Paper ID #6559

Microcontrollers for Mechanical Engineers: From Assembly Language to

Controller Implementation
Mr. Noah Salzman, Purdue University, West Lafayette

Noah Salzman is a graduate student in engineering education at Purdue University. He received his B.S.
in engineering from Swarthmore College, his M.Ed. in secondary science education from University of
Massachusetts, Amherst, and his M.S. in Mechanical Engineering from Purdue University. He has worked
as an engineer and has taught science, technology, engineering, and mathematics at the high school level.
His research focuses on the intersection of pre-college and undergraduate engineering programs.

Dr. Peter H. Meckl, Purdue University School of Mechanical Engineering

Peter H. Meckl is a Professor in the School of Mechanical Engineering, where he has served since 1988.
Peter obtained his BSME from Northwestern University and MSME and Ph.D. degrees from MIT. His
research interests are primarily in dynamics and control of machines, with emphasis on vibration reduc-
tion, motion control, and engine diagnostics. His teaching responsibilities include courses in systems
modeling, measurement systems, and control. In addition, he teaches a course entitled Technology and
Values, which introduces students to the social and environmental impacts of technology through a series
of readings and discussions. Peter received the Ruth and Joel Spira Award for outstanding teaching in
2000. He spent a semester in the Institute of Measurement and Control Engineering at the University of
Karlsruhe, Germany, in spring 2005, conducting research and teaching on autonomous vehicles.

American
c Society for Engineering Education, 2013
 Microcontrollers for Mechanical Engineers: From Assembly
Language to Controller Implementation
From automobiles to robotics to process automation, the electronic control of mechanical
systems is ubiquitous in modern control engineering. These controls are often implemented using
microcontrollers, making understanding and learning how to work with microcontrollers an
important part of the mechanical engineering curriculum related to control theory. For over 25
years1,2, a course on utilizing microcontrollers to control electromechanical systems has existed
at The University.

Many different microcontroller platforms have been utilized in the course, including the Intel
8085, the Intel 80188, and the Motorola 68HC12. Having used the 68HC12 platform for the past
10 years, and with various hardware and software problems leading to student frustration, the
decision was made to upgrade to a new hardware and software platform. After evaluating tools
from Microchip, Texas Instruments, and ARM the decision was made to use the
STM32VLDISCOVERY board utilizing a Cortex-M3 microcontroller from ST
Microelectronics3 along with the Keil4 development environment from ARM. To facilitate easy
connections to existing laboratory equipment, a custom enclosure and interface circuitry were
developed. This provides easy access to digital and analog I/O and a serial interface to
communicate with a terminal program running on a PC.

These tools were chosen for several reasons. Keil makes a free version of its development tools
available for download, with the only major limitation being the size of the code able to be
compiled. By choosing a free set of development tools, students were able to install the tools on
their personal computers to work on outside of the lab and utilize for their own projects after
completing the course. The STM32VLDiscovery board is readily available and inexpensive
(<$15) allowing students to have their own hardware to work with as well. With over 20 billion
ARM based chips shipped to date5, it is one of the most commonly utilized microcontroller
platforms currently available and is utilized across a variety of industries.

Course Description

The current course focuses on the use of microcontrollers for control system implementation.
The lectures and laboratory assignments have been designed to satisfy the following three
objectives: (1) provide a basic knowledge of microprocessors, their architecture, and their
programming; (2) provide the tools for interfacing microprocessors with peripheral devices,
including digital I/O, analog I/O, and serial communication; and (3) provide experiences in
utilizing microprocessors for real-time measurement and control. Although this course is a
graduate course, upper-division undergraduates are encouraged to participate. While prior
experience with programming and control theory are desirable, the course does not have any
formal prerequisites.

Since mechanical engineering curricula typically do not include courses on assembly language
programming and microcontroller interfacing, this course is designed to introduce these topics to
mechanical engineers. In this way, they can develop their expertise for implementing controllers
toward the end of the semester, without having to take several courses that cover these topics in
depth in the electrical engineering department. The ultimate goal is to have the students control
actual electromechanical systems with microcontrollers. Appendix A contains the sequence of
topics included in the course and accompanying laboratory activities.

