Design, Build and Play - Online Robotics Classes

Full text

Design, Build and Play - Online Robotics Classes
Tiago Caldeira
Instituto Superior T´
ecnico
University of Lisbon
Lisboa, Portugal
tiago.caldeira@tecnico.pt
Manuel Reis Carneiro
Institute of Systems and Robotics
University of Coimbra
Coimbra, Portugal
manuel.carneiro@isr.uc.pt
Abstract—In the last few years, Robotics has been one of the
most sought after activities for teens who were looking to engage
as a complement to their school classes. Many clubs and associ-
ations provided simple activities focused on entertaining young
minds after-school but as those projects were mostly conducted
face-to-face, the students were left without alternatives since the
pandemic started. Design, Build and Play is an online-based
program in which introductory robotics classes are conducted
remotely to introduce key iSTEME concepts.
Index Terms—STEM, Remote Education, Robotics, Innova-
tion, iSTEME
I. INTRODUCTION
Since the industrial revolution that teaching and learning
methods kept evolving, but over the last decades one aspect
was kept almost constant - the classroom. The technology
was progressively been added to the class, introducing tablets,
computers, and interactive boards, essentially developing the
concepts of Smart Education, Smart Pedagogy, and Smart
Classrooms [1]. Those changes kept the room as the center
of education, and despite the advances in many areas, online
classes represented only a small portion of the total classes.
In this research paper, we introduce a method that uses
Educational Robots to teach iSTEME concepts to students.
The objective of the course was to introduce a low-cost edu-
tainment solution [2] that would allow the students to stay safe
at home and engage in a pedagogic activity together with their
families. In this publication, we will start with the inspiration
that leads to this research (Section II). We then introduce
the method used during the course (Section III), the results
(Section IV), and analysis (Section V). Finally, we will present
the conclusions and potential areas for expansion(Section VI).
A. From STEM to iSTEME
The choice of STEM as an acronym keeps evolving, and
different authors define different stages and acronyms which
added other disciplines. In this research, the focus will be
on iSTEME, which consists of a modern evolution of the
traditional STEM and better reflects the scope of this work.
STEM -Science, Technology, Engineering, and Math
results on an acronym change (from original, but less known
SMET) proposed by Dr. Peter Faletra, and was first used
after a meeting on science education held at the US National
Science Foundation in the early ’90s. It started as a program
for talented students and it mostly focused on creating multi-
disciplinary lessons.
iSTEME -innovation, Science, Technology, Engineering,
Math and Entrepreneurship. A new concept that doesn’t
focus on the Language and Arts, but makes Innovation the
starting point before focusing on traditional Science, Technol-
ogy, Engineering, and Math. That is then complemented with
the Entrepreneurship mindset [3].
B. Current Pandemic
The Coronavirus 2019 (COVID-19) made serious impacts
on many aspects of everyday life. The world saw a paradigm
shift in the education system favoring online learning [4] as a
result of the constraints of movement imposed by different
governments worldwide during the pandemic, which often
leads to an unprecedented academic crisis worldwide. The
COVID-19 pandemic has created significant challenges for the
global higher education community, which lead to different
responses all over the world by numerous Universities [5].
Additionally affected by the pandemic were the extracurricular
activities, with many associations and clubs being closed, as
the infrastructure to adapt to the new requirements were not
in place in smaller institutions.
II. IN SP IR ATIO N
The team conducting the research conducted relevant work-
shops in the past, mostly face-to-face and that served as the
starting point. The main inspiration for the design of the online
course was the Bots4You Kit (Fig. 1), which was the center of
many face-to-face training and competitions, therefore some
teaching material was already developed, and the source code
was available as well.
This platform was fabricated using a laser cut machine and
it is easy to customize due to the modular design. This kit is
open-source and is based on the Arduino microcontroller. Ar-
duino is worldwide available and multiple tutorials are already
designed to take advantage of the simplicity to start using it.
Due to its open-source nature, different designs were created
introducing additional features such as low power Bluetooth,
motor controllers, and headers for a simple interface with
sensors.
