International Journal of Emerging Engineering Research and Technology

Volume 6, Issue 11, 2018, PP 24-31

ISSN 2349-4395 (Print) & ISSN 2349-4409 (Online)

Modeling and Control of 2-DOF Robot Arm

Nasr M. Ghaleb1 and Ayman A. Aly1, 2
1
Mechanical Engineering Department, College of Engineering, Taif University, PO Box 888, Taif,
Saudi Arabia
2
Mechanical Engineering Department, Faculty of Engineering, Assiut University, PO Box 71516,
Assiut, Egypt
*Corresponding Author: Nasr M. Ghaleb, Mechanical Engineering Department, College of
Engineering, Taif University, PO Box 888, Taif, Saudi Arabia

ABSTRACT
This paper presents a Modeling, Simulation and Control of a Two Degree of Freedom (2-DOF) robot
arm.This Work is taken from the Final Year capstone project. First The Robot specifications , Robot
Kinematics with Denavit-Hartenberg parameters (DH)for Forward kinematics and Inverse Kinematicsof 2-
DOF robot armwere presented. Then The dynamics of the 2-DOF robot arm was studied to derive the
equations of motion based on Eular-Lagrange Equation of motion. A Control Design was performed using
PID controller for the modeling and control Technique. The models has been done based on
Matlab/Simulink software.
Keywords: Robotics, 2-DOF Robot arm, Kinematic, Dynamic, PID Control and Modeling.

INTRODUCTION most of industrial processes. Most robot

manipulators found in industrial operations are
Robotics is defined practically as the study, controlled by PID algorithms independently at
design and use of robot systems for
each joint.
manufacturing and generally are used to
perform highly repetitive, unsafe, hazardous, There are many control techniques used for
and unpleasant tasks. Robotics many different controlling the robot arm.
functions that used either in industry and The most familiar control techniques are the
manufacture or in complex, clatter and changing PID control, adaptive control, optimal control
environment such as pick and place, assembly, and robust control. As the final goal is to design
drilling, welding, machine tool load and unload and manufacturing real robots, it's helpful doing
functions, painting, spraying, etc. or in A the simulation before the investigations with
delivery in a hospital and Hotels, Discovering real robots, to enhance the final real robot
the space As a results of these different tasks performance and behavior.
there are different robot arm configuration such
as rectangular, spherical, cylindrical, revolute ROBOT SPECIFICATION AND KINEMATICS
and prismatic jointed. Robot Specification
A pick and place robot arm is used to ease A two degree of freedom robot arm is described
process of moving materials and supplying the in Figure(1)
motion required in the manufacturing processes.
The transfer process of the materials is usually which consists primarily of two links with the
being accomplished, using man power and as following specifications in OXY coordinates:
the transfer process is repeated for a period of 𝐿1 = 1 m is the length of the first link.
time, it can cause injuries to the operator. the 𝐿2 = 1 m is the length of the second link.
robot arm preventing injuries and increasing the
efficiency of the work, with reducing the human 𝑚1 = 1 kg is the mass of the first link.
being errors that cost highly time and martial. 𝑚2 = 1 kg is the link of the second link.
The proportional-integral-derivative (PID) 𝜃1 = the rotation angel of the first link.
control has simple structure for its three gains.
𝜃2 = is the rotation angel of the second link.
The control performances are acceptable in the

International Journal of Emerging Engineering Research and Technology V6 ● I11 ● 2018 24