Lectures have been designed to introduce students to various basic concepts that are crucial to
programming and interfacing with microprocessors. These topics include microprocessor
architecture, assembly language programming, digital I/O, serial communication, interrupts, and
analog-to-digital conversion. In addition, since not all students have the same level of control
background, some lectures are devoted to fundamentals of classical control design, including
controller gain selection, integrator windup, and digital implementation. Since actuators and
sensors represent a vital connection between the microprocessor controller and the system to be
controlled, several lectures are devoted to a discussion of stepper motors and drivers, incremental
optical encoders for position measurement, and issues of digital sampling and aliasing.

Numerous lab exercises complement the lectures, culminating in the design and implementation
of real-time controllers for a variety of electromechanical systems. The labs begin with a basic
introduction to the development environment and assembly language programming, followed by
working with the various peripherals of the microcontroller. These include digital I/O, serial I/O,
timer and external interrupts, and analog to digital and digital to analog conversion. The first
third of the exercises focus exclusively on assembly programming, followed by an exercise in
mixed C and assembly programming, with the rest of the course including the controller
implementation done using the C programming language.

The last six weeks of the course are dedicated to a series of two-week projects implementing
controllers on several different electromechanical systems. All of the students design a controller
for a common apparatus (most recently a modified version of the Aeropendulum6), and then
choose two additional setups. Options include an inverted pendulum, a heating and cooling
apparatus, speed control of a DC motor, a cruise control for an air-powered engine, control of a
refrigeration system, and an active mass damper. Each of these projects involves system
characterization, modeling in Simulink, and developing an appropriately-tuned controller to
achieve a set of performance specifications.

Given the somewhat unorthodox use of assembly programming in a controls course for
mechanical engineers and the recent update to the microcontroller platform, we set out to answer
the following research questions:

1) What aspects of the course, both content and pedagogy, did the students find the most
valuable?
2) How does learning assembly language programming help students to understand and use
a microcontroller?

Methods

To help answer these questions, a survey was written to collect three different kinds of
information. Students were asked to provide demographic information on their status (graduate
or undergraduate), major, graduate degree program, and prior experience with coursework in
controls and working with microcontrollers. The second part of the survey consisted of three
Likert-type questions to measure students’ perceptions of the value of the various content areas
presented in the course, their confidence at utilizing skill and knowledge acquired in the course,
and the usefulness of the different learning activities employed in the course. Finally, the
students were asked to provide their views on the usefulness of learning assembly programming,
the most positive aspects of the course and opportunities for improvement via open-response
items. The survey was administered online using the Qualtrics survey software, and descriptive
statistics were generated using Microsoft Excel.

Results

The type of degree being pursued by the respondents and their majors are shown in Tables 1 and
2, respectively. Table 3 shows the prior controls coursework experience, and shows that the
majority of respondents have taken at least one course in controls prior to this course. However,
3 of the respondents had no prior controls experience. Five of the respondents had worked with a
variety of microcontrollers prior to taking this course, as shown in Table 4.

Table 1: Type of degree of respondents (n=13)

Degree Type Count
MS Non-thesis 5
MS Thesis 4
Ph.D. 2
Undergraduate 2

Table 2: Major of respondents (n=13)

Major Count
Mechanical Engineering 9
Mechanical Engineering Technology 1
Civil Engineering 1
No Response 2

Table 3: Prior controls coursework reported by respondents (n=13)

Number of Prior Control Courses Count
None 3
1 4
2 3
3 or more 3

Table 4: Prior microcontrollers used by respondents (n=5)

Microcontroller Count
Arduino 2
AVR 2
Texas Instruments MSP430 1
Intel 8051 1
Cypress PSOC 1
PIC 1
Figure 1 indicates the respondents’ perceptions of the helpfulness of various content areas
presented in the course. With the exception of microcontroller architecture, which is not a strong
focus area of the course, the respondents indicated that areas that focused on the low-level
functioning of the microcontroller (Interrupts, Interfacing, and Assembly Programming) were the
most helpful aspects of the course. Least helpful was modeling and simulation, a topic that is a
significant part of the project activities but receives relatively little formal treatment in the
lectures.

