Servo system [Picture]
[Dissassembled servo system]
[Servo system functional scheme]
For the purpose of investigating of complex electrical drives, a
laboratory servo system has been built, which is capable of independent
and adjustable generation of variable load and transmission elasticity,
friction and backlash.
The two identical permanent-magnet synchronous motors are used,
one as a driving machine and the other as a loading one. They are
coupled by a replaceable elastic shaft. Five elastic shafts of equal
length and different diameter may be replaced in order to vary the
transmission stiffness coefficient. Flywheels with replaceable inertial
disks are mounted on the shafts of both motors, thus allowing arbitrary
adjustment of the motor and load moment of inertia. The driving/load
machine can be blocked by tightening the flywheels to the drive
Braking devices for friction generation in the axial and radial
sliding bearings are placed on the load side of the drive. A rolling
bearing may occasionally be substituted for the axial sliding bearing.
The desired fricton force level can be obtained by screwing/unscrewing
pressing screws. The torque sensor is mounted behind the braking
device on the load side of the drive. The angle of backlash may
be varied by exchanging the cotters of four different widths in
the clamp on the load side of the drive. However, if screws are
used instead of exchangeable cotters to connect the shaft and clamp,
a drive without backlash is obtained.
The superimposed speed, position, and force controllers are implemented
in the control computer. The motor position is measured by incremental
encoders with the accuracy of 0.0036o = 13' (100000 cps) providing
good insight in to the friction dinamics.
Mobile robots Pioneer DX, AT
Two Pioneer 3DX, one Pioneer 2DX, and one Pioneer 3AT mobile platforms built by MobileRobots Inc. are used in our research.
For more information please refer to the link above.
Robosoccer team [
Robot soccer is a game similar to the "classical" soccer, but the players
are robots. Each team consists of five robot players (7.5x7.5 cm size).
Robots have to be controlled remotely from the host computer (using
wireless communication). The color camera connected to the computer
is used for getting the information about the position of individual
players and the ball. The match between two teams (each having its
own camera and computer is running without human intervention, with
the exception of foul or stalemate situations.
Thus, robotic soccer is a challenging research and pedagogical domain
that involves many technical and or scientific problems. Fast real-time
and accurate image processing algorithms are required. Mechanical and
electronic systems must be robust and reliable. Low-level control system
of robot players must be able to control the players precisely enough,
guarantying non-oscillatory and fast movement of the robots. And finally,
high-level strategic control usually involves some aspects of AI.
Ball and beam [
The control system of the ball and beam process should enable the ball
positioning at any point on the beam. A ball is placed on a beam, where
it is allowed to roll with one degree of freedom along the length of the
beam. A DC motor is coupled to the beam via the nylon line, which enables
control of the beam angle. To control position of the ball along the beam
two variables have to be measured: ball position and beam angle. The beam
is coated with resistive material, which is supplied with voltage. This
serves as a potentiometer arrangement for ball position measurement. Beam
angle can be estimated, but it is much more reliable to measure it.
Measuring is currently implemented using simple, but robust vision system
that consist of a low cost USB web camera and a PC.
Magnetic levitation [
The apparatus includes laser feedback and high flux magnetics to affect
large displacements and provide visually stimulating tracking and regulation
demonstrations. The system is quickly set up in the open loop stable and unstable
(repulsive and attractive fields) configurations shown. By adding a second
magnet, two SIMO plants may be created, and by driving both actuators with
both magnets, MIMO control is studied. The inherent magnetic field nonlinearities
may be inverted via provided real-time algorithms for linear control study or
the full system dynamics may be examined. Disturbances may be introduced via
the second drive coil for demonstrating system regulation in SISO and SIMO
For more information visit the above link.
The laboratory pump plant
has been built with the purpose of investigating the effects in
processes of fluid transfer and storage and especially the effects
of interconnected variables in MIMO processes. The plant consists
of two 2 m high columns, which are connected by pipes. Two pumps
driven by the squirrel cage induction motors controlled by the frequency
converters provide the water flow through the plant. Water circulates
in a closed loop. By closing and opening manual valves, several
different process configurations can be arranged, which make the
plant very flexible. There are several sensors of basic process
variables (pressure sensors and flowmeters). All of them are standard
industrial sensors (not laboratory sensors) and we have taken care
that they have different principles of operation. For example, there
are three different types of flowmeters: differential pressure flowmeter,
inductive flowmeter and flowmeter based on Coriolis force. There
are also two controlled valves, which are used for disturbance generation
in controlled process loop.
