Abstract: This study presents the processes undertaken in the
design and development of an intelligent omni-directional mobile robot
using four custom-made mecanum wheels. The mecanum wheel developed
consists of nine rollers made from delrin. All mecanum wheels are
independently powered using four units of precisian gear DC motors and
the wheel/motor assemblies were mounted directly to the robot chassis
made using an aluminum frame. A four channel high power H-bridge using 2
units of LMD 18200 motor drivers IC circuit was design, built and
interfaced to a BasicStamp (BS2) microcontroller board. Basic mobility
algorithm using BasicStamp software was developed to test the basic
mobility capabilities and test the qualitative view of the system's
mobility performance. An experiment was set-up to analyze the motion
characteristic of the mobile robot motion in Y-axis, X-axis and rotary
motion. Data from the experiment will be used for mathematically model
for mobile robot platform and speed controller modeling and design.
Mecanum wheel mobile robot provides a convenient platform for further
development in the mobile platform. The combination of mechanical design
on the wheel and chassis, motion control and multiple input/output
sensors allow the exploration of large number of control algorithm and
software to be implemented to the robot for practical applications.
Key words: Mecanum wheel, mobile robot, microcontroller, motion
control
INTRODUCTION
This project utilized the Mecanum wheel design pioneered in 1973 by
Mecanum AB's Bengt Ilon. Mecanum wheel is based on the principle of
a central wheel with a number of rollers placed at an angle around the
periphery of the wheel. The angled peripheral roller translates a
portion of the force in the rotational direction of the wheel to force
normal to the wheel directional. Depending on each individual wheel
direction and speed, the resulting combination of all these forces
produces a total force vector in any desired direction thus allowing the
platform to move freely in direction of resulting force vector, without
changing the direction of the wheel. Figure 1 shows a traditional
Mecanum wheel design by Ilon with the peripheral roller with 45[degrees]
degree slope held in place from the outside.
Using four of mecanum wheels provides omni-directional movement for
a vehicle without needing a conventional steering system Slipping is a
common problem in the mecanum wheel as it has only one roller with a
single point of ground contact at any one time.
Due to the dynamics of the mecanum wheel, it can create force
vectors in both the x and y-direction while only being driven in the
y-direction. Positioning four mecanum wheels, one at each corner of the
chassis (two mirrored pairs), allows net forces to be formed in the x, y
and rotational direction (Fig. 2).
[FIGURE 1 OMITTED]
A difficulty with this strategy is that there are four variables to
control three degrees-of-freedom. In this case the system is said to be
over determined and it is possible to create conflicts in the actuation.
As a result of the constraints associated with the mecanum wheel some
form of controller is required to produce satisfactory motion.
[FIGURE 2 OMITTED]
MATERIALS AND METHODS
As this was a complete Mechatronic project incorporating
mechanical, electronic and software development, the different areas
were developed synergistically thus allowing interactions between the
disciplines to be viewed and managed. It also meant that all three core
disciplines needed to be developed to a certain stage before any one
area could be further worked on. Although it was physically possible to
use other means to develop the core areas independently, a synergistic
approach tends to be more efficient. Even though this parallel design
approach was used, the areas of development shall be discussed in
sections assuming that other sections have already been completed to a
certain level and are referenced where necessary.
Developments and implementation: The development for this project
can be divided into the major process, the mechanical design for mecanum
wheel and mobile robot chassis, electronics design for 4 channel motor
driver and interfacing with BasicStamp microcontroller board and
software development for motion control.
Mechanical design: The mecanum wheel been develop consist of nine
roller with diameter of 110 mm. Each roller diameter is 20 mm at the
center and 16 mm at each end. All rollers are made by engineering
plastic call delrin. The roller was hold by roller holder made by
stainless steel and the center hub was made aluminum. Design structure
of the mecanum wheel shown in Fig. 3. All mecanum wheels are
independently powered using four units of precisian gear DC motor and
the wheel/motor assemblies were mounted directly to the robot chassis.
Typical mecanum wheel mobile robot platforms are square or
rectangular, attach with wheel with +45[degrees] roller and wheel with
-45[degrees] roller on each side. The omni-directional capabilities of
the platform depend on each wheel contact firmly with the surface and
some of the mecanum wheel mobile robots are equipped with suspension
system. For simplicity our mecanum wheel and motor assembly are mounted
directly on the platform chassis made from aluminum frame and metal
plate.
