Implementation of Pick & Place Robotic Arm for

Warehouse Products Management

2021 IEEE 7th International Conference on Smart Instrumentation, Measurement and Applications (ICSIMA) | 978-1-7281-7523-2/21/$31.00 ©2021 IEEE | DOI: 10.1109/ICSIMA50015.2021.9526304

Nazib Sobhan Abu Salman Shaikat

Department of Mechatronics Department of Mechatronics
Asian Institute of Technology Engineering
Pathumthani, Thailand World University of Bangladesh
[email protected] Dhaka, Bangladesh
[email protected]
l

Abstract—Automation is the future of the industrial shipbuilding, food manufacturing, and chemical industries,
manufacturing process for the upcoming fourth industrial etc.
revolution. This automation implicates intelligent machines,
robots, the Internet of Things (IoT), and the Artificial Robotic pick and place are an automatic process,
Intelligence (AI). The pick and place robotic manipulator has where picking up and placing is completed in a very
often utilized robot systems for industry. The application of the minimal time with precision, which in turn speeds up the
fourth industrial revolution is started with pick and place production rates. In this generation, various kinds of
robotic manipulator. This paper addresses the design and end-of-arm tooling options are available. However, pick
implementation of pick & place robotic hand manipulator and place robots are the ones, which can be modified to
operation for automatic warehouse product management. fit for its exact production requirements. In factory lines,
Mainly the big warehouse management system is quite complex these kinds of robots can move all sorts of products like
and difficult for manual control and that’s why an automatic small, large, heavy, or hard-to-handle manufacturing
system is introduced to ease the process. The proposed products. Pick and place can be different types such as
automated scheme is successfully cover with 5 DOF robotic it can be a spherical robot which can provide one linear and
manipulator and various mathematical calculations is done to two rotational movements, and the cylindrical robot that
run the model. Finally, a prototype has been built to validate provides movement in vertical, horizontal, and rotational axes,
the model, and eventually, a GSM-based mobile application is
a SCARA robot (fixed robots with 3 vertical axes rotary arms)
introduced to regulate the operation. This work can not only
or an articulated robot. These types of robots are quicker and
bring the precession but also provide the guarantee of the
warehouse system is free from artificial waste and ensuring
works can be done in less time. Moreover, the safety of the
perfect warehouse products management system. working area is increasing and these robots can do a lot more
tiresome works. In this work, the pick & place robot is
Keywords— Robotic Hand Manipulator; Product designed for the automatic warehouse management system.
Management; GSM; Industrial Automation.

II. LITERATURE REVIEW

I. INTRODUCTION The warehouse automation is a new concept that requires
In upcoming Industry 4.0 robot plays a significant role in warehouse products movements automatically. Many
the industrial manufacturing process. Mainly, with shorter researchers and engineers have designed & developed
time and precision, the robot can perform a specific task automatic warehouse management pick and place robot
precisely. In the market, different sorts of robots can be found operations over the past few decades. The concept of
to perform different tasks. Some robots can perform tasks warehouse automation is discussed by Chuan.et.al (2007).
similar to humans, which are known as “Androids”, though From this model, it is discussed RFID which is one of the
several robots aren’t manufactured like the human model. For latest drifts in industry. For the fourth industrial revolution, the
example, industrial robots are mainly designed for continuing possible application can range from warehouse to library
repetitive tasks, which isn’t like with human-like construction. management. This research is aimed to create a self-governing
Even some robots can be controlled remotely from a long robot with an RFID signal application. As the main
distance by a human operator. components, this project comprises with microcontroller and
RFID system. This robot can identify the items by reading the
According to that definition of Roboticist Anca Dragan of tag on the items. This robot picks up an item and goes to the
UC Berkeley, robotic decisions should be made by themselves destination with the help of a line follower module to place the
and it should be useful enough. Decisions should be like item in a specific location. The novel concept is proposed for
dodging obstacles like trees or other objects. Robotics a flexible robot arm (kato.et.al (2013) which represents the
design deals with the design, construction, operation, and design for pick and place applications of lightweight products.
usage of robots, as well as computer systems for their sensory Mainly, a one-stepper motor and two DC motors are used for
feedback, control, and data processing. These technologies are robotic actuation. For a five DOF robotic manipulator, the
responsible to develop new types of machineries which can visual servo control is done to perform pick-and-place tasks is
replicate human actions, that in turn substitute for humans. In discussed in Sai.et.al (2015). Recently the latest hybrid
the sphere of robot technology, the pick and place robots are switched technique including reactive servo control structure
widely used for manufacturing processes especially in car is introduced, encouraged by a sensing-and-reaction behavior
manufacturing (eg: BMW, Mercedes Benz, and Honda etc.),

