International Journal of Mechanical & Mechatronics Engineering IJMM E-IJENS Vol:10 No:04
Review of Automated Machines towards
Devising A New Approach in Developing
Semi-Automated Grass Cutter
Sivarao, T.J.S.Anand, Hambali, Minhat, Faizul

Abstract — This article presents a review of researches done
on the subject of automated tractor. An autonomous tractor is a
vehicle that can operate without or with minimal human
control, self-propelled and guided automatically along a desired
path. The benefits from such a system are useful for agriculture
industry by reducing labor cost and time, as well as improving
output efficiency by eliminating human errors. Many
researches and inventions have been made, with the results
ranging from successful, encouraging to some that are
impractical for commercial implementation for certain reasons.
These implements include sensor, global navigation satellite
system, machine vision, laser triangulation, ultrasonic
transmitter and geomagnetic controller, as well as actuator and
servo motor.
Index Term— Automated grass cutter, Proportional
Integral Derivative (PID), Geomagnetic Direction S ensor
(GDS ), Mower.
I. INT RODUCT ION
Lawn mowing is an important part of the process to keep
terrain with grass well-manicured and tidy. Places such as
soccer field, garden, home lawn, golf course and many others
require constant lawn mowing and grass cutting. One of the
complicated tasks is mowing, with large amount of time and
effort needed to complete it. Depending on geographical
location and weather trends such as rainfall and temperature, it
is generally necessary to move a lawn on a weekly basis. The
duration to complete the task takes from minutes to hours,
depending on the size of the particular lawn and the mo ving
equipment available. The required long hours and the
complicated job would make the operator to succumb to
fatigue. It is a fact that humans have some physical limitations
regarding environmental factors such as weather conditions,
including temperature, pressure, humidity and so on which
T his work was supported in part by the University T eknikal Malaysia
Melaka (UT eM) towards working on UT eM funded project titled
„engineering of mechanical control kit for unmanned grass cutter‟,
Project number: PJP/2009/FKP(8A)-S536.
Ir. Sivarao, Anand, Hambali, Minhat, are the academic members while
Faizul is a lab technician at Faculty of Manufacturing Engineering,
Universiti T eknikal Malaysia Melaka (UT eM), Hang T uah Jaya, 76100
Durian T unggal, Melaka, T he correspondence author can be contacted
via e-mail: kpe_siva@yahoo.com or sivarao@utem.edu.my).
105204-9191 IJMME-IJENS © August 2010 IJENS
1
causes low blood pressure and poor fluid intake [1].
In committed situations, robots can do and should do the
job for us. Robots are not only used in outer space, but, there
are many places on earth where it is too dangerous for man to
work at like in the deep sea, nuclear plants or waste dumps.
Hence, there is also a growing need for robots in supporting
human beings in their daily life at home or at work [2]. These
are also reasons for the limited workforce available in this
sector. The operation of a lawn is a skill and labor intensive
task and furthermore the shortage and aging workforce in
agriculture results in a decrease of skilled machine operators.
Therefore, the development of driverless lawn tractor is of
commercial significance and societal importance.
There has been a need for some time for a lawn tractor that is
truly automatic. Although many types of automatic lawn
mowers have been proposed in the past, none have met with
any real degree of commercial success [3]. One of the reasons
being that the proposed systems cost too much, that many
small landscaping operators would not be able to afford. For
those in developing countries where the cost of labor is low,
the investment against return would not be favorable. Besides
that, systems that apply technology such as Global
Positioning System (GPS) and motion-sensor are too
sophisticated for any ordinary employee to operate. This paper
reviews the previous inventions done by other researchers in
finding the various automated tractors and the newly proposed
cost effective method for the simple and useful automated
tractor.
II. REVIEW OF PREVIOUS W ORK
Conventional tractors are manual driven, where a driver is
required to maneuver the direction of the tractor on the desired
course. For activities such as in agriculture seed planting or
regular grass cutting process, the path traveled by the tractor
is mostly fixed in a certain pattern predetermined by the driver
or operator. Therefore, the criteria to develop an autonomous
tractor has the primary requirement that the tractor be guided
to travel on a desired path with minimal human intervention.
With the development of technology, many scientists and
researchers have studied, designed, or invented various
autonomous vehicles which can self-propelled and guided
automatically while human only have to monitor or control
through some remote means.
I J ENS
 International Journal of Mechanical & Mechatronics Engineering IJMM E-IJENS Vol:10 No:04 2

