THE OFFICIAL JOURNAL OF ENGINEERING RESEARCH

INTELLIGENT SHOPPING CART

Kevin Dwight H. Duroja, Michael Ryan R. Cadorniga,
Paul Eugene R. Cagara, and Shiela Mae V. Quinamot

ABSTRACT

This study aimed to create a shopping cart that will automatically follow
the shopper so that there will be no human intervention upon pushing
the cart. Also, this study aimed to create a budget system application
for the shopper to be aware of his/her budget. The system is composed
of a cart which houses the whole circuit, microcontroller, infrared,
radio frequency receiver, ultrasonic sonar driver, windows tablet that
has the budget system application, and a barco de scanner which scans
the barcode of the products. The other system is a wearable gadget
for the shopper which accommodates the infrared and radio frequency
transmitters. The study employed both qualitative and quantitative
methods of research to produce the essential data. The methods used
were experimental method, which involved expansion and integration
of existing studies, and descriptive approach, which involved evaluation
through implementation and actual installation. The result of the
evaluation confirms that the device functioned all of its features. The
evaluators were interested on the idea of the Intelligent Shopping Cart
being used in supermarkets and surplus warehouse. With this device,
the shoppers are more aware of their budget upon shopping and are
able to move conveniently without pushing the cart.

Keywords: Design project, intelligent shopping cart, infrared, radio

frequency transceivers, ultrasonic sonar driver

PULSAR•291
THE OFFICIAL JOURNAL OF ENGINEERING RESEARCH

INTRODUCTION

As technology progressed, lives have been significantly improved

due to the emergence of laborsaving and intelligent utilities. Supermarket
shopping is considered one of the biggest causes of human problems in
modern day to day life. Also, it is one of the biggest expenses faced
by any family or household. People are not concern and they take less
effort in purchasing items without noticing that they run out of budget.
They will only realize this problem when they are in the cashier already.
A shopping cart is a bag or basket on wheels for carrying shopping
purchases used by supermarket customers. It is provided by the retail
store for a customer to use in collecting purchases (Foster, et al., 2009).

Most modern shopping carts are made of metal or a combination

of metal and plastic and have been designed to nest within each other
in a line to facilitate collecting and moving many at one time and also
to save on storage space. The carts can come in many sizes, with larger
ones able to carry a child. There are also specialized carts designed for
two children, and electric mobility scooters with baskets designed for
disabled customers. In the United States, customers are allowed to leave
the carts in the parking lot, and store personnel will return the carts
to the storage area. In Europe and Canada however, coin (or token)
operated locking mechanisms are provided to encourage shoppers to
return the carts to the exact location after use (Awati, S.B et al, 2012).

However, the shopping carts in major stores have experienced

little changes and served only simple purposes ever since they were first
manufactured. Pushing these carts around becomes noticeably painful
when they carry heavy loads. Even though people are not moving fast in
stores, there is a risk of collision if obstacles arise out of sight. Moreover,
products can be extremely difficult to locate before people get familiar
with the store layout. Considering all the disadvantages discussed
above, the researchers aimed to implement an intelligent shopping cart
that provides great convenience and efficiency to customers (Fan, et
al., 2013).

292•PULSAR VOLUME 3 • MARCH 2014

 THE OFFICIAL JOURNAL OF ENGINEERING RESEARCH

Researchers at IBM recently assembled several of the high-

tech machines for a demonstration at their Industry Solutions Lab in
Hawthorne. Among them were the smart shopping cart, a computerized
produce scale called “Veggie Vision,” and a fascinating projection
tentatively dubbed the “Everything Display.” Some are being tested in
stores while others are in various stages of development. Other companies
including NCR, Fujitsu and Hewlett-Packard are also working on similar
products, sometimes in partnerships (Jones, et al., 2003).

This device is not popular in the Philippines. Most of the time,

the shoppers at the supermarket have experienced some difficulties on
pushing the shopping cart and budgeting their money. This problem
must be solved in order for the shoppers to be convenient at the
supermarket. This device aims to lessen the human effort by producing
a shopping cart that can automatically follows the user and calculate
the amount of the total items that they want to purchase.

