Controller Designing for Quanser Qube Servo 3

Muhammad Ashir Muhammad Dawood Bilal

Faculty of Electrical Engineering Faculty of Electrical Engineering
GIK Institute of Engineering Sciences and Technology GIK Institute of Engineering Sciences and Technology
KPK, Pakistan KPK, Pakistan
Email: u2022353@[Link] Email: u2022365@[Link]

Hassaan Ahmed Rana Muhammad Osairum

Faculty of Electrical Engineering Faculty of Electrical Engineering
GIK Institute of Engineering Sciences and Technology GIK Institute of Engineering Sciences and Technology
KPK, Pakistan KPK, Pakistan
Email: u2022208@[Link] Email: u2022412@[Link]

Abstract—This report focuses on the modeling, simulation, B. Integral Controller

and hardware implementation of classical control strategies—PI,
PD, and PID—for the Quanser Qube Servo 3. Using MAT- The integral controller uses the accumulated error over time
LAB/Simulink, different controllers were developed and applied to eliminate steady-state error. The control action is defined
to the Qube’s position control problem under step and ramp as:
inputs. The performance of each controller was assessed based
on steady-state error and system response. Experimental testing Z t
was conducted on the hardware to validate the simulated results u(t) = Ki e(τ )dτ (2)
and determine the most effective control strategy. 0

where Ki is the integral gain. This controller corrects

I. I NTRODUCTION persistent errors but may lead to overshoot if used excessively.
In Linear Control Systems, Proportional (P), Integral (I), and C. Proportional-Integral (PI) Controller
Derivative (D) controllers are widely used to regulate dynamic
systems. The proportional controller generates control output The PI controller combines proportional and integral terms.
based on the current error between the desired and actual It provides both fast response and elimination of steady-state
output. It ensures a fast response but may result in a steady- error:
state error. Z t
The integral controller accumulates past errors over time u(t) = Kp e(t) + Ki e(τ )dτ (3)
and generates a control action that eliminates the steady-state 0
offset. However, excessive use of the integral term can lead to It adjusts the output based on both the current error and the
a sluggish response or overshoot. past accumulated error, improving overall system accuracy.
The derivative controller predicts the future trend of the
error and introduces damping to reduce overshoot and improve
stability. Combining these techniques, such as in PI, PD,
and PID configurations, yields better performance in terms
of speed, accuracy, and robustness.

II. C ONTROLLER D ESIGN T ECHNIQUES

A. Proportional Controller
Fig. 1. PI Controller Block Diagram
A proportional controller produces an output proportional
to the current error signal. The control law is defined as:
D. Proportional-Derivative (PD) Controller
The PD controller enhances transient response by adding
u(t) = Kp e(t) (1)
a derivative component that anticipates future error changes.
The control law is:
Here, Kp is the proportional gain, and e(t) is the instan-
taneous error. This controller is simple and ensures quick de(t)
response, but it cannot eliminate steady-state error. u(t) = Kp e(t) + Kd (4)
dt
 Here, Kd is the derivative gain. This controller improves  
damping and stability, making it suitable for systems requiring Ki 22.4
GP I
openloop (s) = Kp + · (9)
fast response. However, it is sensitive to high-frequency noise. s s(0.15s + 1)
The controller was simulated in MATLAB/Simulink for a
unit step input. The resulting steady-state error was:

ess = −0.0022 (10)

The PI controller effectively reduced the steady-state error

while maintaining a moderate response speed.

Fig. 2. PD Controller Block Diagram B. PD Controller Design

The PD controller improves transient performance and
E. Proportional-Integral-Derivative (PID) Controller damping. It is defined as:
The PID controller is a comprehensive control strategy
combining P, I, and D actions. It is defined by: CP D (s) = Kp + Kd s (11)
Z t
de(t) The tuned values were:
u(t) = Kp e(t) + Ki e(τ )dτ + Kd (5)
0 dt
It offers rapid response, zero steady-state error, and im- Kp = 0.09311, Kd = 0.0456 (12)
proved damping. PID controllers are widely used due to their
flexibility and robustness across different systems. The open-loop transfer function becomes:

22.4
GP D
openloop (s) = (Kp + Kd s) · (13)
s(0.15s + 1)
The system response to a step input showed a final steady-
state error of:

ess = −0.0016 (14)

