Design and Efficiency Mapping of an Electric Drive for Mobile Robotic Container Platform for Use in Industrial Halls

Full text

Design and Efficiency Mapping of an Electric Drive
for Mobile Robotic Container Platform for Use in
Industrial Halls
Lukáš Krčmář, Ondřej Mach and Josef Černohorský
Faculty of Mechatronics, Informatics and Interdisciplinary Studies
Technical University of Liberec
Liberec, Czech Republic
lukas.krcmar@tul.cz, ondrej.mach@tul.cz, josef.cernohorsky@tul.cz
Abstract— This article discusses specific design approach to
the electric drive used in mobile robotics. The results are shown
at the example of autonomous robotic platform, which is
considered for use in interoperation logistics in manufacturing.
Major technical limitation with the current state of knowledge
are still the energy storage systems. Due to their limited specific
and volumetric density, combined with higher production cost it
is necessary to define critical points for significant energy
savings. One of the most significant power consuming parts of
mobile robots is its own drive unit. It is necessary to take into
consideration every single detail of the electric drive design.
From the driving dynamics simulations to the selection of
components such as motors, gearboxes and transmission topology
so as to achieve the optimal design with high efficiency.
Keywords— Autonomous mobile platform; control system;
measurement and data acquisition; electric drive; efficiency
mapping; lithium batteries
I. INTRODUCTION
The problem of material handling and transportation inside
the production facilities is now, at a time of implementation of
industry 4.0 principle, an often discussed topic [1][2]. Many
production facilities use their own specific containers for the
transport of material, which must be moved from storage to
production premises at the exact time intervals and vice versa.
Using the material handling equipment, which requires its
own personnel for transport is quite inefficient and sometimes
very slow. This problem solves the mobile robotic platforms
that is used to transport the material. Many factories already
uses guided or fully autonomous robotic container systems,
which supplies material for production line for all time.
Our problem is concern to autonomous robotic container
platform for input material handling in large industrial halls.
This platform is supposed to be moved on the flat floor with
inertial navigation system [3][4][5]. The platform
autonomously drives with the full container from material
supply line to the production machine and it returns back with
empty one. The travel speed is limited to 8 km·h-1. This value
is suggested as maximum driving speed for safety operation to
avoid very dangerous collisions between mobile platform and
staff. The weight of fully loaded container is 40 kg, empty
container weight is 12 kg and the mobile platform weight is
25 kg. The specific industrial container used in this application
is significantly higher than wider and therefore the big
challenge is in driving stabilization. This also leads to the
limited acceleration and braking to 1.11 m·s-2.
II. MOBILE PLATFORM CONCEPT
Fundamental requirements for the mobile platform for
material handling in industrial halls are made up of a balanced
mix of performance, range, maneuverability and production
costs. Floors of industrial buildings in which the platform is to
be used are assumed to have a smooth surface with a minimum
of unevenness, some of which are slip-resistant, particularly in
sloping planes.
As for indoor mobile platform chassis conception, a
wheeled chassis is considered as optimal. There are exist
numerous options of wheeled chassis concepts [6][7][8] such
as synchronous drive chassis, differential drive chassis, car
chassis with Ackermann steering or a tricycle drive chassis
come to mind.
The major disadvantage of synchronous drive platform for
this particular application is a complicated mechanical design
which does not offer here any major benefits.
The car chassis concept with Ackermann steering shown in
Fig. 1 brings very good energy efficiency, but its does not
allow a full 360° degrees on site rotation maneuverability of
the platform, which is one of the requirements for the platform.
The tricycle drive chassis concept in Fig. 2 has similar
features to the car chassis. In the case that only one wheel is
driven, and this wheel also determines the direction of travel it
would be possible to rotate tricycle platform on the spot.
The differential drive platform concept in Fig. 3 offers the
most benefits for this particular application. The only serious
disadvantage in this case is a low traction of two wheels.
• Low cost simple and robust design
• Relatively accurate odometry with integration error
• Minimum cornering losses
• Good maneuverability with rotation on the spot