Conceptual Framework

INPUT PROCESS OUTPUT

• Encode the • Compare • Activate/deactivate the
shoppers’ the distance motor based on the
budget between the data analyzed by the
• Barcode user and the microcontroller.
Scanner cart. • Activate if the distance is
• Compare greater than 0.5m
the data • Deactivate if the distance
collected is less than 0.5m
from the
keypad and • An alarm will activate if:
the items – the amount exceeds from
scanned from the allotted budget.
the barcode. – the distance between
the cart and the user is
greater than 2m.

Figure 1. Framework of the Study

PULSAR•293
THE OFFICIAL JOURNAL OF ENGINEERING RESEARCH

The intelligent shopping cart allows the shoppers to input their

budget in order for them to be aware of their allotted budget. In order
for the shoppers to be aware of the price, they will need to use the
barcode scanner attached at the shopping cart. Then the barcode scanner
is connected to the tablet which has software that would analyze the
barcode being scanned and then allowed the shopper to monitor all of
the items that they purchased. The infrared transmitter and receiver
installed both in the user and the cart respectively will compare the
distance between the user and the cart.

The motor will be activated and deactivated with the desired speed
based on the sensor. If the shopper is too far from the cart, an alarm will
be activated both in the user alarm and cart. Also if the allotted budget
that has been scanned by the barcode scanner and inputted at the tablet
exceeds the amount, an alarm will be activated to aware the shopper
that they already exceeds to their desired budget.

This project will involve two systems and it will be installed in both
the shopper and the cart. The cart, which has a maximum load of fifty
(50) kilograms, will automatically follow the shopper and will avoid
any obstacles in front. Since the microcontroller has a limited memory,
the number of products to be scanned in the barcode scanner will be
limited at twenty (20) items. The budget that has been entered by the
user as well as the total amount being incremented by every product
scanned will be shown at the tablet.

Objectives of the study

This study aimed to lessen the effort of the shoppers at the

supermarket. It helps the shoppers to buy their items without the
inconvenience of pushing the cart with heavy load. The intelligent
shopping cart is the one who will carry all the items that shopper
wanted to purchase and it has the ability to follow the shopper in an
independent way. Also, this study offers the shoppers the convenient
way on how to budget their money by updating the shopper about

294•PULSAR VOLUME 3 • MARCH 2014

 THE OFFICIAL JOURNAL OF ENGINEERING RESEARCH

their current cost that they want to purchase. This intelligent shopping
cart can sum up the amount of the items that have been placed into
the cart. In that way, the shoppers can now budget their money and
they will be aware if their allotted money is sufficient. The researchers
wanted to design an intelligent shopping cart that is able to make the
shoppers more comfortable when shopping. The researchers wanted to
test the functionality of the cart to know its capabilities. Furthermore,
the researchers wanted to explore its marketability and commercial
viability.

METHOD

This study used both qualitative and quantitative methods of

research to analyze the collected data to address the objectives of this
research, specifically experimental survey and descriptive designs. The
first method used was experimental development. The experimental
development method is a repetitive process that uses trial and error
approach. It was suitable for the study because the researchers used
experimental and scientific approaches all throughout the development
of the project. The gathered data was monitored and was used for
further development of the project. The study would focus on how to
gather data from repeated experiments before and after treatment to
ensure the functionality of the system. The experimental development
design involves the development of infrared receiver module, ultrasonic
sensor, microcontroller, motors, the tablet, and all other components.
Descriptive method is the gathering of qualitative and quantitative
data in testing the hypothesis or answer questions about the subject
of the study. Descriptive approach of evaluation helps the researchers’
response of solutions to problems.

The study was conducted at the University of the Immaculate

Conception, Annex campus located at Bonifacio Street, Davao City. This
study was evaluated at a specific local supermarket. The researchers
tested the functionality of the system on grocery stores to ensure that
the features of the system will be beneficial for the respondents. The

PULSAR•295
THE OFFICIAL JOURNAL OF ENGINEERING RESEARCH

design tested its versatility and reaction towards people that were
currently shopping at the said location.

There were thirty (30) respondents and evaluators in this said

project. The project was evaluated by six (6) shoppers, six (6) market
supervisor, six (6) cashier or teller, six (6) engineering faculty, and six
(6) UIC students. The researchers demonstrated the procedures on how
to use it. The shoppers tested the shopping cart. A device is attached
to the selected shopper in order for the cart to follow the signal. For
the evaluation of the performance of the intelligent shopping cart,
the respondents rated the accuracy of the device, the range of the
connectivity during the activation of the cart, and functionality of the
barcode scanner.