From most of the researches, many types of guidance „machine vision‟ video camera scans the immediate
systems and obstacle detection system are developed and surroundings and watches for obstacles. The global
applied to the vehicle, such as GPS, vision capture by camera, positioning system determines the tractor‟s precise location on
infrared or sonar wave sensor, and all these devices collect the the field. The inertia sensor senses the movement of the tractor
data and input information and send for analysis and in the field and calculates its path. The computer guidance
calculation by computer which then convert it into command control uses „sensor fusion‟ to combine all of the information
for controlling the actuator and motor [4-6]. When the actuator coming from the different sensors. The electro-hydraulic
and motor are activated, it will control the wheel steer with steering valve receives orders from the computer and controls
certain angle or direction and the sensor attached to the the steering to cater the precise need [8].
wheels and actuator will give the feedback information to the
main computer to analyze and give a response or command for
the next step. This cycle will repeat until the job is finished.
These would be a good solution for autonomous vehicle but
since the cost of the installation and investment is higher, so it
cannot be a popular commercial product, and the maintenance
cost would be high as well. Operator who controls this system
also needs professional and technical skill training, and it
becomes expenditure which requires more time and money.
From the U.S. Patent by Pansire, (1980), an invention of self-
propelled and self-guiding lawn mower was invented [7]. A
fully automatic lawn mower which is self-propelled and guiding
based on the principle of magnet following a passive ribbon or
wire of ferromagnetic material. The concept of providing
guidance for a machine utilizing the principal of a magnet
located on the machine predetermined path defined by a
passive wire of ferromagnetic material or permanent magnet Fig. 1. Basic elements of agricultural vehicle automation system (Reid et
that impregnated material laid on or in ground. The drive al, 2000).
wheels are coupled to a continuously rotating drive shaft by
individual solenoid actuator, the drive shaft is turned by a
battery powered electric motor and the guidance system
generating steering command through activating and
deactivating the actuators. This invention is limited with the
magnetic force is enough or not to attract and maintain the
lawn mower stable and the travel path is limited by the
predetermined path.
Reid, et al, (2000) from North America had done some
research on an agricultural automatic guidance [8]. This was
then the recent research in agricultural vehicle guidance
automation in North America. The conceptual framework of an
agricultural vehicle guidance automation system as shown in
Fig. 1 which includes navigation sensors together with
navigation planner, vehicle motion models, and steering
controllers as can be visualized in Fig. 2. In the research, they
had studied the various types of electronic device and steering
controller. For the steering controller, the actuator which Fig. 2. Sensor fusions on an agricultural tract or.
converts the control signal from a feedback controller to an
appropriate mechanical adjustment in steer angle. Hydraulic In 2002, Batavia et al., was carried out a research on
steering system was developed around 1980, after that, the autonomous coverage operation in semi-structured outdoor
electro-hydraulic steering system was developed for environment [9]. This study focused on the comprehensive
agricultural vehicles. In 1998, Proportional Integral Derivative navigation system capable of extended coverage operations in
(PID) steering controller for agricultural tractor guided by a semi-structured environments. The system is able to track
Geomagnetic Direction Sensor (GDS) was further developed to specified paths with high accuracy, detect small obstacles
achieve suitable steering performance. The wheel position reliably, and plan coverage patterns to completely cover a
sensor calculates the angle that the wheels are turned. The specified area, where, the user interface can be displayed