978-1-7281-7523-2/21/$31.00 ©2021 IEEE

Authorized licensed use limited to: Amrita School Of Engineering - Kollam. Downloaded on October 18,2024 at 06:52:26 UTC from IEEE Xplore. Restrictions apply.
. 156
 2021 IEEE 7th International Conference on Smart Instrumentation, Measurement and Applications (ICSIMA2021)

with logic-based approaches is proposed to handle this issue A. System Framework

proficiently. This system uses both position-based and image- The pick & place robot hand controlling system is
based reactive planning methods. For moving the objects from controlled by microcontroller and for demo arduino mega is
one place to other in industrial automation, the planar robot is used. In this automatic controlling and detection system,
very appropriate (Wen-Bo Li (2015)). In this paper, 4 DOF arduino is connected with a potentiometer. Here the
SCARA robot is developed to implement pick-and-place tasks potentiometer makes an angle and the arduino controlling unit
of rectangular and circular work pieces. For smart warehouse makes an angle and voltage step down power supply
management, COSO framework approach is developed for the automatically in servo motor 1 rotate the base. As well as step
evaluation of internal control activities. Some researchers use by step coordinates the angle of the robot hand & position the
image processing techniques to find actual objects (Roshni final part. Finally, pick & place as well return to the exact
et.al (2017)), whereas some other researchers are worked with place. Figure 3.1 shows the block diagram of the automatic
computer vision techniques to sort out different objects to pick & place robot for implementing our warehouse system:
place in different predetermined locations by color and height
(Shaikat et. al (2020)). In 2016, Andhare.et.al. (2016)
presented research work on warehouse management.
According to this, the axis x, y is originated by converting
pixel coordinates into real-world co-ordinates by using 2D
transformation. In this work, computer vision is used, where
the task assignment problem of warehouse robots in the smart
warehouse (TAWRSW) is investigated. The research work of
Kattepur.et.el-2018 shows a model, which indicates the Fig. 2. Block diagram of warehouse management by implementing Pick
interactions with the workflow of Industry 4.0 warehouse and place robot operation
operations by using the coexisting programming language
Orc. This work mainly focuses on the technique of B. Robot Operation
distinguishing the object from it surrounding using android For the mechanical structure of this work, the servo motor-
based mobile applications and robotic operation through its based pick & place robot hand is used. With this other
center of gravity. A system workflow of a warehouse robotic accessories like the hand, joints, and parts are used. Mainly
system is given below: the Solidworks are used to make the framework. The figure
represents a 3D Mechanical design pick & place robot. This is
used for making operation of pick & place & specified object.
There are parts names such as Base, Hand Manipulator, Grip,
etc. The mechanical design of the robotic structure is given
below:

Fig. 1. A system workflow of a warehouse robotic system

Firstly, products are carried out in a conveyor belt. Then, if the

product is similar to the original product and has no defect,
then it moves to another conveyor belt. Otherwise, it will go Fig. 3. Mechanical 3D design of pick and place robot
back to the first position. After the original product will reach,
the 5-DOF robotic arm will pick the product and place it to a The mechanical section of this project is designed with the
specific location, which will pick by cart. It will move the following components which are shown in table.
product to the specific storing place and delivery notes will be
taken afterwards. Then it will move to the place, where the TABLE I. MECHANICAL EQUIPMENTS
product will pick up by the FEFO system. After that, another
cart will take that load and moves the products to loading COMPONENTS QUANTITY DESCRIPTION
docks. Then this loading dock will be shifted by carrying van. Used to transfer pick & place
Afterward, these goods are selected based on the expiry date. Gripper 01
from one place to another
Used for moving to the whole
Robot Hand 01
system.
III. METHODOLOGY
For the development of the model robot, a prototype A mechanical gripper is a significant part of this hand
system has been developed. The establishment for an manipulator and the gripper is operated to grabbing the object
automatic warehouse management by implementing pick & which is the end effector of a robotic arm. In this project, we
place robot hand operation, a system detection with design and build a gadget that can be added to the robotic arm
controlling system and an arduino operation is framed. Lastly, or any mechanism which needs grippers. Mainly the gripper is
Assembling all the individual systems and operation in a unit a tool that allows the grasping of an object to be manipulated.
place, our work is developed. The human hand can be an example to realize about gripper.
The gripper enables the holding, tightening, handling, and