Modeling and Control of 2-DOF Robot Arm

Robot Kinematics
Forward Kinematics
The Forward kinematics of a robotic arm is
determined a group of parameters called
Denavit-Hartenberg (DH) parameters which
used for deriving the homogenous
transformation matrices between the different
frames assigned on the robot arm structure. The
DH parameters for a two degree of freedom
robotic arm are defined as follows:
Figure 1. Two degree of freedom Robot Arm
Table 1. DH-parameters for the 2-DOF robotic arm
Link ai αi di 𝜽𝒊
1 L1 0 0 𝜃1
2 L2 0 0 𝜃2
The homogenous transformation matrices for the 2-DOF robotic arm shown in Figure(1) are derived
as follows:
cos 𝜃1 − sin 𝜃1 0 𝐿1 cos 𝜃1
𝟎 sin 𝜃1 cos 𝜃1 0 𝐿1 sin 𝜃1
𝟏𝑻 = (1)
0 0 1 0
0 0 0 1
cos 𝜃2 − sin 𝜃2 0 𝐿2 cos 𝜃2
𝟏 sin 𝜃2 cos 𝜃2 0 𝐿2 sin 𝜃2 (2)
𝟐𝑻 =
0 0 1 0
0 0 0 1
Using the Eq. (1) and (2), the homogenous transformation matrix 02𝑇can be derived as follows:
cos(𝜃1 + 𝜃2 ) − sin(𝜃1 + 𝜃2 ) 0 𝐿1 cos 𝜃1 + 𝐿2 cos(𝜃1 + 𝜃2 )
𝟎 sin(𝜃1 + 𝜃2 ) cos(𝜃1 + 𝜃2 ) 0 𝐿1 sin 𝜃1 + 𝐿2 sin(𝜃1 + 𝜃2 )
𝟐𝑻 = ( 3)
0 0 1 0
0 0 0 1
Therefore, can be derived either algebraically or
𝑛𝑥 𝑜𝑥 𝑎𝑥 𝑝𝑥 geometrically. The geometrical approach is
𝑹
𝑛𝑦 𝑜𝑦𝑎𝑦 𝑝𝑦 considered to be much easier for robot arms of
𝑯𝑻 = 𝑛𝑧 𝑜𝑧 𝑎𝑧 𝑝𝑧 (4) high degrees of freedom. In our Case, we solved
0 0 0 1 the inverse kinematics equations for the 2-DOF
𝟎 𝑹
robotic arm shown in Figure(2) using the
𝟐𝑻 =𝑯 𝑻 (5) geometrical method.
From Eq. (5), the position coordinates of the
manipulator end-effector is given by:
𝑃𝑥 = 𝐿1 cos 𝜃1 + 𝐿2 cos(𝜃1 + 𝜃2 ) (6)
𝑃𝑦 = 𝐿1 sin 𝜃1 + 𝐿2 sin(𝜃1 + 𝜃2 ) (7)
And the end-effector's orientation matrix is
defined by the first three rows and three
columns of the transformation matrix in Eq.(3).
Inverse Kinematics
The inverse kinematics of a robotic arm is a
solution of finding the robot arm joint variables
of given the position Cartesian coordinates of Figure 2. Two degree of freedom Robot Arm Inverse
the end-effector. The mathematical equations Kinematic
used to solve the inverse kinematics problem From Figure2, a mathematical equation for