105204-9191 IJMME-IJENS © August 2010 IJENS

I J ENS
 International Journal of Mechanical & Mechatronics Engineering IJMM E-IJENS Vol:10 No:04
directly on vehicle or remotely. Human intervention was
occasionally necessary due to persistent obstacle detection
false positives and hardware failures. Future work will
concentrate on improving localization to reduce the
dependence on high accuracy (and high cost) GPS, and
additional testing.
An automatic guidance for agricultural vehicles in Europe
research was done by [10]. This research was concerned with
automatic guidance of agricultural vehicles with implements of
various solutions. One of the guidance systems for vehicles
consist sensor, controller, and actuator. There were several
sensors inserted in the systems and the controller, PID type,
translates the sensor position deviation signal into a voltage
which is used to open a valve, forcing the hydraulic ram in the
steering circuit to change the steer angle of the front or rear
axle, in case of side shifting the implement, an additional ram to
adjust the position of the implement relative to the tractor.
Automatic guidance systems witnessed in Fig. 3 for side-
shifted implements have a limited control band and, as a
consequence, still need a driver to steer the vehicle within this
band to enable automatic correction of position deviations
[11]. The driver must keep the path and compensate lateral
forces produced by implement shifting.
Fig. 3. Modular structure for a system to guide agricultural vehicle
autonomously. T he figuratively portrayed components indicate available
elements [11].
A study on guidance of a forage harvester with GPS was
done by Stoll and KutzBach, (2000) [12]. A self-propelled
forage harvester was the test vehicle for this research; it was
equipped with an automatic steering system, and Real Time
Kinematic Global Positioning System (RTK-GPS) was used as
the positioning sensor. The GPS measurements were exported
to the navigation system. The adjustment signal derived from
the steer angle calculations are transmitted via the parallel port
adapter to the hydraulic steering system. A proportional valve
105204-9191 IJMME-IJENS © August 2010 IJENS
3
controls the oil flow in the steering cylinder and thereby the
positions the steering wheels accordingly and the position of
the vehicle is fed back by GPS measurements. The steer angle
is limited to ±10.6º and a potentiometer mounted at the left rear
wheel side measures the actual steered angle and transmits the
information back to the hydraulic steering control system. The
measured steering angle of the left rear wheel was converted in
the steering system to the mean steering angle. The result
showed that the positioning accuracy was sufficient for
vehicle guidance even though the dynamics of the high
mounted GPS antenna were not compensated. A standard
deviation of driving precision of less than 100 mm is acceptable
for several agricultural applications; however, further research
will consider methods for improving driving performance.
Nelson, (1997) obtained the U.S. Patent for an invention of
automated lawn mower [13]. A self-propelled lawn mower
which utilizing a rotating directional loop antenna to determine
its positioning within the cutting area by measuring the angle
between transmitters placed in the cutting area. The orientation
is determined by comparing present and previous positions;
the stored path information is compared with the calculated
position to determine the steering signal which directs the
mower to move in desired path. The direct current drive motor
replaced with a clutch and gear mechanism to the drive wheel
directly from the engine, an actuator would select forward or
reverse gears. An operator needs to monitor and control the
process via the remote control and computer. A further
improvement of the information transmitting is required and the
safety features need to be added.
A research on an agent of behavior architecture for
unmanned control of a farming vehicle was done by [14]. They
presented a hybrid agent of behavior architecture to deal with
the autonomous navigation of a farming vehicle. The vehicle
autonomously works local piloting; follow a tentative path
predefined by human operators through a set of intermediate
positions. A wireless LAN is used to get a bi-directional
communication between the vehicle and human supervisor.
Sensor-fusion algorithms are proposed to overcome the lack of
GPS signals so as to obtain continuous and precise
positioning. The autonomous tractor focused on the st eering
wheel and on the brake and clutch pedals, maintaining the
manual driving mode. Different solenoids/electro-valves and
actuators were integrated for the steering control system. In
the experiment, gradual action was achieved by pulse width
modulation (PWM) control of the three-way-valve that
commands the steering hydraulic circuit and its schematic is
shown in Fig. 4. A closed-loop control system is required to
reach and keep the steer angle reference. It was found that, the
control response over-shoot resulting in delayed dampen in
the hydraulic steering system which was caused by non-linear
characteristics of the hydraulic system.
I J ENS
 International Journal of Mechanical & Mechatronics Engineering IJMM E-IJENS Vol:10 No:04 4

resources). (3) Turns are the number of turns made by the

robot as it moves. The number of turns should also be as low
as possible for the same cost reasons as mentioned above.
Similarly Wong et al (2003) studied the topological coverage
algorithm for mobile robots [17]. It uses natural landmarks in
the environment to construct a planar graph representing a
decomposition of reachable surface into simple sub regions
that can be covered by a zigzag pattern. The grid based
methods generally used for navigation in coverage
applications can require considerable memory and computation
and is susceptible to odometry error. On the other hand, the
topological map used in this paper is proven to be robust
against sensor and actuator noise. It is also compact as its
resolution corresponds to the complexity of the environment.