Having in mind the importance to make the course multi-
disciplinary, instead of using a commercial kit, each element
of the robot’s body was designed during the course.978-0-7381-1321-0/21/$31.00 ©2021 IEEE
2021 IEEE World Conference on Engineering Education (EDUNINE) | 978-1-6654-0302-3/21/$31.00 ©2021 IEEE | DOI: 10.1109/EDUNINE51952.2021.9429157
Authorized licensed use limited to: b-on: Universidade de Coimbra. Downloaded on May 20,2021 at 13:52:30 UTC from IEEE Xplore. Restrictions apply.

Fig. 1. Bots4You Robot design.
A. Design and 3D Printing
In a virtual world, the design possibilities are unlimited,
with an infinite amount of configurations and shapes that
could be selected for the robot. Once the preferred option is
selected, it can be fabricated making use of digital fabrication
machines [6] like laser cutters, CNC machines, or 3D printers.
Before starting the course, a previously prepared parametric
design optimized for 3D printing was selected and then
optimized to improve the material distribution and ensure that
the output was easy to be printed [7]. This base design allows
room for improvements and attachments.
During this training, the 3D design will be done with the
Autodesk Tinkercad online tool (Fig. 2). Tinkercad is an easy-
to-use 3D CAD design tool that allows to quickly turn ideas
into CAD models that could be shared or 3D printed [8].
Due to the low-cost nature of the course, the chosen
fabrication method of the different elements was 3D Printing in
a low-cost FDM printer (Ender3, Creality3D) using a common
biodegradable polyester (PLA) which is the most widely used
filament material in 3D printing and has the advantage of being
created from renewable raw materials.
Fig. 2. TinkerCad - Example of the 3D Design interface
B. Electronics and Programming
Tinkercad can also be used to simulate different electronic
circuits as seen in Fig. 3. Different basic electronic concepts
were introduced with this tool as hands-on experiments where
students were encouraged for instance to observe the effects
of resistance, current, voltage, and polarity in their simulated
circuits.
The advantage of using simulation for hands-on electronics
learning is that there is always room for errors without
physically damaging components or students getting injured.
Furthermore, simulation is a great tool to easily and inexpen-
sively project and test more complex circuits.
Fig. 3. TinkerCad - Simulation of electronic circuits.
Apart from just simulating discrete component circuits,
the Tinkercad electronics simulation tool also provides pro-
grammable components (specifically a virtual Arduino mi-
crocontroller), which can be programmed both in a visual
language or in a text-based Arduino-like language, which can
be observed in Fig. 4.
The advantage of Tinkercad as a microcontroller program-
ming tool is that a beginner user can start his/her programming
journey by just focusing on the desired interaction between
sensors and actuators. For this, the user just needs to drag
and drop code-blocks sequentially without having to bother
with the hassle of syntax. Tinkercad automatically generates
Arduino-like text code from the developed code-blocks algo-
rithms which can be run in physical Arduino microcontrollers,
being a real advantage for more experienced users.
Fig. 4. Developed code blocks algorithm and automatically generated Arduino
text-code from Tinkercad
Authorized licensed use limited to: b-on: Universidade de Coimbra. Downloaded on May 20,2021 at 13:52:30 UTC from IEEE Xplore. Restrictions apply.

C. From Innovation to Entrepreneurship
Inspired by iSTEME methodology, it is important to conduct
students from their ideas into potential projects and business
endeavors. By organizing this course around a clear path
from the innovation going towards entrepreneurship, it allowed
students to learn Science, Technology, Engineering, and Math
while keeping their focus on business opportunities. Moreover,
such an approach allows the students to define their own path,
allowing them to customize their learning experience.
III. COURSE
As introduced in the previous section, the course was
the structure in 12 steps (weekly meetings), going from the
introduction of the topic, until the design and implementation
of a project.
The overall program starts by introducing Robotics, dis-
cussing the different components of a robot, and the connec-
tion between the hardware, software, and the actions which it
will perform.