Technical Flowchart

BLOCK DIAGRAM

The system consisted of a DC battery, voltage regulator, 12v Wiper

motor, infrared and RF transceiver and receiver, microcontroller,
ultrasonic sensor, tablet, barcode scanner, LED, and a buzzer. The
DC battery supplies all the components that need power. The voltage
regulator provides the appropriate voltage that a specific component
needs. The DC motor is responsible for the mechanical part of the cart.

296•PULSAR VOLUME 3 • MARCH 2014

 THE OFFICIAL JOURNAL OF ENGINEERING RESEARCH

The IR Transmitter (shopper) will be the one that transmits the data
to the IR Receiver (cart). It will determine the position of the user
and cart and will send the information to the microcontroller. The
microcontroller will then process that information and gives signal to
the motor. The motor will then respond with what the microcontroller
sends. The speed of the motor will vary, stop and start again depending
on the information. An ultrasonic sensor is the detector for the
obstruction. A LED and a buzzer served as a warning device when the
cart is more than 2 meters away from the user. Barcode scanner will
scan the products and the price. The software in the Tablet will be
responsible for monitoring the items being purchased and prompts when
the amount exceeds. The researchers will start with the manipulation
of the infrared transmitter and receiver and ultrasonic sensor. Since,
these modules are responsible for the movement of the cart; there is a
need to conduct series of testing in order to test for the functionality of
these devices. The researchers investigated more about programming
these devices. Afterwards, the researchers synchronized these devices
with the motor. Modifications were made in case of any inaccuracy of
the system. The barcode scanner was connected to the Tablet via USB
and a software was created for the monitoring of the purchases of the
shopper. The final construction of this design was the fabrication of the
cart. The components must be intact and placed under it so that it will
not be a nuisance.

The Likert scale shown below will be used for the statistical
treatment of the data. It is used to measure the level of accuracy and
functionality of the intelligent shopping cart. The highest rating is 5
which correspond as “Strongly Agree” and the lowest rating is 1 which
is labeled as “Strongly Disagree”. The scale level describes the response
of the evaluators based on their observation and evaluation of the
system at actual implementation and application.

PULSAR•297
THE OFFICIAL JOURNAL OF ENGINEERING RESEARCH

Table A. Interpreting the mean level

Mean Level Responses Description Interpretation

4.20 - 5.00 Strongly Very High It means that the system

Agree exhibit the feature being
asked all the time
3.40 - 4.19 Agree High It means that the system
exhibit the feature being
asked most of the time
2.60 - 3.39 Moderately Moderately High It means that the system
Agree sometimes exhibit the
feature being asked
1.80 - 2.59 Disagree Low It means that the system
rarely exhibit the feature
being asked
1.00 - 1.79 Strongly Very Low It means that the system
Disagree never exhibit the feature
being asked

298•PULSAR VOLUME 3 • MARCH 2014

 THE OFFICIAL JOURNAL OF ENGINEERING RESEARCH

RESULTS AND DISCUSSION

Table 1. Mobility of the shopping cart based on the sensors

Descriptive
Statements Mean Interpretation
Rating

1. The cart will automatically 4.3 Strongly It means that the

follow the shopper within Agree system exhibit the
0.5m to 2m. feature being asked
all the time
2. The cart can detect obstacles 4.6 Strongly It means that the
0.5m away. Agree system exhibit the
feature being asked
all the time
3. It can detect barcodes of the 4.7 Strongly It means that the
products. Agree system exhibit the
feature being asked
all the time

It means that the

Strongly system exhibit the
Overall Mean 4.5
Agree feature being asked
all the time

Table 1 shows the shopping cart’s capability to follow the shopper

based on the evaluation of thirty (30) respondents. As shown in the
table, the evaluators strongly agree with the statements; the overall
mean yielded to 4.5. This implies that the system is a unique idea
because it will help the shoppers move freely and lessen the labor of
pushing the cart while shopping at the grocery.

As to the individual items, statement 1 and 2 got a mean value of

4.3 and 4.6 respectively which corresponds “strongly agree”. Although,
they rated it high, qualitative data reveals issue on the movement of
the cart that it moves side-by-side and the obstruction sensors are very
sensitive in detecting an obstacle in front. Thus, the reason is that the
receiver sensors are positioned on the side. When the left sensor is

PULSAR•299
THE OFFICIAL JOURNAL OF ENGINEERING RESEARCH

activated, the right motor will activate and vice versa. Furthermore,
the researchers decided to put the sensors that way, so that when the
shopper turn left and right, the cart quickly respond to their action.
As the transmiter can emit signals to any of the sides, the sensors
can detect. Lastly, the barcode scanner yielded a mean value of 4.7
which meas “strongly agree”, shows that the scanner performs well on
scanning the products barcode.