Fig. 1. Car chassis concept with Ackermann four wheel
steering system of autonomous container platform [2].
Fig. 2. An example of tricycle drive chassis concept [9].
Fig. 3. An example of differential drive chassis concept for
autonomous mobile robot [10].
III. MOBILE PLATFORM DRIVING FORCE CALCULATION
For designing the high efficiency drive of the mobile
container platform it is necessary to know all technical
parameters of platform and the cargo itself as well as a precise
description of the driving cycle of the platform.
The driving dynamics of mobile platform are simulated
using generally known mathematical and physical apparatus
for a wheeled vehicle [11]. Equation is based on the
equilibrium of forces where on one side of the equation is the
available tractive effort reaching wheels Fw and on the other
side is the sum of all driving resistances Ri (1).
wi
i
FR
(1)
After substituting the driving resistances Ri and with the
use of the Newton's second law for acceleration resistance
w rl air acc g
F R R R R   
(2)
where Rrl is rolling resistance, Ra is air resistance, Racc is
resistance to acceleration and Rg is grade resistance of the
vehicle.
2
22
1
n
w i i
air i
d d d w
TJ
e
Z R m x m g sin
r r r





       




(3)
The acceleration resistance Racc is divided into two parts,
namely the acceleration of the sliding part and the resistance of
the rotating parts. Due to the small weight of the rotating parts,
the resistance of the rotating parts was neglected.
The tractive effort reaching wheels Fw can be expressed as
the sum of the electric motor torque Tm divided by the dynamic
radius of the wheel rd, where ii is gear ratio and ηi is overall
powertrain efficiency with engaged gear (4).
m i i
wid
Ti
Fr



(4)
The electric motor power Pm required to overcome the
driving resistances of the platform is based on (5) where v is
actual vehicle speed.
m
mt d
T
P F v v
r
   
(5)
TABLE I. DATA FOR ROLLING RESISTANCE CALCULATION
Calculation Parameters
Mobile platform
Full
container
Empty
container
e [m]
0.0077
0.002
0.002
r [m]
0.062
0.002
0.002
m [kg]
25
40
12
Calculated data is shown in the following Table II.

TABLE II. THE RESULTS OF THE CALCULATION
Parameter
Symbol
Unit
Value
Rolling resistance
Rrl
N
46.8
Grade resistance
Rg
N
68.32
Acceleration resistance
Racc
N
16.66
∑ of all resistances
∑Ri
N
131.78
Motor torque required
Tm
Nm
9.88
IV. ELECTRIC DRIVE FOR MOBILE PLATFORM
Autonomous mobile container platform due to the
anticipated workload should be equipped with maintenance-
free brushless drive with BLDC or PMSM motors with
advanced control system.
The proposed differential platform shares the torque between
the two independently driven wheels, the torque per motor is
equal to 4.94 Nm in worst case scenario. This value can be
achieved by conventional approach which involves the
integration of a compact planetary gearbox with the industrial
servodrive. The disadvantage of this approach is the
significant cost of increased design complexity and additional
losses in the transmission it seems to be appropriate design
solutions in the form of a single module drive unit with a pair
of independent in-wheel motors. The other design approach
which should be considered is an use of in-wheel motors.
There must be pointed out that the common design of gearless
drive usually has low efficiency at very low speeds.
Particularly if there is not available enough space for using
wheels with larger diameter which is necessary for installation
of a multi-pole in-wheel motors.
A. Traction drive
Due to the requirements for low cost of the system the
Leadshine iSV-B23180 compact industrial servo drive with an
integrated controller was chosen for this application. It is a
36 V low voltage BLDC drive with integrated encoder. The
control signal input is by the step-dir The internal PID
controllers of the servodrive are configurable via software.
B. Planetary Gearbox
BR automation 8GP30-060 is an industrial compact
planetary gearbox with a 10:1 gear ratio. The efficiency
reported by the manufacturer is close to 95%. The gearbox is
fully maintenance-free with a lifetime load. Due to the drive
mechanical design of the mobile platform, the minimal radial
and axial forces acts on the gearbox shaft. The producer
guarantees a maximum working life of 30,000 hours.
V. SERVO DRIVE PID CONTROLLER TUNING
The torque is transmitted to the wheels from gearbox shaft
through the ribbed belt. However, this elastic member of the
torque transmission chain causes difficulty in adjusting the PID
controller of the servo drive.
Fig. 4. Oscillations of belt drive in target position as a result of
autotuning tool supplied with the servo drive.
The default setting values of the PID controllers or the result of
the autotuning tool in the supplied software shown were not
usable in combination with the belt drive of the platform. For
this reason, the SISOTOOL toolbox in the Matlab environment
was used to identify and set up the controllers. Subsequently,
the PID controller setting was transferred to the servo drive.
TABLE III. PARAMETERS OF LEADSHINE ISV-B23180 SERVO DRIVE
Parameter
Symbol
Unit
Value
Rated power
Pn
W
180
Rated torque
Tn
Nm
0.6
Peak moment
Tm_max
Nm
1.5
Rated speed
nn
RPM
3000
Peak speed
nmax
RPM
4000
Weight
Mm
kg
1.54
TABLE IV. PARAMETERS OF BR AUTOMATION 8GP30-060 PLANETARY
GEARBOX
Parameter
Symbol
Unit
Value
Gear ratios
i
-
10:1
Rated output torque
Tg
Nm
15
Max. output torque
Tg_max
Nm
24
Max. input speed up to 7.5 Nm
nmax
rpm
4500
Max. input speed up to 15 Nm
nmax
rpm
4500
Max. input speed
nmax
rpm
13000
Noise
-
dB
58
Min. operating temperature
υmin
°C
-25
Max. operating temperature
υmax
°C
90
Weight
Mg
kg
0.9