100%
90%
80%
70%
60%
50%
1=Not Helpful
40% 2
30%
20% 3
10%
0% 4
Interrupts

Programming

C Programming
Interfacing

Controller Design
Microcontroller

Modeling and
Implementation

5=Very Helpful
Simulation
Architecture
Assembly

Controller

Figure 1: Helpfulness of various content areas reported by respondents (n=13)

Figure 2 shows the respondents’ confidence in various skills learned in the course. The
homework, labs, and projects all have a strong focus on programming, and students reported the
most confidence in their ability to complete tasks related to programming. The students reported
the least confidence in their ability to model a system and design a controller, suggesting that
there is room for improvement in the instruction related to these areas.

100%
90%
80% 1=Not
70%
Confident
60%
50%
2
40%
30% 3
20%
10% 4
0%
C Programming
Generating a Program

Modular Programming

Decomposing a Problem for

Circuit Construction
Assembly Programming

Debugging a Program

Controller Design

Modeling a System

5=Very
Confident
Flowchart

Debugging

Figure 2: Confidence in skills learned during the course reported by respondents (n=13)
Figure 3 shows the helpfulness of various instructional activities reported by the respondents.
Labs, lectures and projects were all considered helpful, with 90% or more of the respondents
rating these a 4 or 5 on a five point scale. Homework was slightly less helpful, probably because
the homework assignments generally involved writing assembly or C code to prepare for that
week’s lab without having access to the lab equipment.

100%

90%

80%

70%
1=Not Helpful
60%
2
50%
3
40%
4
30%
5=Very Helpful
20%

10%

0%
Labs Lectures Projects Homework
Figure 3: Helpfulness of different learning activities reported by respondents (n=13)

Open Response Results

In addition to the demographic and Likert-type questions included on the survey, the students
were invited to share their thoughts via open response questions on a variety of topics. These
included students’ perceptions of the value of learning assembly programming, overall
assessment of the value of the course, and an opportunity to describe challenging aspects of the
course and suggestions for improvement.

When asked how learning assembly programming helped them to understand how the
microcontroller worked and their ability to connect assembly programming to the rest of the
course material, all 12 of the respondents indicated that they felt it was valuable in some way.
They said it “connects more to the working of the microcontroller than C language”, “helps me
to better visualize how microcontrollers process in reality through memory and registers”, and
“assembly coding help me understand the basic operation of a microcontroller.” In addition to
the respondents indicating that learning assembly helped them to generally understand the
operation of the microcontroller, many also indicated specific low-level functions of the
microcontroller that learning assembly helped them to understand, such as “how serial and
digital IO works”, ”how a C-program is decomposed into machine level code”, and “how various
components help in functioning of microcontroller like timers, interrupts, IO etc. and how best
we can utilize them.” Only one respondent felt that the assembly programming was “a little bit
disconnected” from the rest of the course.
Most of the respondents felt that the best part of the course was the ability to apply the material
presented in the lectures the solution of actual control problems in the labs and projects. One
respondent wrote: “The best parts of the work is the interaction of class and lab, we can learn
from lecture and get further idea of intuition from labs. When on lecture, professor tries to make
us better understood the materials…so that we can not only try to understand them from the
powerpoint but also can see the structure of what he tries to describe(sic).” Others appreciated
the labs and projects “because they were able to draw a connection to most of the topics
covered”, and “it enables us to put what we learn in class into practice.” Another respondent
wrote, “I have taken many controls courses…but none of them bridged the gap between the
actual implementation of controls on the physical system. Different components that I have
learnt together seem to be more connected after this course.” Finally, one of the respondents also
appreciated the ability to work with an ARM Cortex-M3 microcontroller, as the “ARM Cortex-
M3 is popular right now.”