25 International Journal of Emerging Engineering Research and Technology V6 ● I11 ● 2018

Modeling and Control of 2-DOF Robot Arm

solving the elbow joint angle 𝜽2 can be derived arm: Eular-Lagrange method and Newten-Eular
using Pythagoras theorem as follows: method. in this work , We used the Eular-
Lagrange method which depends on calculating
𝑝𝑥 2 + 𝑝𝑦 2 = 𝐿1 2 + 𝐿2 2 + 2𝐿1 𝐿2 cosθ2 (8) the total Kineatic and Potential Energies of the
1 robot arm to determine the Lagrangian (ℒ) of
cosθ2 = (𝑝𝑥 2 + 𝑝𝑦 2 − 𝐿1 2 − 𝐿2 2 ) (9)
2𝐿1 𝐿2 the whole system in order to calculate the force
or torque applied of each joint.With the
sinθ2 = ± 1 − cosθ2 2 (10) Lagrangianℒ, we can solve the Euler- Lagrange
equation which relies on the partial derivative of
Therefore, kinetic and potential energy properties of
sin θ2 mechanical systems to compute the equations of
𝜃2 = ± 𝑎𝑡𝑎𝑛 (11)
cos θ2 motion and is defined as follow:
For The joint variable 𝜃1 : 𝑑 𝜕ℒ 𝜕ℒ
F= − (21)
𝑑𝑡 𝜕 𝜃 𝜕𝜃
𝑝𝑥 = (𝐿1 + 𝐿2 cosθ2 ) cosθ1 − 𝐿2 sinθ1 sinθ2 (12)
where F is the external force acting on the
𝑝𝑦 = 𝐿2 sinθ2 cosθ1 + (𝐿1 + 𝐿2 cosθ2 )sinθ1 (13)
generalized coordinate, represents the torque
𝐿1 + 𝐿2 cosθ2 −𝐿2 sinθ2 applied to therobot and ℒ is the Lagrangian
∆= (14)
𝐿2 sinθ2 𝐿1 + 𝐿2 cosθ2 equation of the motion, given in Eq.(22)
𝑝𝑥 2 + 𝑝𝑦 2 = (𝐿1 + 𝐿2 cosθ2 )2 + (𝐿2 sinθ2 )2 (15) ℒ 𝑞 𝑡 ,𝑞 𝑡 = 𝐾𝐸 𝑞 𝑡 , 𝑞 𝑡 − 𝑃𝐸 𝑞 𝑡 (22)
𝐿1 + 𝐿2 cosθ2 𝑝𝑥 To solve the Lagrangian Eq.(22),we need first to
∆sinθ1 = (16)
𝐿2 sinθ2 𝑝𝑦 calculate the kinetic Energy 𝐾𝐸 and potential
Energy 𝑃𝐸 as follow:
𝑝𝑥 −𝐿2 sinθ2
∆cosθ1 = (17) 1
𝑝𝑦 𝐿1 + 𝐿2 cosθ2 𝐾𝐸 = 𝑚𝑥 2
2
∆sin θ1 𝐿1 +𝐿2 cos θ2 𝑝 𝑦 −𝐿2 sin θ2 𝑝 𝑥
sinθ1 = = (18) 𝑃𝐸 = 𝑚𝑔𝑙
∆ 𝑃𝑥 2 +𝑝 𝑦 2

∆cos θ1 𝐿1 +𝐿2 cos θ2 𝑝 𝑥 +𝐿2 sin θ2 𝑝 𝑦 The velocity is determined by taking the
cosθ1 = = (19) derivative of the position respect to time. so the
∆ 𝑝 𝑥 2 +𝑝 𝑦 2
position in the end of the link is known using
sin θ1 𝐿1 +𝐿2 cos θ2 𝑝 𝑦 ±𝐿2 sin θ2 𝑝 𝑥
𝜃1 =𝑎𝑡𝑎𝑛 = 𝑎𝑡𝑎𝑛 (20) known familiar variables :
cos θ1 𝐿1 +𝐿2 cos θ2 𝑝 𝑥 ±𝐿2 sin θ2 𝑝 𝑦
𝑥1 = 𝐿1 𝑠𝑖𝑛𝜃1
ROBOT DYNAMICS
𝑦1 = 𝐿1 cos 𝜃1
The dynamic model of a robot is concerned with
the movement and the forces involved in the 𝑥2 = 𝐿1 𝑠𝑖𝑛𝜃1 + 𝐿2 sin⁡
(𝜃1 + 𝜃2 ) (23)
robot arm and establishes a mathematical
𝑦2 = 𝐿1 cos 𝜃1 + 𝐿2 cos(𝜃1 + 𝜃2 )
relationship between the location of the robot
joint variables and the dimensional parameters Then substitute in the kinetic's energy equation:
of the robot. There are two methods for 1 1 1 1
𝐾𝐸 = 𝑚1 𝑥1 2 + 𝑚1 𝑦1 2 + 𝑚2 𝑥2 2 + 𝑚2 𝑦2 2 (24)
performing the dynamics equations of a robot 2 2 2 2