The program could be summarized into:
1) Introduction to Robotics
2) Inspiration and Project
3) Sensors and Actuators
4) TinkerCad and 3D design
5) Circuit Design and Simulation
6) Programming: Blocks and Code
7) Assembly the components
8) 3D Printing
9) Electronics Test
10) Programming the Robot
11) Testing the Robot
12) Project Presentation
For the implementation, several components were selected
and divided into three categories ’See’, ’Think’ and ’Act’ to
simplify the students’ understanding (Fig. 5).
The ’Think’ part is done by the Arduino Romeo mini
BLE (which includes low power Bluetooth and a dual-motor
controller), and the ’Act’ comes mostly from the two DC
Motors, used for the robot motion. The gray-scale sensor (Line
Detection) and the ultrasonic sensor (Distance Measurement)
Fig. 5. What a robot is? (’O que ´
e um Robˆ
o’) - Slide Example (in Portuguese)
- See, Think, Act (’Ver’, ’Pensar’, ’Agir’).
are the sources of information, being, therefore, the parts of
the robot used to ’See’.
Those sensors allow not only the robot to follow a line,
but also be able to detect and avoid obstacles or solve mazes.
Additionally, a couple of LED lights and one buzzer were
added.
A. Idealization
Once the students understand the concepts, they can imagine
their robots and plan their innovation, and look for future
business opportunities. Questions such as ’What should the
robot do?’, ’What sensors should it have?’, ’How many people
would benefit from such design?’ were combined with their
own personal interests and answers to other questions such as
’What do I want to build?’, ’Why do I want to build a robot?’.
Depending on their objectives and previous knowledge, they
focus on specific components of the course, for example, a
student might prefer to program rather than to design.
B. Design
In Fig. 6, the model of the Robotics Base is seen being
altered in TinkerCad. The shape and size can be easily
modified by students, for instance, they can add holes or other
connections to sensors.
Fig. 6. Design to support the Motors.
To understand the basic concepts, different parts of the
robot were designed in CAD software. Due to the simplicity,
the robot wheels were selected as the first practical exercise.
Students were then invited to submit their designs to be 3D
printed and delivered to them, as shown in Figure 7.
Fig. 7. Some wheel designs proposed by the students.
Although it would never be possible to provide a 3D printer
for each student, they were taught basic concepts of FDM-
based fabrication such as 3D printer control, dimensional
Authorized licensed use limited to: b-on: Universidade de Coimbra. Downloaded on May 20,2021 at 13:52:30 UTC from IEEE Xplore. Restrictions apply.

accuracy, and the need for tolerances when designing 3D
printable parts. The participants were then encouraged to
share the 3D STL files of any parts they would want to add
to their robots, which were fabricated using PLA filament.
Figure 8 shows the 3D printed wheels designed by some of
the participants.
Fig. 8. 3D printed wheels designed by some of the participants using FDM
technology and PLA filament
C. Circuits
The main goal of teaching electronics concepts to the young
students was not for them to learn the mathematical formulas
that rule the phenomenon but instead to sharpen their intuition
on the physical processes that rule the circuits.
This example (Fig. 9) was implemented by the students to
understand how electric current directly affects the brightness
of a LED (or the velocity of a DC motor). Furthermore, each
circuit that the students designed allowed them to come in
contact with different components (LEDs, resistors, poten-
tiometers, photoresistors, switches, etc.) while seeing how they
influence the circuit: None of the students will ever forget
that a LED needs a series resistor connected to it to limit the
amount current passing through it, otherwise it may fail, as
they observed in the simulation.
Fig. 9. Left: Different states of a LED (turned off, burnt, intense glowing
and week glowing). Right: Using a photo-resistor to control the velocity of a
motor through light intensity.
D. Programming
Teaching programming in hardware and problem-oriented
way instead of directly teaching syntax allowed to keep the
engagement of all participants during the modules. Instead of
directly presenting all the syntax, each command is presented
as soon as the students feel the need for new tools (e.g.
Scratch [9]).
After learning how to interface a Sonar sensor with Arduino
and how to read distances, the students were challenged to
program a car-parking sensor: a device that would light up
a LED whenever the distance measured by the sensor was
inferior to a given threshold. With such a challenge, they were
lead to feeling the need for a command that could evaluate the
veracity or falsity of a condition. This was the moment where
if/else statements were introduced. Results of this challenge,
from one of the participants, can be observed in Fig. 10.