VI. ELECTRIC DRIVE EFFICIENCY MAPPING
In order to determine the efficiency maps of the drive
system of the mobile platform was used a laboratory
dynamometer Dynofit ASD 6.3K-4. Two independent
measurements were made, one for servo drive only and the
second for the set of servo drive with planetary gearbox. The
most accurate measurement of the servo drive with gearbox
and belt drive was not performed at this time but it is scheduled
to the near future. Servo drive was powered during
measurement from the power supply set to battery nominal
voltage of 36 VDC.
VII. ELECTRIC DRIVE TESTING ON A TEST TRACK
The next step was to measure the values of the energy
consumption in real operational condition. In order to
determine the energy consumption from the traction battery, a
test track with a distance of 25 m was created. The track
contained a marked start and target position.
The following course is shown in Fig. 7. The mobile
platform with the empty container start from the starting
position and was autonomously navigated by its inertial system
to the target position. The drive path is straight, after 20th
second, at the end of path, the mobile platform with empty
container have to be parked between the two obstacles. This
maneuver was carried out between 20-35th second. Platform
rotation on the spot is not recorded as velocity. In the target
position the container was manually loaded by 30 kg load.
Subsequently, after 62th second the platform autonomously
maneuvered back from the target position between two
obstacles and in 76th second is already directed back towards
the starting position. The total path length is 50 m. The voltage
and current flows from drive only were recorded by the data
logger Junsi Powerlog 6s with 0.2 second sample rate.
With an empty container the acceleration power peak of
mobile platform is 180 W, when running at steady speed of
8 km·h-1, the mean power consumption of the drive is 58 W
with 65 % efficiency. With a fully loaded container, the
acceleration power peak of mobile platform is 195 W and the
steady speed mean power consumption is 76 W with 67 %
efficiency. The mean power value over the entire test track is
then 33 W and the mean drive efficiency is 52 %.
Fig. 5. An efficiency map graph of Leadshine iSV-B23180
servo drive (BLDC motor with integrated controller).
Fig. 6. An efficiency map graph of Leadshine iSV-B23180
servo drive with 8GP30-060 planetary gearbox. The two
blue lines represents the course of efficiency during the
test drive of the empty platform on the test track.
Fig. 7. Graph with the test track measured data, electric drive power consumption and mobile platform speed.