After simplification,
1 2 1 2 1 2 2
𝐾𝐸 = (𝑚1 + 𝑚2 )𝑙1 2 𝜃1 + 𝑚2 𝑙1 2 𝜃1 + 𝑚2 𝑙2 2 𝜃1 𝜃2 + 𝑚2 𝑙2 2 𝜃2 + 𝑚2 𝑙1 𝑙2 𝑐𝑜𝑠𝜃2 𝜃1 𝜃2 + 𝜃1 (25)
2 2 2

The Potential energy equation is defined as:

𝑃𝐸 = 𝑚1 𝑔𝑙1 cos 𝜃1 + 𝑚2 𝑔(𝑙1 cos 𝜃1 + 𝑙2 𝑐𝑜𝑠(𝜃1 + 𝜃2 )) (26)
Substitute Eq.(25) and(26) in Eq.(21) to form the lagrangian equation as:
1 2 1 2 1 2 2
ℒ = (𝑚1 + 𝑚2 )𝑙1 2 𝜃1 + 𝑚2 𝑙1 2 𝜃1 + 𝑚2 𝑙2 2 𝜃1 𝜃2 + 𝑚2 𝑙2 2 𝜃2 + 𝑚2 𝑙1 𝑙2 𝑐𝑜𝑠𝜃2 𝜃1 𝜃2 + 𝜃1 −
2 2 2
𝑚1 𝑔𝑙1 cos 𝜃1 + 𝑚2 𝑔 𝑙1 cos 𝜃1 + 𝑙2 𝑐𝑜𝑠 𝜃1 + 𝜃2 (27)
To calculate the force applied to the robot, we form the Lagrange-EularEq.(21)with the Lagrangian(ℒ)
𝑑 𝜕ℒ 𝜕ℒ
Fθ1,2 = −
𝑑𝑡 𝜕 𝜃 𝜕𝜃1,2
1,2

International Journal of Emerging Engineering Research and Technology V6 ● I11 ● 2018 26

Modeling and Control of 2-DOF Robot Arm

After simplification, the force applied at joint 1is given by

𝐹𝜃1 =
𝑚1 + 𝑚2 𝑙1 2 + 𝑚2 𝑙2 2 + 2𝑚2 𝑙1 𝑙2 cos 𝜃2 ) θ1 + 𝑚2 𝑙2 2 − 𝑚2 𝑙1 𝑙2 cos 𝜃2 θ2 −
2
𝑚2 𝑙1 𝑙2 sin 𝜃2 ( 2𝜃1 𝜃2 + 𝜃2 ) − 𝑚1 + 𝑚2 l1 gsinθ1 − 𝑚2 𝑙2 𝑔𝑠𝑖𝑛(𝜃1 + 𝜃2 ) (28)
and the force applied at joint 2 is given by
𝐹𝜃2 = (𝑚2 𝑙2 2 + 𝑚2 𝑙1 𝑙2 cos 𝜃2 )θ1 + 𝑚2 𝑙2 2 θ2 − 𝑚2 𝑙1 𝑙2 sin 𝜃2 𝜃1 𝜃2 − 𝑚2 𝑙2 𝑔𝑠𝑖𝑛(𝜃1 + 𝜃2 ) (29)
The motion of the systemis given by the following formof a nonlinear equation:
𝐹 = 𝐵 𝑞 + 𝐶 𝑞, 𝑞 + 𝑔 𝑞 (30)
where,
F θ1
𝐹=
F θ2
( 𝑚1 + 𝑚2 𝑙1 2 + 𝑚2 𝑙2 2 + 2𝑚2 𝑙1 𝑙2 cos 𝜃2 ) (𝑚2 𝑙2 2 − 𝑚2 𝑙1 𝑙2 cos 𝜃2 )
𝐵 𝑞 =
𝑚2 𝑙2 2 + 𝑚2 𝑙1 𝑙2 cos 𝜃2 𝑚2 𝑙2
2
− 𝑚2 𝑙1 𝑙2 sin 𝜃2 ( 2𝜃1 𝜃2 + 𝜃2 )
𝐶 𝑞, 𝑞 =
− 𝑚2 𝑙1 𝑙2 sin 𝜃2 𝜃1 𝜃2
− 𝑚1 + 𝑚2 l1 gsinθ1 − 𝑚2 𝑙2 𝑔𝑠𝑖𝑛(𝜃1 + 𝜃2 ) )
𝑔 𝑞 =
−𝑚2 𝑙2 𝑔𝑠𝑖𝑛(𝜃1 + 𝜃2 )
θ1
𝑞=
θ2
1
MATHEMATICAL MODELING OF ACTUATING 𝜔 𝑠 = 𝐾𝑡 ∗ 𝐼𝑎 𝑠 ∗ (32)
(𝐽 𝑚 𝑠+𝑏𝑚 )
SYSTEM
The PMDC motor Transfer Function is derived,
The Permanent Magnet Direct Current PMDC Subtituting the electrical part characteristics
motor is used to actaute the system, which has Eq.(44) in Eq.(45), gives Eq.(46):
an Electrical part and a Mechanical part, as seen
1
in Eq.(44) and Eq.(45), Describing the Electrical 𝐾𝑡
𝐿𝑎 𝑠+𝑅𝑎
[𝑉𝑖𝑛 (𝑠) − 𝐾𝑏 𝜔𝑚 ] = 𝐽𝑚 𝑠𝜔 + 𝑏𝑚 𝜔 (33)
and Mechanical characteristics of PMDC motor,
respectively, Using Newton's law, Kirchhoff’s Rearranging Eq.(46), The PMDC motor open
law and Ohm’s law. loop transfer function is obtained, given by
1
Eq.(47), This equation shows that the PMDC
[𝑉𝑖𝑛 (𝑠) − 𝐾𝑏 𝜔𝑚 ] ∗ = 𝐼𝑎 𝑠 (31) motor is a second order system.
𝐿𝑎 𝑠+𝑅𝑎