Distillation column [Picture]
A batch distillation
column has been built with the purpose of investigating advanced
control algorithms on highly nonlinear and time variable processes.
Laboratory distillation column has been designed and tested in Croatian
pharmaceutical company PLIVA. The column is operated as a batch
distillation column, which means that the column is introduced with
'batch' and then the distillation process is carried out. The column
can operate at a constant or variable reflux ratio. In the first
case, the composition of the distillate will be time variable, while
in the second case it is possible to achieve constant composition
of the distillate.
Distillation can be carried out on binary mixtures, as well as on
solutions that are more complex. The quantity of distillate is limited
by the volume of the column reboiler (400 ml), while the achievable
purity of distillate is limited by the number of column trays (10).
Three-mass oscillatory system [Picture]
The three mass oscillatory
system is realized by three carts on a rail connected with springs.
Carts are designed in such manner that weight of each cart can be
easily changed by putting or removing extra weighting brass plates.
Each brass plate weights 0.5 kg. Springs can be mounted interchangeably
and are supplied in three nominal stiffnesses: 175 Nm, 395 Nm and
788 Nm. Thus resonance frequencies of the oscillatory system can
The oscillatory system is driven by the brushless servo drive which
is connected to the endmost spring by a short lever (10 cm) mounted
on the motor shaft in order to increase its' output displacement.
Servo drive can be controlled manually by using simple function
generator, or by using a PC desktop computer with PCL - 812PG acquisition
card. Only displacement, and therefore the velocity, of the endmost
cart can be measured. Displacement measurement is realized by using
the incremental encoder.
This laboratory model can be used for modelling and identification
as well as control of the poorly damped oscillatory systems.
The laboratory shell-and-tube heat exchanger has been designed for
testing the properties of advanced identification and control algorithms.
It enables the examination of the effects of dead-time and process
parameter variations. Water from the cold water tank is pumped into
the pipeline by means of a pump. Cold water is heated in the heat
exchanger, which consists of a number of parallel, mutually isolated
A specially designed delay pipeline connected to the heat exchanger
outlet ensures the required process dead time. Cold water from the
fresh water supply network may be supplied to the cold water tank
continuously, thereby maintaining constant water temperature in
Water in the hot water tank is heated by an electric heater to the
desired temperature, which is maintained by a thermostat. Hot water
driven by another pump, which is operated by a frequency converter
supplied three-phase squirrel cage motor, flows through the heat
exchanger in a direction opposite to the cold water flow.
Water temperature is measured by resistance thermometers and water
flow by the flow meters. Signals of all temperature and flow measuring
devices are conducted to a PC via an I/O card. Cold water temperature
at the heat exchanger outlet may be altered by varying the hot water
flow, which is linearly dependent on the frequency converter frequency.
The control algorithms (e.g. GPC and pole placement) have been developed
using Real-Time Workshop of the MATLAB/SIMULINK program package.
The algorithm output represents the reference frequency value, which
is transmitted to the frequency converter via a D/A converter.
[Functional scheme of control system with MATLAB]
Boost converter has function of higher harmonic active filter in
ac or dc one phase supply network. Fundamental concept of boost
converter control is based on input-current and output-voltage cascade
control. Conventional and modern control algorithms of boost converter
have been investigated and they ensure low level of higher harmonics
in supply network. Moreover, input power factor greater than 0.95
when converter operates on ac network as well as high input impedance
when converter operates on dc network is achieved. Applied algorithm's
validation is checked by simulation on mathematical model with instantaneous
values and partly on 2.5 kW physical laboratory model.
APC turret [Picture]
For the purpose of developing and investigating two-axis simultaneous
speed and position control algorithms, as well as friction and backlash
effects, an APC turret has been adapted. Previous control solution,
that utilizes analog speed controllers, DC drives and analog speed
sensors has been abandoned and replaced with more contemporary one
that utilizes PC computer with appropriate PC cards, brushless servo
drives (servo motors and converters) and digital displacement sensors
realized by incremental encoders and resolvers, thus enabling position
control and better tracking.
The control system is realized with maximum fidelity to the real
operating conditions in APCs (speed control with speed reference
obtained from adapted turret joystick), with additional control
options: position control and computer generated position or speed
reference. Since this model is intended also for educational purposes,
a user friendly interface is beeing developed with possibility of
generating various types of references.