Fig. 4. Schematic of the hydraulic system (Garcia-Perez, L., et al,

2007).
A model recognition and validation for an off-road vehicle
electro hydraulic steering controller was done by Zhang, Q.
(2002). Automated steering control is essential for autonomous
off-road vehicles which use an electro hydraulic (E/H) actuator
to implement the steering control as can be seen in Fig. 5. The
design and validation of an E/H steering controller through a
combination of system identification, model simulation and
field test was presented. A kinematic model of the steering
linkage geometry provided the gain between the hydraulic
actuator and the front wheel. The system model was used to
close the steering control loop based on the feedback signal
from the hydraulic steering actuator rather than from the front
Fig. 5. Hydraulic steering actuating system on the test vehicle (Zhang et
wheels. The simulation and vehicle test result indicated that,
al, 2001).
the developed steering controller was capable of efficiently
handling the non-linearity and dynamic asymmetry of the E/H
steering control system. It was also found to be capable of
achieving prompt and accurate steering control on an off-road
vehicle regardless of the nonlinear nature and the changing
ground conditions [15].
Recently in 2008, Cohen et al. tried the simulation of moving
lawn area which is continuous planer area by a mobile robot.
The robot is given as input a bitmap of a known geometric area
and derives an optimal covering path by implementing and
improving the well known on-line Full Scan Spanning Tree
Covering (STC) algorithm [16]. Their algorithm is shown in Fig.
6.
Fig. 6. Grid approximation of a given work area and the spanning tree
An efficient area coverage algorithm should meet several
(Cohen et al, 2008)
performance measurements. (1) Run-time should be fast and
efficient – which is accomplished by the robot being able to III. PROPOSED AUT OMAT ED MODEL
calculate its movement on-line. For very large areas if run-time In a further study of the subject, a field study simulation was
is slower than linear, the robot will not be useable. (2) The carried out before mechanizing a full scale semi-automated
over-cover measurement is the relative size of the area that is grass cutter. The proposed low cost solutionised principle
covered more than once by the robot. This measurement would be the reference in developing the other conceptual
should be as low as possible. As the over-cover measurement designs in nearest future. This proposed project has proven to
grows, the overall time to cover the entire area rises, which achieve a relatively low cost solution in grass cutting with a
increases the cost of using the robot (money, energy, time and simple but functional, cable-guided mechanism.