Authorized licensed use limited to: Amrita School Of Engineering - Kollam. Downloaded on October 18,2024 at 06:52:26 UTC from IEEE Xplore. Restrictions apply.
157
 2021 IEEE 7th International Conference on Smart Instrumentation, Measurement and Applications (ICSIMA2021)

releasing of an object. The mechanical design of the robotic

hand gripper is given below:

Fig. 4. Mechanical design of robotic hand gripper

C. Circuit Layout
The essential electrical apparatus are given below:

TABLE II. ELECTRICAL APPARATUS

Components Quantity Description

Arduino Mega 01 Used to control motor.
Used to control the system by
Servo Motor 05 Potentiometer
Potentiometer 05 Used to control servo motor
Power Supply 01 Converted AC to DC
Fig. 6. System Flow Chart
Voltage Used to control the functions of
Regulator 05 input and output circuits. At first, the robotic hand searches for the object. When the
Used to display for shown the sensor detects any signal, it gives the signal to robot and the
LED Display 05 output. robot is rotate and change the gripper position by using
potentiometer. Finally, the gripper picks the object and places
The circuit layout is given below. this in desired location.

IV. MATHEMATICAL ANALYSIS

A. Denavit-Hartenberg (D-H) Method
This robot having established a coordinate system to each
link of manipulator 4X4 transformation matrix relating two
successive two successive coordinate systems can be
established. The resulting transformation matrix is given,
i −1
Ai = T ( z, d )T ( z,θ )T ( x, a)T ( x,α )
After calculating from D-H homogeneous transformation
matrices, the resulting transformation matrix is given by,
cθi − cα i sθ i sα i sθ i ai cθ i 
 sθ cα i cθi − sα i cθ i ai sθi 
i −1
Ai =  i
0 sα i cα i di 
 
Fig. 5. Circuit Layout of Robotic Manipulator 0 0 0 1 

cθ 234 0 − sθ 234 a3cθ 23 + a2cθ 2 

sθ 0 cθ 234 a3 sθ 23 + a2 sθ 2 
1
A4 =  234
D. Working Flow Chart  0 −1 0 0 
 
From the working flow chart, the total warehouse  0 0 0 1 
management system is described. The flowchart is given
below.
The D-H parameters of a 5 DOF manipulator are given
below:

Authorized licensed use limited to: Amrita School Of Engineering - Kollam. Downloaded on October 18,2024 at 06:52:26 UTC from IEEE Xplore. Restrictions apply.
158
 2021 IEEE 7th International Conference on Smart Instrumentation, Measurement and Applications (ICSIMA2021)

TABLE III. D-H PARAMETERS OF 5-DOF MANIPULATOR The given θ1, θ2 and θ3, the values of theta is used to
compute the position of point Q. By this method, it would be
For the model the a1=15 cm; a2=7 cm; a3= 5.3 cm and for easy to find the position vector of any point in the end effector.
the movements d1=10 cm, d5=20cm. The output of this model The angle θ1+ θ2+θ3 is the end effector orientation angle.
is given below:
Joint i αi ai di θi
1. -π/2 a1 d1 θ1
2. 0 a2 0 θ2
3. 0 a3 0 θ3
4. -π/2 0 0 θ4 The resultant matrix of direct kinematics is given below:
5. 0 0 d5 θ5

The above result mentions that the robot is in a steady

state. When it rotates a hundred eighty degree counter clock
wise direction, it provides a transformation from the fourth C. Inverse Kinematics
coordinate system to the first coordinate system. The value of
The location of end effector is needed for calculating the
θ1, θ2 ,and θ5 is required to form the operation.
inverse kinematics.
The robot is moving from zero degree to one eighty degree
for the demonstration and it is moving hundred and eighty
degree rotation in forward and reverse positions. It represents
the data below:

Fig. 7. Inverse Kinematic Solution for Robotic Manipulator

To calculate the inverse kinematics the calculation of D-H

method analysis is essential. To solve the inverse kinematics,
vector loop method becomes more efficient for analysis
robotic manipulator. Using vector algebra, the position vector
of wrist center P can be calculated with origin Q. The
equations are given below:

Px = qx − a3cφ
Py = qy − a3sφ
Thus, the orientation angle Φ is found which is
combination of Theta1, Theta 2 and Theta 3.
Φ=θ1+θ2+θ3. Taking the considerations of X and Y axis,

B. Direct Kinematics Px = a1cθ1 + a2cθ12

The direct Kinematics expressed the end-effector
coordinate system by position vector Q. The position vector Q Py = a1sθ + a2 sθ12
is calculated with respect to the base coordinate system. The
forward or direct kinematics is shown by following matrix: The transformation matrix is given by below,
 qx  0 a1cθ 1+ a 2 cθ 2+ a3 cθ 123   cφ − sφ 0 qx 
q   0   a sθ + a sθ + a sθ   sφ
 y  = 0 A3   =  1 1 2 2 3 123  cφ 0 q y 
 qz  0  0 
0
A3 = 
0 0 1 0
       
1 1   1  0 0 0 1

Authorized licensed use limited to: Amrita School Of Engineering - Kollam. Downloaded on October 18,2024 at 06:52:26 UTC from IEEE Xplore. Restrictions apply.
159
 2021 IEEE 7th International Conference on Smart Instrumentation, Measurement and Applications (ICSIMA2021)

works properly but in real life it is affected by some

cθ123 − sθ123 0 a1cθ1 + a 2 cθ12 + a3cθ123 
attenuation. To overcome this attenuation the PID tuner is
 sθ cθ123 0 a1sθ1 + a 2 sθ12 + a3 sθ123  applied.
=  123
 0 0 1 0  E. Euler Angle Representation
  Euler theorem is used to get general displacement of a rigid
 0 0 0 1  body with one point fixed with rotational axis body. Each
The output of the inverse kinematics of this manipulator is rotation is to find an axis through the fixed point oriented at
given below: particular polar angles θ, Φ and Ψ. These three angles
represent roll, pitch and Yaw. For three successive rotation,
X’=Ax with A=BCD
The total Euler angle is shown below:
 cosψcosφ −cosθsinφsinψ cosψcosφ +cosθ cosφsinψ sinψsinθ
A=−sinψcosφ −cosθsinφsinψ −sinψsinφ +cosθcosφcosψ cosψsinθ
 sinψsinφ −sinψcosφ cosθ 
D. Wrist Center Position For manipulation the angle would be considered, as θ=30,
The wrist center position with respect to and the position Ψ=-90 and Φ=90. The output curve would be,
vector is given below:
 Px   q x − d 6 wx 
 P  q − d w 
0
p = op =  y   x 6 y

 Pz   q x − d 6 wz 
   
 1  = 1 
The gripper coordination system is given below:
Fig. 10. The resultant curve of X,Y and Z axis

F. PID Control
The PID tuner is used to overcome the signaling error.
The three parameters which include proportional, derivative
and integral are used to tune the PID. The PID tuning model
for this manipulator is given below:
Fig. 8. Gripper Coordination System and Wrist Center Position

Transforming coordinate system into position vector is

given below.
0
p = 0A3 3 p

The output curve of this manipulator is divided in three

sections and these sections are related with directions of Fig. 11. MATLAB Model of PID Tuning and Process Control
manipulator and movement velocity. The direction-curve is
given below. The experimental PID tuning model is given below:

The manipulator velocity curve is below:

Fig. 9. Robotic Arm Velocity Movement

Fig. 12. Experimental PID Tuning Model
When the signal comes manipulator works and it stays in
desired position until the new signal comes. In simulation it The output of the PID model is given below:

Authorized licensed use limited to: Amrita School Of Engineering - Kollam. Downloaded on October 18,2024 at 06:52:26 UTC from IEEE Xplore. Restrictions apply.
160
 2021 IEEE 7th International Conference on Smart Instrumentation, Measurement and Applications (ICSIMA2021)

Fig. 16. Final set up of Robotic hand with mobile application

Fig. 13. Input and Output curve of PID Tuning Model V. CONCLUSION
This work enables the user to create the shortest time
required to pick and place a product in a specific location.
Moreover, Pick & place robot is not only used for internal
control but also quantity of concession hazard due to revising
the organizational operation as well. Experimental results
validate that the robot can adjust the work pieces with less than
2 pixels error for circular & rectangle work pieces & develop
robot is feasible & valid.