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
[FIGURE 5 OMITTED]
The size of mobile robot approximately 20 cm (width) and 30 cm
(length). Figure 4 show the design structure of the mobile robot.
Electronic design: Four channel bi-directional motor driver been
design to drive all four mecanum wheels. The specifications developed
for the necessary driver board were:
* The circuit should be compatible with a single logic-level PWM
input signal for speed control of each wheel and a single logic-level
input line for the direction of motor rotation for each wheel.
* The circuit should be able to operate with a high PWM carrier
frequency from the microcontroller (20 MHz) to provide inaudible
operation.
* The circuit would require four independent H-Bridge drivers for
bi-directional motion.
* Each H-Bridge driver circuit must be capable of providing
suitable continuous current at 12V DC.
The prototype motor driver was develop using 2 units of LMD18200 IC
manufactured by National Semiconductor that capable of 3 amps continuous
current at up to 55V DC and also had extra integrated features including
current sensing and thermal overload shutdown [2]. Figure 5 show the
single motor driver circuit.
The DC motors used in this platform have built-in 40:1 gear
reduction and speed at 65 RPM at 12VDC. The optical encoders provided
velocity information on each wheel to the micro-controller. A four
channel high power H-bridge driver board was interfaced to a BasicStamp
(BS2) micro-controller board. The overall system hardware architecture
(Fig. 6) shows the connections between hardware components of the mobile
robot platforms.
Microcontroller: In order to give the existing robot any
intelligent functionality some form of on-board processor was essential.
Microcontrollers are ideally suited for such an application as they are
compact, have many built-in hardware features such as timers and UARTS,
have a significant number of digital I/O lines and have low power
requirements. The essential microcontroller specification for this
project was its ability to generate four independent PWM signals. Other
general requirements were; high speed operation to ensure environmental
data could be processed at real-time. BASIC Stamp micro-controllers have
been chosen this project for well-known for their ease of use,
comfortable programming language and easy debugging using a PC. Figure 7
show the BasicStamp (BS2) microcontroller board.
Table 1 and 2 show the relationship between output data of the
micro-controller for speed (PWM) and bi-directional (DIR) control for
all four DC motors and the omni-directional motion of the mecanum wheel
mobile robot.
Experiment setup: With the use of the prototype motor driver board
and test software to programming the microcontroller output as list in
Table 2 and 3, the basic mobility control was gained via programming the
basic motion software to the microcontroller. This setup allowed the
following motions as well as the increase and decrease of speed.
* Forward--all four wheels forward in
* Backward--all four wheels move backward at the same speed
* Right slide--wheel 1 and 4 forward, wheel 2 and 3 backward
* Left slide--wheel 2 and 3 forward, wheel 1 and 4 backward.
* Clockwise--wheel 1 and 3 forward, wheel 2 and 3 backward
* Counter-Clockwise--wheel 1 and 3 backward, wheel 2 and 3 forward
[FIGURE 6 OMITTED]
[FIGURE 7 OMITTED]
The following list in Table 2, show the basic motion of mecanum
wheel mobile robot with their corresponding wheel direction. By varying
the individual motor/wheel speed we can achieve driving direction along
any vector in X-Y axis. The actuation required for these movements can
be seen in Fig. 8.
RESULTS AND DISCUSSION
The testing gave a qualitative view of the system's mobility
performance. The forward and reverse motion as acceptable but did not
utilize any function of the mecanum wheels. Likewise with rotational
motion, the system performed as would be expected of a standard
differential drive platform. The translational motion in X-axis,
however, was not acceptable as the platform would tend to wander in the
y-direction when attempting to traverse sideways. Investigation showed
that the motor systems were working adequately but the mecanum wheels
themselves and mobile robot chasis needed some minor modification.
A qualitative view of the platform's basic mobility for
translational movement showed an increase in performance but at this
stage some form of quantitative analysis was required to review further
minor modifications to the system especially on the mecanum wheel design
and close-loop speed control on each motor. Data from the experiment
will be used for mathematically model for mobile robot platform and
speed controller modelling and design.
[FIGURE 8 OMITTED]
The accuracy of direction and movement of the mobile robot depend
much on slip rate of the wheel on floor' various conditions. As are
result, the real position and orientation of the mobile robot deviate
from the original planned course or path. The other parameters that
influence the accuracy of movement of mecanum wheel mobile robot are
surface contact and traction. Because of slippage problem, motion
analysis for mecanum wheel type vehicle is complicate. Visual dead
reckoning for motion control and odometry are the method to reduce the
motion error.