 (s ) Kt
G speed (s )   (34)
V in (s ) (La J m )s  (R a J m  bm La )s  (R abm  K t K b ) 
2

PIDCONTROLLER DESIGN where𝑒 = qd − q , qd is desired joint angle, Kp,

The nonlinear equation that are derived from Ki and Kd are proportional, integral and
Euler-Lagrange Eq. (30), where inputvariable F derivative gains of the PID controller,
which represents the torque applied to the robot respectively. This PID control law can be
is unknown, so it's requires a control in the expressed via the following equations.
Force applied of the joints to reach the final 𝐹 = 𝐾𝑃 𝑒 + 𝐾𝐷 𝑒 + 𝜉 (36)
position.In our case, we use the classical linear
PID, Particularly we need Two PID controls 𝜉 = 𝐾𝐼 𝑒 ; 𝜉 0 = 𝜉0
since the first arm motion is dependent from the where 𝜉 represents the additional state variable
second arm motion. In fact, still having a strong that is the integral action of the PID control law
interaction between the two arms. The classical formed and it's time derivative is 𝜉 = 𝐾𝐼 𝑒. The
linear PID law is performed; closed-loop equation is obtained by substituting
𝐹 = 𝐾𝑃 𝑒 + 𝐾𝐷 𝑒 + 𝐾𝐼 𝑒 𝑑𝑡 (35) the control action F from Eq.(32) into the robot
model Eq.(30), gives:

27 International Journal of Emerging Engineering Research and Technology V6 ● I11 ● 2018

Modeling and Control of 2-DOF Robot Arm

𝐵 𝑞 + 𝐶 𝑞 , 𝑞 + 𝑔 𝑞 = 𝐾𝑃 𝑒 + 𝐾𝐷 𝑒 + 𝜉 (37) −π 2
θo = π
From the system'smodel Eq.(30), we can have 2
𝑞 = 𝐵(𝑞)−1 −𝐶 𝑞 , 𝑞 − 𝑔 𝑞 +𝐹 (38) so, in our case:
with 𝑓1 = 𝐾𝑃1 θ1𝑓 − θ1 + 𝐾𝐷1 𝜃1 + 𝐾𝐼1 𝑒(𝜃1 ) 𝑑𝑡
−1 𝑓2 = 𝐾𝑃2 θ2𝑓 − θ2 + 𝐾𝐷2 𝜃2 + 𝐾𝐼2 𝑒(𝜃2 ) 𝑑𝑡
𝐹=𝐵 𝑞 𝐹 ⇔ 𝐹=𝐵 𝑞 𝐹 (39)
so, we decoupled the system to have the new However, the complete system equations with
(non-physical)input control would be