Fig. 14. Bode plot of frequency and time of PID Tuning Model
REFERENCES
G. Mobile Application [1] Loh Poh Chuan, Ayob Johari, Mohd Helmy Abd Wahab, Danial Md.
Nor, Nik Shahidah Afifi Md. Taujuddin, Mohd Erdi Ayob, “An RFID
To automate the system a mobile application is introduced Warehouse Robot”, International Conference on Intelligent and
in this project. With this a Bluetooth application is introduced Advanced Systems 2007.
and finally the robot is connected with GSM. The figure of [2] Gustavo Kato, Diego Onchi, Monica Abarca. "Low cost flexible robot
mobile application is given below: manipulator for pick and place tasks", 2013 10th International
Conference on Ubiquitous Robots and Ambient Intelligence (URAI),
2013
[3] Chi-Yi Tsai, Ching-Chang Wong, Chia-Jun Yu,Chih-Cheng Liu,
Tsung-Yen Liu. "A Hybrid Switched Reactive-Based Visual Servo
Control of 5-DOF Robot Manipulators for Pick-and-Place Tasks",
IEEE Systems Journal, 2015
[4] Pratiksha Andhare, Sayali Rawat. "Pick and place industrial robot
controller with computer vision", 2016 International Conference on
Computing Communication Control and automation (ICCUBEA),
2016
[5] N Roshni, T K Sunil Kumar. "Pick and place robot using the centre of
gravity value of the robot using the centre of gravity value of the
moving object", 2017 IEEE International Conference on Intelligent
Techniques in Control, Optimization and Signal Processing (INCOS),
2017.
[6] Abu Salman Shaikat, Suraiya Akter, Umme Salma, “Computer Vision
based Industrial Robotic Arm for Sorting Objects by Color and
Height”, Journal of Engineering Advancements (JEA), 2020.
[7] Ratna Sari, Raymond Kosala, Benny Ranti, Suhono Harso Supangkat.
"COSO Framework for Warehouse Management Internal Control
Evaluation: Enabling Smart Warehouse Systems", 2018 International
Conference on ICT for Smart Society (ICISS), 2018
[8] Chi-Yi Tsai, Ching-Chang Wong, Chia-Jun Yu, Chih-Cheng Liu, and
Tsung-Yen Liu, “A Hybrid Switched Reactive-Based Visual Servo
Fig. 15. Mobile application for regulating the robotic manipulator Control of 5-DOF Robot Manipulators for Pick-and-Place Tasks”,
IEEE SYSTEMS JOURNAL, VOL. 9, NO. 1, MARCH 2015.
With introducing Mobile application, the automation of [9] Peters, Brandon A. C., Carl A. Moore, and Rodney G. Roberts.
warehouse robotic system would become more sophisticated and "Teleoperaton setup for wire-cutting application in the field",
precise. The below figure represents the demo operation of this Proceedings of the 2012 44th Southeastern Symposium on System
Theory (SSST), 2012.
model.
[10] Suman Jana, Arvind Narayanan, Vitaly Shmatikov “A Scanner Darkly:
Protecting User Privacy From Perceptual Applications”, 2013 IEEE
Symposium on Security and Privacy.
[11] D. Braganza, M. L. McIntyre, D. M. Dawson, and I. D. Walker, “Whole
arm grasping control for redundant robot manipulators,” in Proc. Am.
Control Conf., Minneapolis, MN, USA, 2006, pp. 3194–3199.

Authorized licensed use limited to: Amrita School Of Engineering - Kollam. Downloaded on October 18,2024 at 06:52:26 UTC from IEEE Xplore. Restrictions apply.
161