=Paper=
{{Paper
|id=Vol-1450/paper3
|storemode=property
|title=A Maker Approach to Computer Science Education: Lessons Learned from a First-Year University Course
|pdfUrl=https://ceur-ws.org/Vol-1450/paper3.pdf
|volume=Vol-1450
|dblpUrl=https://dblp.org/rec/conf/iwec/Svanaes15
}}
==A Maker Approach to Computer Science Education: Lessons Learned from a First-Year University Course==
https://ceur-ws.org/Vol-1450/paper3.pdf
A Maker Approach to Computer Science Education:
Lessons Learned from a First-Year University Course
Dag Svanæs
Department of Computer and Information Science,
NTNU
NO-7491 Trondheim, Norway
dags@idi.ntnu.no
Abstract. We report from a one-semester introductory course for first-year
computer science students where Arduino, robot programming and app
development with Processing was used to foster engagement and creativity. The
main learning objective for the students was to learn basic hardware and
software skills, while at the same time motivating for further computer science
courses. The course had a total of 250 university students with two teachers and
ten teaching assistants. The major challenges were related to creating exercises,
educational material and a physical work environment for the students that
allowed for creativity in the spirit of the maker culture. We learned that much
effort must be placed on making a high number of well-documented small and
complete examples that the students can use as a starting point in their projects.
We also learned that the teaching assistants should themselves have spent much
time experimenting with the actual technology.
Keywords: Maker culture, Arduino, computer science education, robots.
1 Introduction
The computer science education at the Norwegian University of Science and
Technology (NTNU) has traditionally been fairly theoretical in the first two years,
with a focus on basic courses in mathematics and programming and little time for the
students to explore technology or apply the theory on real-life problems before later in
the study. This has its roots in the German engineering education of the early 1900s,
with its emphasis on a natural science approach to engineering. The earliest
engineering curriculums at NTNU were written in the 1910s and 20s by professors
who had their training in this tradition, and the wisdom of the "theory first" approach
to engineering education has not been challenged until recently.
In [1] Felder contrast this traditional deductive approach to engineering education
with the integrated view:
Copyright © 2015 for the individual papers by the papers' authors. Copying permitted only for private and academic purposes.
This volume is published and copyrighted by its editors.
Make2Learn 2015 workshop at ICEC’15, September 29, 2015, Trondheim, Norway.
15
• "Deductive (Fundamentals --> Applications): Begin the first year with
basic mathematics and science, teach “engineering science” in Years 2 and 3,
and get to realistic engineering problems and engineering practice in the
capstone course.
• Integrated: Introduce engineering problems and projects starting in Year 1,
and bring in the math and science (and communication and economics and
ethics) in the context of the problems and projects." (Ibid, p. 3)
Over the last three years the computer science curriculum at NTNU has been
through a major revision, aiming for a more integrated model. One of the major
motivations for making changes has been an increasing dropout rate in the first two
years of the study, and a general dissatisfaction among the students with the strong
emphasis on theory early in the study.
One of the changes implemented in the new curriculum is a new project-based
"programming laboratory" course in the second semester of the first year. The main
learning objective of this course is for the students to learn basic computer hardware
and software skills, while at the same time motivating for further computer science
courses. The course was run for the first time this year, and we made use of Arduino
[2], robot programming [4], and app development with Processing [3] to foster
engagement and creativity. The course was to a large extent inspired by the maker
culture.
We will here present the structure of the course and some preliminary findings.
2 Course Structure
The course was named "Programming laboratory 1" to signal that it is a hands-on
course. It is followed by a course "Programming laboratory 2" in the following
semester that builds on this course. The course is mandatory for all 250 first-year
computer science students at NTNU, and runs for the full 14 weeks of their second
semester. The students participating in the course have learned basic programming
principles in a first-semester course, but their experience with programming is
limited.
Two teachers and ten teaching assistants were allocated to the course this year.
The course was organized as teamwork in the lab, combined with one two-hour
lecture per week. The course is 7.5 ECTS, corresponding to 25% of their teaching
load. We consequently expected them to use at least one full day on the course per
week, although each team only had access to teaching assistants four hours per week.
The course has no exam, and the grading was passed/not passed (P/NP) to put less
pressure on the students. The course had four mandatory exercises throughout the
semester that all had to be successfully completed to pass.
16
3 The Exercises
The two first exercises were individual, and two last were done in teams of five
students. Starting out with two individual exercises was done to ensure that all
students got the basic skills in electronics and Arduino programming to enable them
to take active part in the following two team exercises. Our motivation for doing this
was previous experience from programming courses with team exercises where we
often found that the teamwork was taken over by one or two students with experience
in programming, alienating the students with less programming experience.
The four exercises were:
1. Street signal (individual, 2 weeks).
2. Musical instrument (individual, 3 weeks).
3. Robot competition (team, 4 weeks)
4. Xbot - A robot and its world (team, 5 weeks)
3.1 Hardware, Software and Building Material
The learning material for the course consisted of an Arduino starter kit for each
student and an electronics and robot kit for each team. In addition each team got
simple cardboard building material for exercise 4.
The starter kit consisted of an Arduino, a breadboard, and basic sensors and
actuators. The team kit contained additional sensors, actuators, and Bluetooth
communication modules, in addition to a Zumo robot [4] with an Arduino on top.