The PD controller provided improved damping and a faster

Fig. 3. PID Controller Block Diagram
response compared to PI.
III. C ONTROLLER D ESIGN
C. PID Controller Design
The plant model of the Quanser Qube Servo 3 system is
represented by the transfer function: The PID controller combines all three control actions. Its
form is:
K 22.4
P (s) = = (6) Ki
s(τ s + 1) s(0.15s + 1) CP ID (s) = Kp + + Kd s (15)
s
This is a second-order system with an integrator and a lag
element. PI, PD, and PID controllers were designed using The tuned gain values were:
MATLAB’s codes values to improve system accuracy and
response. Kp = 0.6277, Ki = 1.064, Kd = 0.0819 (16)
A. PI Controller Design The open-loop transfer function becomes:
The PI controller is expressed as:
 
Ki Ki 22.4
CP I (s) = Kp + (7) GP ID
openloop (s) = Kp + + Kd s · (17)
s s s(0.15s + 1)
Using MATLAB’s pidtune function, the following gains The PID controller showed the best performance in terms of
were obtained: accuracy and stability. The steady-state error for a step input
was:
Kp = 0.0990, Ki = 0.0457 (8)
The open-loop transfer function becomes: ess = −2.8231 × 10−4 (18)
D. Comparison of Controller Performance B. Step and Ramp Responses
The results indicate that the PID controller significantly The performance of PI, PD, and PID controllers was eval-
reduced the steady-state error compared to PI and PD, making uated using step and ramp input signals. The figures below
it the most effective for this application. show the system responses for each controller.

TABLE I
T UNED G AINS AND S TEADY-S TATE E RRORS

Controller Kp Ki Kd ess
PI 0.0990 0.0457 0 −0.0022
PD 0.0931 0 0.0456 −0.0016
PID 0.6277 1.064 0.0819 −2.8231 × 10−4

E. Steady-State Error vs. System Type

The steady-state error of a system varies based on its type
and the type of input signal applied. The following table shows Fig. 5. PI Controller Response to Step Input
the expected error behavior for standard inputs:

TABLE II
S TEADY-S TATE E RROR FOR D IFFERENT S YSTEM T YPES

System Type Step Input Ramp Input Parabolic Input

1
Type 0 ∞ ∞
1 + Kp
1
Type 1 0 ∞
Kv
1
Type 2 0 0
Ka

IV. R ESULTS
Fig. 6. PI Controller Response to Ramp Input
The PI, PD, and PID controllers were implemented and
tested on the Quanser Qube Servo 3 using MATLAB/Simulink.
The effectiveness of each controller was evaluated based on
the steady-state error in response to a unit step input. Both
simulation and hardware validations were carried out.

A. PI Controller Results
The PI controller was able to significantly reduce the steady-
state error. The measured steady-state error was:

ess = −0.0022 (19)

Fig. 7. PD Controller Response to Ramp Input

Fig. 4. Simulink Model of PI Controller for Quanser Qube Servo 3

 The Simulink implementation of the PID controller is shown V. C ONCLUSION
below: This report presented the design, simulation, and hardware
implementation of classical control strategies—PI, PD, and
PID—for the Quanser Qube Servo 3. Each controller was mod-
eled and tuned using MATLAB/Simulink to achieve improved
position control under step and ramp inputs.
The PI controller effectively reduced the steady-state error
but exhibited moderate dynamic performance. The PD con-
troller improved transient response but could not fully elim-
inate steady-state error. The PID controller, which combines
the benefits of both PI and PD, demonstrated the best overall
performance in terms of accuracy, stability, and steady-state
response.
Experimental results validated the simulation models, show-
ing that the PID controller produced the lowest steady-state
error and delivered superior control over the Qube system.
This highlights the practical effectiveness of PID control in
real-time applications, particularly for systems with lag and
integration dynamics like the Quanser Qube Servo 3.
ACKNOWLEDGMENT
Fig. 8. Simulink Circuit for PID Controller on Quanser Qube Servo 3
We would like to express our sincere gratitude to our course
instructor, Dr. Dur e Zehra, for her invaluable guidance, sup-
port, and encouragement throughout the course. Her insights
greatly contributed to our understanding of control system
design and its practical applications.
We also extend our heartfelt thanks to Lab Engineer Sir
Mazhar Hussain for his continuous assistance during the hard-
ware implementation phase. His expertise with the Quanser
Qube Servo 3 hardware was instrumental in successfully val-
idating our simulation results through real-time experiments.
R EFERENCES
[1] GIK Institute, “Linear Control Systems Lab Manual,” Faculty of Elec-
trical Engineering, 2025.
[2] N. S. Nise, Control Systems Engineering, 7th ed., Wiley, 2015.

Fig. 9. PID Controller Response to Step Input

Fig. 10. PID Controller Response to Ramp Input