𝑓 𝑞 = 𝐵(𝑞)−1 −𝐶 𝑞 , 𝑞 − 𝑔 𝑞 +𝐹 (43)
𝐹= 1 (40)
𝑓2 with
however, the physical torque inputs to the 𝑓1 𝐾𝑃1 θ1𝑓 − θ1 + 𝐾𝐷1 𝜃1 + 𝐾𝐼1 𝑒 𝜃1 𝑑𝑡
𝐹= = (44)
system are 𝑓2 𝐾𝑃2 θ2𝑓 − θ2 + 𝐾𝐷2 𝜃2 + 𝐾𝐼2 𝑒 𝜃2 𝑑𝑡
𝑓θ1 𝑓 recalling the physical actual torques
=𝐵 𝑞 1 (41)
𝑓θ2 𝑓2
𝐹θ1 𝑓
=𝐵 𝑞 1 (45)
The error signals of the system are 𝐹θ2 𝑓2
e θ1 = θ1𝑓 − θ1 RecallingEq.(32) and Eq.(38), gives:
e θ2 = θ2𝑓 − θ2 (42)
𝜉1 = 𝐾𝐼1 𝑒 𝜃1 𝑑𝑡 ⇔ 𝜉1 = 𝐾𝐼1 𝑒1
where θ𝑓 is the final positions. The final
position is given by 𝜉2 = 𝐾𝐼2 𝑒 𝜃2 𝑑𝑡 ⇔ 𝜉2 = 𝐾𝐼2 𝑒2 (46)
θ1𝑓 π So, the system equations are
= −π2
θ2𝑓 𝜉1 = 𝐾𝐼1 𝑒1
2
and the initial position is given by 𝜉2 = 𝐾𝐼2 𝑒2
θ1 𝐾𝑃1 θ1𝑓 − θ1 + 𝐾𝐷1 𝜃1 + 𝜉1
= 𝐵(𝑞)−1 −𝐶 𝑞 , 𝑞 − 𝑔 𝑞 + (47)
θ2 𝐾𝑃2 θ2𝑓 − θ2 + 𝐾𝐷2 𝜃2 + 𝜉2
SIMULATION AND RESULTS
Approximate mathematical models can be
obtained and then simulated in combination
with the designed control law, for providing a
more realistic validation of the system behavior
and control performance. First we made a
mathematical model of the PMDC motor to
determine the transfer function and then start the
simulation of the actuator system without
controller using MATLAB /Simulink as shown
in Figure (3)to study the system performance Figure3. Step Response of PMDC without controller.
and then improving the design by adding PID
Controller as shown in Figure(4)and(5) for a
more realistic validation of the system
performance.
The system reaches steady state value of 0.778
rad/s in 1.27 s, with small overshoot and the
system response is very fast as shown in
Figure(3), so we need to add controller to
improve the system behavior. PID controller
simulinkis built to control the angular speed at a
desired set point value (𝐾𝑎,𝑡 = 1)and improving
the system performance as illustrate in Figure4. Figure 4. Simulink model of the actuator
system(PMDC).

International Journal of Emerging Engineering Research and Technology V6 ● I11 ● 2018 28

Modeling and Control of 2-DOF Robot Arm

We have noticed from Figure(5) and (6) that the After tuning by trial and error we got with the
system is second order system (Under damped) PID controller values :kp=40, ki=13,kd=5for the
and the values of the controller are getting with best performance, Also we have noticed from
PID tuning by trial and error with system Figure(8) that the position error reaches at a
characteristicsvalues for the best performance considerable time to zero steady-state of
and behavior:𝐾𝑝 =13, 𝐾𝑖 =39, 𝐾𝑑 =1.056, OS (%) constant input.
< 5, settling time < 2 sec, steady-state error <1%
and 𝐾𝐷𝐶 = 1.