To enable remote control of the robots from the students´ mobile phones, we made
an Android app that executes Processing code on the mobile and communicates with
the robot through Bluetooth.
3.2 Exercise 1: Arduino Street Signal (individual)
In exercise 1 the students were asked to use their Arduino starter kit to program a
pedestrian crossing with a street signal.
Figure 1. Street signal illustration (left) and Arduino implementation (right).
17
Figure 1 shows to the left the illustration used in the exercise text, and to the right an
implementation of the street signal. In this implementation a servo is added to stop the
pedestrians and a buzzer is added for sound. The feedback from the students was that
they liked the exercise very much because it was easy to understand and because it
was a real-world problem.
3.3 Exercise 2: Arduino Musical Instrument (individual)
In exercise 2 the students used the Arduino starter kit and sensors from the team kit to
program an Arduino musical instrument of their own choice.
Figure 2. Theremin (left) and Arduino musical instrument (right).
Figure 2 shows to the left a picture of a Theremin musical instrument that was
described as an inspiration. To the right is an implementation of an Arduino musical
instrument using a combination of sensors (touch, tilt, magnetic and sound).
The students liked the exercise, in particular that the problem was fairly open and
allowed them to be creative in choice of sensors and behavior. They further
appreciated that we had provided a number of working examples on the course wiki
for the sensors, together with examples of how to program tones and sounds through a
loudspeaker.
3.4 Exercise 3: Zumo Robot Competition (team)
In exercise 3 the students worked in teams and were given the task of programming
their Arduino Zumo robot [4] to compete against another robot in the zumo ring. The
zumo robot is actually an Arduino shield on its head. We stages competitions between
the 55 teams, leading to final rounds in one of the lectures. The two winning teams
won cinema tickets for the team members. We used a slightly modified version of the
international sumo robot competition rules [5]. The aim of the robot competition was
simpoly to push the other robot out of the ring. Each match lasted for a maximum of
two minutes, and the maximum weight was 500 grams.
18
The robot competition ran in two classes: (1) autonomous and (2) remote control. In
the autonomous class the robots were programmed to compete with the other robot
only navigating with sensors. For the remote control class we provided Bluetooth
modules that enabled them to control the robot from an app that they programmed in
Processing for their mobile phones. We had made a reference implementation of the
app that they could use as a starting point for their own robot control app.
Figure 3. An Arduino zumo robot on a scale (left) and two robots competing (right).
The students found this exercise very inspiring and engaging, and the teams spent
much time on perfecting their robots.
3.5 Exercise 4: Xbot - a Robot and its World (team)
Exercise 4 was more open than the previous exercise, challenging the students to be
creative. The challenge was to make the robot into an "Xbot", a robot with some
specific properties of the team´s choice. The teams were given an A0 cardboard
(120x85 cm) that should be the Xbot´s world. In addition to the robot, the teams were
required to use the Arduinos of the team members as stationary "helper bots".
Each team documented its Xbot in a 30 seconds video. The videos were shown in
sequence on the final lecture of the course, and the students scored them using a
modified version of the Kahoot system. The members of the two winning teams got a
gift from a local semiconductor company.
Figure 4. "Firebot Sam" (left) and his world (right).
19
In Figure 4 we see "Firebot Sam", the Xbot of the winning team. The robot was
programmed to put out fires in candles placed on the side of its "world". It
communicated through Bluetooth with helper bots that detected heat with temperature
sensors. It then navigated by line-following to the right candle. A servo on the robot
was user to tilt a metal top to put out the candle.
The students enjoyed working with the exercise, and a number of innovative
Xbots were created. Some teams had problems coordinating the work among the team
members though, leading to some frustration.
3.6 General Observations
One of our major challenges was to create exercises, educational material and a
physical work environment for the students that allowed for creativity in the spirit of
the maker culture. We learned that much effort must be placed on making a high
number of well-documented small and complete examples that the students can use as
a starting point in their projects. The planning and preparation of the exercises was
done in the previous semester and involved developing code examples, setting up a
wiki and buying electronics. Two teaching assistants were hired for this purpose in
addition to approx. 20% of the work time of the two teachers and an engineer.
Although the teaching assistants had a background in programming, we gave
them a crash course in Arduino in the previous semester and encouraged them to
experiment on their own. As a consequence they became an invaluable recourse.
4 Conclusions
The presented course was inspired by the maker culture, and we have made extensive
use of Arduino and the open source maker culture around this product. A combination
of structured and open exercises worked well to teach basic skills, and at the same
time opened up for creativity. The course was well received, and we plan to follow
the same overall structure next year with some minor modifications.
Acknowledgments. Thanks to the other course team members Asbjørn Thomassen,
Terje Røsand, Pia Lindkjølen and Inge E. Halsaunet.
References
1. Felder, Richard M. "Engineering education: A tale of two paradigms." Shaking the
Foundations of Geo-engineering Education (2012): 9-14.
2. Evans, Brian. Beginning Arduino Programming. Apress, 2011.
3. Reas, Casey, and Ben Fry. "Processing: programming for the media arts." AI & SOCIETY
20, no. 4 (2006): 526-538.
4. Zumo Robot for Arduino. https://www.pololu.com/product/2506 (retrieved 9. July 2015).
5. Unified Sumo Robot Rules. http://robogames.net/rules/all-sumo.php (retrieved 9. July
2015).
20