Work in Progress: Teaching MATLAB through Authentic Data Collection and Analysis Experiences using self-contained, guided experimental setups with a range of disciplinary themes.

Brian O'Connell

Download Paper | Permalink

Conference: 2022 ASEE Annual Conference & Exposition
Location: Minneapolis, MN
Publication Date: August 23, 2022
Start Date: June 26, 2022
End Date: June 29, 2022
Conference Session: First-Year Programs Division Technical Session 13: Work-in-Progress Postcard Session #2
Page Count: 8
DOI: 10.18260/1-2--41760
Permanent URL: https://strategy.asee.org/41760
Download Count: 244

Request a correction

Paper Authors

biography

Brian O'Connell Northeastern University

visit author page

Brian O'Connell is an Associate teaching professor in the First-Year Engineering program at Northeastern University. His undergraduate degree in Mechanical Engineering came from the University of Massachusetts at Amherst in 2006. He then worked for Kollmorgen Electro/Optical as a mechanical engineer developing periscopes and optronic masts. In 2011, he returned to academia at Tufts University, earning his MS and Ph.D. in Mechanical Engineering for his work with low-cost educational technologies and his development and use of technologies to aid usage tracking in makerspaces to examine them as interactive learning environments. He joined Northeastern in 2017. As well as teaching first-year engineering courses, he continues to design new technologies and curricula for use in his own classroom as well as for K-12 engineering education outreach.

visit author page

Download Paper | Permalink

Abstract

Engineering students primarily use MATLAB to support their research and lab-course experimentation. It was designed and developed for the collection, handling, and analysis of large data sets. MATLAB curriculum at the First-year level does not always take this typical use into consideration, focusing on learning programming concepts and MATLAB commands without regard for that context. This grant and project is an attempt to address that by incorporating experimental data collection and analysis into those lessons.

Four experimental setups were developed as part of this project. Each involves an Arduino-based hardware component that enables the student to collect data. The hardware makes use of open-source sensors, and readily available commercial-off-the-shelf parts as well as rapid prototyping typically available in academic settings to achieve an average cost of less than

Citation
Format

O'Connell, B. (2022, August), Work in Progress: Teaching MATLAB through Authentic Data Collection and Analysis Experiences using self-contained, guided experimental setups with a range of disciplinary themes. Paper presented at 2022 ASEE Annual Conference & Exposition, Minneapolis, MN. 10.18260/1-2--41760

TY  - CPAPER
AB  - Engineering students primarily use MATLAB to support their research and lab-course experimentation. It was designed and developed for the collection, handling, and analysis of large data sets. MATLAB curriculum at the First-year level does not always take this typical use into consideration, focusing on learning programming concepts and MATLAB commands without regard for that context. This grant and project is an attempt to address that by incorporating experimental data collection and analysis into those lessons. 

Four experimental setups were developed as part of this project. Each involves an Arduino-based hardware component that enables the student to collect data. The hardware makes use of open-source sensors, and readily available commercial-off-the-shelf parts as well as rapid prototyping typically available in academic settings to achieve an average cost of less than <$50 per unit. They are designed to be rugged and compact to allow for distribution to students as take-home components of their assignments. A provided MATLAB class definition allows students to construct an object in their program for each experiment. This allows for single command interactions with the hardware rather than copy-pasting large portions of example code, allowing students to focus more on the MATLAB lesson elements rather than debugging advanced examples. The class provides callbacks to handle hardware connection, data collection, and database connection. The centralized database provides students a resource for larger datasets, compiled from the experimental runs of all students in the course, in their data analysis. 

Currently, four experiments have been developed. One is an environmental experiment for examining the sun exposure around. This, along with the system clock, allows for the collection of UV light levels, GPS location, and timestamp. After uploading several datapoints and downloading the full class dataset, they can learn about MATLAB visualization options by experimenting with various graphing options, making guided decisions about how best to represent the data. The second electrical engineering experiment in signal processing and analysis. One microcontroller generates an “unknown” signal to be decoded and a second injects randomized noise into the system simulating radio interference, both pass through a resistor array to signal physical interference. Students will be able to experiment with signal filtering and iterative analysis as well as comparison to other processed signals uploaded by other students to try and decode the “unknown” signal. The third is a bio-engineering-inspired human response experiment. This involves a hand-eye coordination experiment that measures reaction time to microseconds and students will be asked to collect several data points from a range of individuals. They will then analyze and graph the data while also filtering out invalid responses, some of which will likely naturally occur from experimentation while others will be randomly generated for discovery. The last is a mechanical/civil engineering inspired experiment for static bend tests. The setup is a desktop static bend setup for breaking small samples, approximately the size of popsicle sticks. This is intended to involve all elements; noisy sensor signals, potentially invalid test runs, and data visualization to name a few.    


Development of the hardware and software occurred during summer 2021. Beta testing will begin in Spring 2022 courses with refinement and redesign during the summer session. Formal implementation will occur in a fall 2022 honors course. Direct comparison will be unavailable due to the class and lack of feasibility of assigning the new and old curriculum simultaneously. Therefore several evaluations of efficacy will take place in several phases across a range of avenues. Pre and post-testing will be utilized for initial comparisons among that cohort as well as comparison to past classes. Students will be recruited for follow-up interviews after their lab courses about their comfort with MATLAB and perceived capabilities. Past students will be recruited to pilot interview/survey protocols as well as collect that data from students who experienced the previous curriculum. Students from other sections of the Fall 2022 first-year engineering courses that use dissimilar MATLAB curriculums will be recruited as another comparison. System capabilities and quality will be evaluated based on the performance of the hardware, particularly regarding its portability and reliability, the quality and reliability of the database, and reliability of the software as well as error reporting and other usability feedback from the students. 
AU  - Brian O'Connell
CY  - Minneapolis, MN
DA  - 2022/08/23
PB  - ASEE Conferences
TI  - Work in Progress: Teaching MATLAB through Authentic Data Collection and Analysis Experiences using self-contained, guided experimental setups with a range of disciplinary themes.
UR  - https://strategy.asee.org/41760
DO  - 10.18260/1-2--41760
ER  -