Fig. 8. A closer look at the laboratory dynamometer Dynofit
ASD 6.3K-4 station used for drive efficiency mapping.
VIII. CONCLUSIONS
Material transport in modern industrial halls is now, in the
time of implementing an ideas of smart factory of Industry 4.0
a very discussed topic. Mobile platforms for use in industrial
areas not only reduce dependence on human labor, but also
save space in production lines and warehouses thanks to the
need for significantly smaller transit corridors and dedicated
handling spaces.
The one of the most limiting technology for such mobile
platforms is still an energy storage system. Almost all mobile
robotic applications are somehow limited by the volumetric
and gravimetric density of their energy storage system.
Therefore, the focus on the maximum efficiency of mobile
platform drives and other consumptions is still very important.
We have mapped the efficiency of the proposed servo drive
and the servo drive assembly with the planetary gearbox. From
the measured data, the planetary gearbox efficiency can be
considered to exceed 90 % over the whole range. But the
measured Leadshine iSV-B23180 BLDC servo drive efficiency
is only about 70 % at its rated load which is not an industry
leading value. There should be also considered a significant no
load power consumption of the servo drive which is about
2.75 W.
The other servo drive parameters are, however, very good,
the control is reliable and drive itself is very quiet. The drive
also has a sufficient torque reserve for overloading the mobile
platform up to tested total weight of 150 kg.
ACKNOWLEDGEMENT
The research reported in this paper was supported by Project
FV10099 – Application of the principles „Industry 4.0“in the
spinning mills and by Student Grant Competition of Technical
University of Liberec.
REFERENCES
[1] S. Zug, S. Wilske, C. Steup and A. Lüder, "Online evaluation of
manipulation tasks for mobile robots in Industry 4.0 scenarios," 2015
IEEE 20th Conference on Emerging Technologies & Factory
Automation (ETFA), Luxembourg, 2015, pp. 1-7.
[2] P. Jandura, J. Černohorský, and A. Richter, “Electric Drive and Energy
Storage System for Industry Modular Mobile Container Platform,
Feasibility Study,” IFAC-PapersOnLine, vol. 49, no. 25, pp. 448–453,
2016.
[3] T. Fauser, S. Bruder and A. El-Osery, "A comparison of inertial-based
navigation algorithms for a low-cost indoor mobile robot," 2017 12th
International Conference on Computer Science and Education (ICCSE),
Houston, TX, USA, 2017, pp. 101-106.
[4] V. Fedak, F. Durovsky, R. Uveges, K. Kyslan, and M. Lacko,
“Implementation of Robot Control Algorithms by Real-Time Control
System,” International Journal of Engineering Research in Africa, vol.
18, pp. 112–119, 2015.
[5] J. Cernohorsky, P. Jandura, and O. Mach, “Mobile robot localization and
object description,” 2017 18th International Carpathian Control
Conference (ICCC), pp. 503-506, 2017.
[6] J. Bacik, F. Durovsky, M. Biros, K. Kyslan, D. Perdukova, and P.
Sanjeevikumar, “Pathfinder – Development of Automated Guided
Vehicle for Hospital Logistics,” IEEE Access, pp. 1–1, 2017.
[7] P. Brabec, R. Voženílek, V. Záda, and L. Beran, “Construction of a
Special Mobile Robot that Uses a Special Travelling Wheel,” MATEC
Web of Conferences, vol. 40, p. 03004, 2016.
[8] J. Hrbáček, T. Ripel and J. Krejsa, "Ackermann mobile robot chassis
with independent rear wheel drives," Proceedings of 14th International
Power Electronics and Motion Control Conference EPE-PEMC 2010,
Ohrid, 2010, pp. T5-46-T5-51.
[9] Robotic Autonomy, Carnegie Mellon University,
https://www.cs.cmu.edu/~rasc/RA/Images/TrikeBot-Sketch.jpg
[10] 2WD Mobile Platform for Arduino,
https://www.robotshop.com/uk/2wd-mobile-platform-arduino.html
[11] P. Jandura, J. Brousek, and M. Bukvic, “The concept of a highly
efficient powertrain for an electric vehicle with respect to vehicle
driving dynamics,” 2015 International Conference on Electrical Drives
and Power Electronics (EDPE), pp. 422-429, 2015.