Close loop control using model base method or classical cascade PID
for speed controller of mobile robot must be integrated into control
algorithm to maintain set-point speeds in both the x and y-directions.
The combination of these two speeds gives the actual speed and more
importantly direction of the robot.
Future work: Mecanum wheel mobile robot provided a convenient
platform for continued development. The combination of mechanical design
on the wheel and chassis, motion control and multiple input/output
sensors allow the exploration of large number of control algorithm and
software to be implemented to the robot for practical application.
We are currently developing a tele-operated mecanum wheel mobile
robot using RF module for data communication between robot and host
computer. The mobile robot will implement closed-loop wheel speed
control using fuzzy logic algorithm for robot omni-directional
kinematics and motion control.
The mecanum wheel also will be used as our platform for Autonomous
Guided Vehicle (AGV) in Flexible Manufacturing System (FMS) at Robotic
and Automation laboratory in KUKUM. The used of mecanum wheel AGV with
the lifting mechanism on the AGV platform will be part of FMS system and
optional with the current AGV that used differential drive and gantry
robot at the ASRS station
CONCLUSION
This paper presents an overview over the primary design stage of
omni-directional mobile robot using mecanum wheel. The strength of this
wheel is the enhanced maneuverability of the mobile robot that needs
extreme maneuverability in congested environment. Mecanum wheel mobile
robot research addresses many problems in robotics such as sensor
integration, real-world modelling, actuator and sensor control, path
planning and navigation, task-level planning and execution and the
control of the robotic system as a whole. Moreover, building mecanum
wheels mobile robot provides a stringent test bed for new concepts and
approaches in both mechanical design for mecanum wheel and overall
mobile robot chassis and also the design for electronic hardware and
software.
This design and development of an omni-directional platform, using
mechatronics system and mecanum wheel to implement intelligent
behaviours and maneuvers, with the help of a microcontroller interfaced
with sensors.
REFERENCES
1. Fiegel, O., A. Badve and G. Bright, et al., 2002. Improved
mecanum wheel design for omni-directional robots. Proc. Australasian
Conf. Robotics and Automation, 27-29 Nov., pp: 117-121.
2. Regan, T., 1999. A DMOS 3A, 55V H-Bridge: LMD18200. National
Semiconductor Application Note 868.
3. Phillips, J.G., 2000. Mechatronic design and construction of an
intelligent mobile robot for educational purposes. Master of Technology
Thesis, Massey University, Palmerston North, New Zealand, pp: 150.
(1) Jefri Efendi Mohd Salih, (1) Mohamed Rizon, (1) Sazali Yaacob,
(1) Abdul Hamid Adom and (2) Mohd Rozailan Mamat
(1) School of Mechatronics Engineering, Kolej Universiti
Kejuruteraan Utara Malaysia 01000 Kangar, Perlis, Malaysia
(2) Terengganu Advanced Technical Institute, Jalan Panchor, Teluk
Kalong 24000 Kemaman, Terengganu, Malaysia
Corresponding Author: Mohamed Rizon, School of Mechatronics
Engineering, Kolej Universiti Kejuruteraan Utara Malaysia, 01000 Kangar,
Perlis, Malaysia. Tel: 60-4-9798333, Fax: 60-4-9798334
Table 1: Switch control on microcontroller output
Output Pin Data
0 PWM input for M1 (High = 100% PWM)
1 Direction input for M1 (High = Clockwise, Low = Counter
Clock-Wise)
2 PWM input for M2 (High = 100% PWM)
3 Direction input for M2 (High = Clockwise, Low = Counter
Clock-Wise)
4 PWM input for M3 (High = 100% PWM)
5 Direction input for M3 (High = Clockwise, Low = Counter
Clock-Wise)
6 PWM input for M4 (High = 100% PWM)
7 Direction input for M4 (High = Clockwise, Low = Counter
Clock-Wise)
Table 2: Motor control for basic motion
M 1 M 2 M 3 M 4
PWM_1 DIR_1 PWM_2 DIR_2 PWM_3 DIR_3 PWM_4 DIR_4 Basic Motion
High High High High High High High High forward
High Low High Low High Low High Low backward
High Low High High High High High Low right slide
High High High Low High Low High High left slide
High High High Low High High High Low Turning
clockwise
High Low High High High Low High High Turning counter-
clockwise