Fig. 10. The car-parking sensor challenge to introduce if/else statements. Top:
Tinkercad simulation of the circuit and physical implementation. Bottom: De-
veloped block-based algorithm with if/else statement in Scratch (Portuguese:
if [read ultrasound in cm] <20, then [LED - HIGH] else [LED - LOW] ).
IV. RES ULTS
For simplicity, and since these classes served as an initiation
to a wide range of iSTEME topics, it was not expected that
students would have the time to design and print their own full
robot. To expedite the process of having a working robot, each
student was provided with a basic pre-designed 3D printed
robotic base, which was aimed at being modular and easily
expandable.
They were also provided with a couple of sensors (one ultra-
sonic distance sensor and 2 lines following sensors) that would
enable them to develop basic interaction between their robot
and the physical world and a microcontroller (Rome BLE
mini, which allied to its low form-factor, has an integrated
motor controller and BLE capabilities), as well as a battery
holder. Starting from this basic layout (Fig. 11), students were
encouraged to expand their robot with more sensors and to
design and 3D print hardware fixtures that would suit their
project goals.
Authorized licensed use limited to: b-on: Universidade de Coimbra. Downloaded on May 20,2021 at 13:52:30 UTC from IEEE Xplore. Restrictions apply.

Fig. 11. The basic robotics base that was provided to all students
In Figure 12, a robot of one of the students can be observed.
This student created his own non-conventional wheels, as well
as designed and 3D printed a mount for 2 sonars, aiming at
being able to develop a wall following algorithm for maze
navigation. Other students created fixtures to hold buttons,
LEDs or buzzers, or even created basic line following robots
aiming at participating in robotic competitions in the future.
Fig. 12. Top: One of the robots developed by a student, with custom wheels
and a fixture for 2 sonar sensors with the goal of creating a robot that could
navigate a wall-maze. Bottom: The sensor mount designed in Tinkercad
V. ANA LYSIS
When conducting the pilot training with 10 students (8-16
years old), the design time, to design a simple structure (that
could be printed or cut using a laser) to hold two motors,
four light sensors, and one ultrasonic sensor, it would take a
few hours for a regular student to execute. Even with initial
planning and drawing on paper, just the implementation of the
design in CAD software would need a long time.
The course was hosted using Zoom and successfully gen-
erated multiple prototypes, as displayed in the Figure 13.
Fig. 13. Example of a remote class in zoom (blurred for privacy)
To ensure that the cycle (from innovation to Entrepreneur-
ship) could be easily completed, a survey of the students’
interest was performed at the end of the course (Fig. 14) with
the results indicating that only a few students are considering
from now a career in robotics, but most of them don’t discard
the possibility to continue the project, and one day creating a
company.
I am interested in... No Maybe Yes
Studying Robotics 2 5 3
Continuing the Project 2 3 5
Creating a company 2 2 6
Fig. 14. Survey on Future Objectives
During the course, different surveys were conducted to
evaluate different parameters, such as the impact of the topic
introduced and the current level of motivation. It is interesting
to observe the evolution (Fig. 15), where the motivation
decreased a little with the theoretical concepts and quickly
improved once the robot starts moving (week 7). When the
students were inquired on the main reason for the jump in the
motivation levels, they claim that it took too long for the robot
to start moving, and they wished they could have it moving
earlier on the course.
When discussing the impact of the online course on their
learning, only 30% of the students said that they would rather
do the course in person, while the majority was happy with
the online alternative.
Authorized licensed use limited to: b-on: Universidade de Coimbra. Downloaded on May 20,2021 at 13:52:30 UTC from IEEE Xplore. Restrictions apply.

1 2 3 4 5 6 7 8 9 10 11 12
40
60
80
100
Session Number[#]
Motivation [%]
Fig. 15. Motivation levels over the course duration
VI. CONCLUSION AND FU TU RE WORK
This 12 weeks course was idealized as a tool to help students
of different ages and backgrounds to develop a wide range of
STEM-related skills such as 3D design, digital manufacturing,
development of electromechanical systems and programming,
allied to innovation and entrepreneurship components. All
these concepts were wrapped up in the versatility of mobile
robotics, proving once again the versatility of such a subject.
