal of Engineering Science and
“Sequential assimilation of soil moisture from Technology, Vol. 2 No. 8, pp. 3955-63.
atmospheric low-level parameters”, Journal of Applied Sutar, S. (2012), “Irrigation and fertilizer control for
Meteorology, Vol. 32 No. 8, pp. 1335-51. precision agriculture using WSN: energy efficient
Bruhn, F.C., Kratz, H., Warell, J., Lagerkvist, C.I., Kaznov, V., approach”, International Journal of Advances in Computing
Jones, J.A. and Stenmark, L.A. (2008), “Preliminary design and Information Researches, Vol. 1 No. 1.
for a spherical inflatable microrover for planetary Valente, J., Barrientos, A., del Cerro, J. and Sanz, D. (2011a),
exploration”, Acta Astronautica, Vol. 63 Nos 5/6, pp. 618-31. “A waypoint-based mission planner for a farmland coverage
Capraro, F., Patino, D., Tosetti, S. and Schugurensky, C. with an aerial robot – a precision farming tool”, paper
(2008), “Neural network-based irrigation control for presented at the 8th European Conference on Precision
precision agriculture”, IEEE International Conference on Agriculture (ECPA’11).
Networking, Sensing and Control, (ICNSC 2008) 6-8 April, Valente, J., Barrientos, A., del Cerro, J., Rossi, C., Sanz, D.
pp. 357-62. and Garzón, M. (2011b), “Practical issues and
Kayacan, E., Bayraktaroglu, Z. and Saeys, W. (2011), improvements in farmland coverage with aerial vehicles”,
“Modeling and control of a spherical rolling robot: a paper presented at the First Workshop on Research,
decoupled dynamics approach”, Robotica, Vol. 30 No. 4, Development and Education on Unmanned Aerial
pp. 1-10. Systems (RED-UAS 2011).
Ke, L., Huibo, J. and Mingyan, L. (2011), “ZigBee wireless Zhan, Q., Cai, Y. and Liu, Z. (2008), “Near-optimal
sensor network using physics-based optimally sampling for trajectory planning of a spherical mobile robot for
soil moisture measurement”, Third International Conference environment exploration”, IEEE Conference on Robotics,
on Communications and Mobile Computing, pp. 505-8. Automation and Mechatronics, pp. 84-9.

65
 Moisture measurement in crops using spherical robots Industrial Robot: An International Journal
Juan D. Hernández et al. Volume 40 · Number 1 · 2013 · 59 –66

About the authors Antonio Barrientos received the Msc Engineer degree in
Automatics and Electronics in 1982 and a PhD in Robotics
Juan D. Hernández obtained the Electronics Engineering in 1986, both from Universidad Politécnica de Madrid. In 2002
(BSc equivalent) degree from Pontificia Universidad Javeriana he obtained an MSc degree in Biomedical Engineering from
in Cali, Colombia. Currently, he is a PhD student in Automation Universidad Nacional de Educación a Distancia. He has been
and Robotics at Universidad Politécnica de Madrid, where he Professor of Robotics, Computers and Control Engineering at
works as a researcher at the Robotics and Cybernetics Research Universidad Politécnica de Madrid since 1988 and has been
Group. His research interests include software development working for more than 30 years in robotics, developing industrial
for micro-controller and FPGA architectures for embedded and service robots. Currently his main interest areas are
applications, such as those used in mobile robots, mechanics aerial and field robotics. He is author of several textbooks in
theories (kinematics and dynamics) and algorithms for robotics and manufacturing automation, and co-author of white
modelling and simulation of serial/branched rigid-body robotic books on the state of robotics and manufacturing in Spain. He is
systems. also co-author of more than 100 scientific papers in journals and
conferences. He has been reviewer of several international
Jorge Barrientos is studying for an Electronics and journals and currently is Editorial Board Member of the
Automation Engineering degree at Universidad Politécnica International Journal of Advanced Robotic Systems. Currently, he is
de Madrid. Since 2011, he has been a member of the the head of the Robotics and Cybernetics research group of the
Centre for Automation and Robotics, Technical University of
Robotics and Cybernetics research group of the Centre for
Madrid – Spanish National Research Council, and a senior
Automation and Robotics, Technical University of Madrid –
Downloaded by New Mexico State University At 07:53 31 January 2016 (PT)

member of the IEEE.

Spanish National Research Council.
David Sanz has a BSc degree in Telecommunication
Jaime del Cerro received the Msc Engineer degree in Engineering and an MSc degree in Robotics, both from
Automatic and Electronic in 1996 and the PhD in Universidad Politécnica de Madrid. Currently he is working on
Robotics in 2007 from Universidad Politécnica de Madrid. his PhD thesis, as a researcher at the Robotics and Cybernetics
He is Professor of Control theory at Universidad Europea Research Group of the Centre for Automation and Robotics,
de Madrid. His research is focused on GNC systems and multi Technical University of Madrid – Spanish National Research
robot systems applied to both aerial and ground vehicles. Council. His research interests include autonomous aerial
Jaime del Cerro is the corresponding author and can be vehicles, obstacle & avoidance systems and cognitive
contacted at: [Link]@[Link] architectures.

To purchase reprints of this article please e-mail: reprints@[Link]

Or visit our web site for further details: [Link]/reprints

66
 This article has been cited by:

1. Kelly T. Sanders, Sami F. Masri. 2016. The Energy-Water-Agriculture Nexus: The Past, Present and Future of Holistic Resource
Management. Journal of Cleaner Production . [CrossRef]
Downloaded by New Mexico State University At 07:53 31 January 2016 (PT)