Figure8. The position error.

Figure 5. System characteristics(With PID controller The force or torque applied can be shown in
tuning). Figure(9) is zero at steady state.

Figure6. Step response of the system with PID

controller.
Then we have built and studied the steady state
dynamic model with PID control of single link
by MATLAB simulation as illustrate in Figure 9. The torque error.
Figure(7). Then we have studied the two degree of
freedom with PID control as a whole system by
trial and error,
The PID parameters are manually tuned to get
the best performance of the system. The best
behavior and performance of the controller
parameters values are as follows :
𝐾𝑝1 = 250 𝐾𝑝2 = 250
𝐾𝑖1 = 200 𝐾𝑖2 = 200
𝐾𝑑1 = 30 𝐾𝑑2 = 30
By our approach trial and error, we have noticed
that 𝐾𝑝1 is related to direct error and to speed of
Figure7. PID Controller of One Single Link. evolution, 𝐾𝑝1 is related to speed of interaction

29 International Journal of Emerging Engineering Research and Technology V6 ● I11 ● 2018

Modeling and Control of 2-DOF Robot Arm

with change in states and 𝐾𝑝1 is related to By the experiment, we got that the slightest
overall error cancellation. change in the controller parameters yields more
overshoots and oscillations due to highly
We have seen in the Figures (10) and (11) that sensitive to initial and final positions.
the position error reaches zero at a reasonably
fast time.

Figure13. Torque of Theta 1.

Figure10. Theta 1 error. CONCLUSION

The main topic of this paper is modeling,
simulation and control of two degree of freedom
robot arm.
The mathematical models of the actuator and
whole system are illustrated. Lagrangian and
Euler-Lagrange used to derive a dynamic model
that simulated the actual robot movement in real
life and obtain sufficient control over the robot
joint positions. The system was controlled in
order to reach a desired joint angle position
through simulation of PID controllers using
MATLAB/Simulink.
Also, the result showed that a slight changes in
initial joint angle positions of the robot arm
Figure11. Theta 2 error. resulted in different desired joint angle positions
and this necessitated that the gains of the PID
controllers need to be adjusted and turned at
every instant in order to prevent overshoot and
oscillation that associated with the change in
parameters values.
REFERENCES
[1] S. H. M. V. Mark W. Spong, Robot Modeling
and Control, WILEY, December 2005.
[2] N.Jazar, R. (2010). Theory of Applied
Robotics.
[3] Okubanjo, A. A., Modeling of 2-DOF Robot
Arm and Control. (FUTOJNLS, Volume-3 ,
December 2017 .
[4] David I. Robles G., PID control dynamics of a
Figure12. Torque of Theta 1. robotic arm manipulator with two degrees of
freedom, August 17, 2012.
And from Figures (12) and (13), we have seen [5] KhalilIbrahim, Ayman A. Aly and Ahmed Abo.
the force or torque applied slight overshot and Ismail ,”Position Control of A Single Flexible
quickly stabilize. Arm”, International Journal of Control,

International Journal of Emerging Engineering Research and Technology V6 ● I11 ● 2018 30

Modeling and Control of 2-DOF Robot Arm

Automation and Systems, (IJCAS), Vol.2, Simulation and Control of 2-R Robot. Global
No.1, pp.12-18, 2013 Journal of Researches in Engineering, Volume
[6] Aalim M. Mustafa & A AL-SAIF, Modeling, 14, 2014.

Citation: Nasr M. Ghaleb and Ayman A. Aly, "Modeling and Control of 2-DOF Robot Arm", International
Journal of Emerging Engineering Research and Technology, 6(10), pp.24-31
Copyright: © 2018 Nasr M. Ghaleb. This is an open-access article distributed under the terms of the
Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in
any medium, provided the original author and source are credited.

31 International Journal of Emerging Engineering Research and Technology V6 ● I11 ● 2018