105204-9191 IJMME-IJENS © August 2010 IJENS

I J ENS
 International Journal of Mechanical & Mechatronics Engineering IJMM E-IJENS Vol:10 No:04
Fig. 7. T he motion of the tractor .
It simply follows the principle of tying a cow with long rope
to a pole in the middle of a field. When the sun sets, the cow
would spiral the rope around the pole while grazing and forces
the cow to move to pole centre. The principle of the proposed
semi-automated tractor system is shown in Fig. 7 resembles the
same grazing cow principle inversely. It is to be operated
through a spool that is fixated at the centre of the field and
connected to the tractor by steel cable. This way, the tractor
would move in a circular motion on the field while unwinding
the cable from the spool. Besides that, to prevent the tractor
being pulled towards the spool, a spring mechanism is installed
to pull the wheels to the opposite direction of the turn. Thus,
the cable would be pulled and tensed resulting in the tractor
moving in a radial. Once it reaches the extreme largest diameter,
then the continuous movement of the tractor will spiral inwards
as to perform second level cutting of the same profiled field.
From the proposed simulation results, it was found that the
model works fine with rolling out the spool while cutting
outwards and rolls back inwards when the rope reaches to the
longest end.
IV. CONCLUSION
From the critical review of the related literature, advantages
of each system are studied with their respective design
constraints, cost and limitations . Therefore, a strategized
system is proposed to develop a successful automated grass
cutting systems, in which the proposed could be implemented
commercially with minimal production and operating cost.
Thus, the grass of a very spacious field can be cut easily in
circular motion without having human to operate the tractor.
On the other hand, since the tractor is unmanned, then,
whether sunny or rainy day will not effect the cutting
operation and the tractor can be continuously operated almost
at any whether conditions. But then, since the cable guided
tractor may be very traditional and not safe enough, therefore
more investigations are to be carried out to employ the same
principle without a cable. For this, detailed conceptual design
study is to be carried out and published in next nearest future.
A CKNOWLEDGMENT
The authors/researchers would like to thank the top level
105204-9191 IJMME-IJENS © August 2010 IJENS
5
management of Centre for Research & Innovation
Management of UTeM for providing the initial research grant
to initiate the semi-automated crass cutter. Also, the authors
are indebt of thanking Faculty of Manufacturing Engineering,
especially the top level management, Dr. Rizal Salleh, Dr. Mohd
Warikh and Mr. Akramin for providing continuous support
and encouragement towards completion of the first phase of
the project.
REFERENCES
[1] Bravo, R., “T ired From Mowing the Lawn”, Journal of Pediatric
Health Care, 24: 2010, 123–126.
[2] Schierer, E. “Humanoid Robots for Human Life Support”,
Proceedings of IFAC Conference on Supplemental ways for
improving International stability through automation 15 -17 June
2006, Ed. P. Kopacek, 101 – 105, Elsevier.
[3] Arkin, E.M., Fekete, S.P., Mitchell, J.S.B. “T he lawnmower
problem”, Proceedings of the 5th Canadian Conference on
Computational Geometry, 1993, 461-466.
[4] Arkin, E.M. Fekete, S.P. Mitchell, J.S.B. Approximation
algorithms for lawn mowing and milling, Computational Geometry
17: 2000, 25-50.
[5] Sipahioglu, A. Kirlik, G., Parlaktuna, O. and Yazici, A., Energy
constrained multi-robot sensor-based coverage path planning using
capacitated arc routing approach, Robotics and Autonomous
Systems 58: 2010, 529-538.
[6] Kim, T and Sarma S.E., “Optimal Sweeping Paths on a 2 -Manifold:
A New Class of Optimization Problems Defined by Path
Structures”, IEEE T ransactions on Robotics and Automation, 19
(4): 2003, 613-636.
[7] Pansire, D.G. “Self-propelled Self-guiding Lawn Mower.” U.S.
Patent 4, 1980, 180,964.
[8] Reid, J.F., Zhang. Q., Noguchi, N., and Dickson, M. “ Agricultural
Automatic Guidance Research in North America.” Computers and
Electronics in Agriculture. Vol. 25, 2000, pp. 155-167.
[9] Batavia, P.H., Roth, S.A., and Sanjiv Singh . “ Autonomous
Coverage Operations in Semi-structured Outdoor Environments.”
Unpublished Research, Robotics Institute Carnegie Mellon
University, 2002.
[10] Keicher, R. and Seufert, H. “Automatic Guidance for Agricultural
Vehicles in Europe.” Computers and Electronics in Agriculture. 25:
2000, 169-194.
[11] Jahns, G. “Automatic Guidance of Agricultural Field Machinery.”
Proceedings of the Joint International Conference on Agricultural
Engineering & T echnology Exhibition, 1997, Dec. 15–18,
Bangladesh, 70–79.
[12] Stoll, A. and Kutzbach, H.D. “Guidance of a Forage Harvester with
GPS.” Precision Agriculture. 2: 2001, 281-291.
[13] Nelson, R.G. “Automated Lawn Mower.” U.S. Patent.5, 1999,
974,347.
[14] Garcia-Perez, L., Garcia-Alegre, M.C., Ribeiro, A., and Guinea, D.
“An Agent of Behaviour Architecture of Unmanned Control of a
Farming Vehicle.” Computers and Electronics in Agriculture, 2007,
60: 39-48.
[15] Zhang, Q., Wu, D., Reid, J.F., and Benson, E.R. “ Model
Recognition and Validation for an Off-road Vehicle Electro
hydraulic Steering Controller” Mechatronics. 12, 2002 845-858.
[16] Cohen, M.W., Sirotin, I. and Rave, E. Lawn Mowing System for
Known Areas, International Conference on Computational
Intelligence for Modelling Control and Automation, CIMCA, 2008,
539 – 544.
[17] Sylvia C. Wong, S.C., MacDonald, B.A., “A topological coverage
algorithm for mobile robots”, Proceedings of the 2003 IEEE -
International Conference on Intelligent Robots and Systems, Las
Vegas, 2003, 1685–1690.
I J ENS