Table 2. Weight capacity of the system

Descriptive
Statements Mean Interpretation
Rating

1. The cart can carry a 4.5 Strongly It means that the

maximum of 50 kilograms of Agree system exhibit the
weight. feature being asked
all the time

It means that the

Strongly system exhibit the
Overall Mean 4.5
Agree feature being asked
all the time

Table 2 shows the movement of the cart when it has a load of 50

kilograms. It can be observed on the gathered data with a mean of 4.5
in which the evaluators strongly agree. This illustrates that the system
is capable to carry a load of 50 kilograms.

300•PULSAR VOLUME 3 • MARCH 2014

 THE OFFICIAL JOURNAL OF ENGINEERING RESEARCH

Table 3. Features of the intelligent shopping cart (Audio indicators)

Descriptive
Statements Mean Interpretation
Rating

1. 2 meters away from the 4.4 Strongly It means that the

shopper (The cart will alarm) Agree system exhibit the
feature being asked
all the time
2. It will alarm if the sum 4.7 Strongly It means that the
exceeds the input budget. Agree system exhibit the
feature being asked
all the time

It means that the

Strongly system exhibit the
Overall Mean 4.5 Agree feature being asked
all the time

Table 3 shows the result of the alarms upon the detection of the
status of the cart and the database of the products. The overall mean
value is 4.5 means that the evaluators’ observes that the alarm triggered
when the conditions were met.

PULSAR•301
THE OFFICIAL JOURNAL OF ENGINEERING RESEARCH

Table 4. Features of the intelligent shopping cart (Budget System

Application)

Descriptive
Statements Mean Interpretation
Rating

1. It can display the products’ 4.7 Strongly It means that the

prices in the tablet. Agree system exhibit the
feature being asked
all the time
2. It can ADD/SUBTRACT 4.7 Strongly It means that the
product prices. Agree system exhibit the
feature being asked
all the time
3. It can input the desired 4.7 Strongly It means that the
budget in the tablet. Agree system exhibit the
feature being asked
all the time

It means that the

Strongly system exhibit the
Overall Mean 4.7 Agree feature being asked
all the time

Table 4 shows the evaluation on the software application of the

database of the products. The overall mean of 4.7 indicates that the
application is fully functional and it satisfies all the features indicated.
Thus, it shows that the evaluators were satisfied with the features of
the device.

302•PULSAR VOLUME 3 • MARCH 2014

 THE OFFICIAL JOURNAL OF ENGINEERING RESEARCH

CONCLUSION AND RECOMMENDATION

Based on a series of testing, the cart can move effectively in an
open and spacious aisle with less people. Narrow, cramped and crowded
aisles were found out to be ineffective because the obstruction sensors
are sensitive and the sensors will not activate the motor.

According to the feedbacks of the evaluators, the researchers have

come up with the recommendation and improvement of our design. The
price must be displayed after scanning the product so that the shopper
will be aware on the quantity that he/she will input. The motor used
in our design is just a cheap wiper motor. If we used a high class motor
and at the same time it is being welded properly, the movement of the
cart will be smooth and responsive.
 

LITERATURE CITED

Awati, S.B., et al. (2012). Smart Trolley in Mega Mall January

4, 2013 from http://www.ijetae.com/files/Volume2Issue3/
IJETAE_0312_82.pdf

Fan, D., et al. (2012). Smart Shopping Cart, June 23, 2013 from
http://courses.engr.illinois.edu/ece445/projects/spring2013/
project22_proposal.pdf

Foster, D., et al. (2009). Grocery shopping with an Autistic

Child: Problems & Solutions, June 13, 2013 from http://
voices.yahoo.com/grocery-shopping-autistic-child-problems-
solutions-3308440.html?cat=25

Jones, A., et al. (2003). Smart carts, Veggie Vision in grocery stores
to come, retrieve July 2013 from http://usatoday30.usatoday.
com/tech/news/techinnovations/2003-09-26-future-grocery-
shop_x.htm

PULSAR•303