The time required to complete the labs and projects was the most frequently described challenge
presented by the course with one student writing “a lot of time needed to prepare for the lab as
well as debugging.” Another student wrote “The most challenging part to me is programming
language. I have learned C++ during my undergraduate courses long ago, and I have forgotten
them totally.” Others found the control theory included in the course challenging, writing:
“Controls systems was new to me and it took a lot of time before I understood it.” Several also
indicated modeling in Simulink was challenging. These results are consistent with those shown
in Figure 1. Finally, several respondents also indicated some frustration dealing with the
hardware and software bugs that accompanied moving to a new microcontroller platform
requesting that we “ensure that all of the supplied code indeed functions properly. The most
frustrations and waste of time stemmed from finding errors in the library code. Once you lose
confidence that the supplied code was functioning properly it was increasingly difficult to
efficiently debug new programs.” Other suggestions for improvement mostly focused on
providing more examples or scaffolding for the content areas that they felt were challenging.

Conclusion

Based on the results of the survey, the respondents felt that this was a useful course and
appreciated the opportunities that it provided to apply control theory to the solution of real-
world, electromechanical control problems. By learning assembly language programming,
students were able to understand the underlying function of the microcontroller, and they felt it
was a useful learning experience. The results of the survey also suggest that as the course exists
right now, there is not enough time spent on modeling and controller design, especially for
students with limited prior controls experience. Based on these results, future versions of the
course will contain more structured assignments and activities to help students become more
confident in these areas.
Bibliography

1. Shoureshi, R. & Meckl, P. H. Control Education in the Era of Microprocessors. in Proceedings of the 1992
American Control Conference 455–460 (1992).
2. Meckl, P. H. Teaching Mechanical Engineers about Embedded Programming. in Proceedings of the ASEE 2003
IL/IN Sectional Conference (2003).
3. STM32VLDISCOVERY - STMicroelectronics. at <http://www.st.com/internet/evalboard/product/250863.jsp>
4. Keil Embedded Development Tools. at <http://www.keil.com/>
5. Company Profile - ARM. at <http://www.arm.com/about/company-profile/index.php>
6. Enikov, E. T. & Campa, G. USB-powered portable experiment for classical control with MATLAB Real-Time
Windows Target. in 119th ASEE Annual Conference and Exposition, June 10, 2012 - June 13, 2012 (American
Society for Engineering Education, 2012).
Appendix A: Course Sequence and Laboratory Activities

Session Topic Laboratory Activity

Introduction to Microprocessors and
Microcontrollers; Number Systems, Binary
1 Arithmetic
2 Digital Logic; Memory and Memory Addressing
Microprocessor Architecture and Internal
3 Operations
PC Familiarization, Software
4 Instruction Set Overview, Addressing Modes Development Environment
5 Assembler Directives; Branching and Looping
6 Subroutines & the Stack, Passing Parameters
Assembly Language Programming,
7 Peripherals, I/O Interfaces Digital I/O
8 Parallel I/O Ports
9 Serial Communications, UART, ASCII Conversion
10 Interrupts
11 Interrupts (cont.); Timers
Keyboard & Console I/O Operations,
12 Mixed-Language Programming Serial Communications
13 Digital-to-Analog Conversion
Interrupts: Application to Frequency
14 Analog-to-Digital Conversion Counting
15 Review of Systems Concepts
16 Review of Classical Controller Design Mixed-Language Programming
17 Midterm Exam
18 PID Control
Digital-to-Analog and Analog-to-Digital
19 Ziegler-Nichols Tuning & other Tuning Algorithms Conversion
20 Discrete Controller Implementation Digital Controller Implementation
21 Sampling, Sampled Data Systems
22 Aliasing, Anti-Aliasing Filters
23 Saturation and Anti-Windup
24 Digital Filtering Electromechanical Control Project I
25 A More Detailed Look at Peripherals
26 Actuators, Stepper Motors
27 Stepper Motors (cont.)
28 Stepper Motor Drivers
29 Optical Encoders Electromechanical Control Project II
30 Microcomputer Peripherals, Bus Standards

Electromechanical Control Projects

Engine Cruise Control Inverted Pendulum Heating/Cooling System
DC Motor Control Adaptive Refrigeration Cycle Active Vibration Control