Furthermore, the course was created and delivered in such
a way that, during the COVID-19 Pandemic, all participants
would be safe since the course was fully remote. In this work,
we discussed some of the challenges and opportunities that
remote learning offers and how our course is ideal to keep
developing young minds with new knowledge and state of
the art multi-disciplinary challenges offered by educational
robotics.
Contrary to what happens with most educational robotics
courses, our goal is to challenge the young participants to
create their own projects from scratch. Instead of just lecturing
them theoretical knowledge on electronics or programming
languages’ syntax, we encourage them to look for answers
in a problem-oriented way through challenges that they need
to solve. Furthermore, instead of just giving them a limited
robotic base, we propose teaching them how to design and fab-
ricate their own hardware, how to project their own electronics,
and how to create innovative projects. All these factors are
assets that will help train the next generation of entrepreneurs.
From our pilot training with 10 students, our remote mobile-
robotics course proved to have a real advantage in motivating
our students to Engineer and Entrepreneurship careers, with
most of them wanting this course to continue with more
detailed modules on each of the covered topics.
Due to the low-cost characteristics of this course, it proves
ideal to be implemented in schools as a tool for innovative
education for engineering, even more in the current times were
remote solutions and activities must exist to keep students
engaged and occupied by providing this course and offer this
kit (Fig. 16).
A web version is being implemented towards the user ex-
perience automation, with the objective to enlarge the amount
of users. Moreover, it is important to start moving the robot
earlier in the course.
ACKNOWLEDGMENT
The project was developed with the support of ADoT -
Associac¸˜
ao Desenvolver o Talento - which accepted to run the
pilot program as well provide access to survey their members.
REFERENCES
[1] V. Uskov, J. P. Bakken, L. Aluri, R. Rachakonda, N. Rayala, and
M. Uskova, “Smart pedagogy: Innovative teaching and learning strategies
in engineering education,” in 2018 IEEE World Engineering Education
Conference (EDUNINE), pp. 1–6, 2018.
[2] Hsien-Tang Lin and Tzai-Hsing Kuo, “Teaching programming technique
with edutainment robot construction,” in 2010 2nd International Confer-
ence on Education Technology and Computer, vol. 3, pp. V3–226–V3–
229, 2010.
[3] T. Caldeira and J. Montalvo, “Automated design platform for educational
robots,” in 2019 International Conference on Innovations in Engineering
Education - 2019 Advances in Science and Engineering Technology
(ASET) International Conferences (IEE-2019 ASET), (Dubai, United Arab
Emirates), Mar. 2019.
[4] D. D. e. a. S. Shetty, C. Shilpa, “Academic crisis during covid 19: Online
classes, a panacea for imminent doctors,” Indian J Otolaryngol Head Neck
Surg, 2020.
[5] J. Crawford and K. e. a. Butler-Henderson, “Covid-19: 20 countries’
higher education intra-period digital pedagogy responses,” Journal of
Applied Learning & Teaching, vol. 3, no. 1, pp. 9–28, 2020.
[6] J. Montalvo, “Digital fabrication as an opportunity for social transfor-
mation in cali colombia,” Proceedings from the Fab13 Research Stream,
2017.
[7] J. Montalvo and T. Caldeira, “Smart manufacturing of 3D printed robots,”
in 2019 International Conference on Advanced Materials, Design and
Manufacturing - 2019 Advances in Science and Engineering Technology
(ASET) International Conferences (AMDM-2019 ASET), (Dubai, United
Arab Emirates), Mar. 2019.
[8] Autodesk, “Tinkercad - free online collection of software tools.”
[9] M. M. Lab, “Scratch is a project of the lifelong kindergarten group at the
mit media lab.”
Fig. 16. Additional photos of the robot (front and side views)
Authorized licensed use limited to: b-on: Universidade de Coimbra. Downloaded on May 20,2021 at 13:52:30 UTC from IEEE Xplore. Restrictions apply.