Download

Download Full Text (593 KB)

Description

The development of neural network curriculum in the Electronics Engineering Technology (EET) program began as project for the Kansas National Space Grant College and Fellowship Program – Opportunities in NASA STEM FY 2020-2024 grant. In this grant, NASA funds supported curriculum development and research in affiliated universities. In the project, faculty and students met to explore online resources and participate in online courses on neural networks that focused on the Keras application programming interface (API).

A portion of the developed curriculum was first used in Spring 2022 in EET 549 Advanced Microcontrollers and EET 745 Advanced Microprocessor Systems and Applications. The topics of regression, classification, convolutional neural networks, and recurrent neural networks were covered. In EET 549, students were assigned a final project in which a line follower robot was programmed using a neural network.

In 2023 and 2024, EET faculty and students became involved in a NASA Rapid Response Research (R3) grant. In this grant, it was proposed that neural-symbolic neural networks could be used to augment physics-based predictions of the aurora borealis with sightings made by citizen scientists. Neural-symbolic neural networks are those that implement logic predicates.

As part of the R3 grant, curriculum was developed for predicate logic and neural-symbolic neural networks. This was used in EET 549 and 745 in Spring 2024. In addition, curriculum on tiny machine learning (TinyML), the implementation of neural networks on memory and speed constrained devices such as microcontrollers, was also introduced.

Department of Primary Author

Automotive and Engineering Technology

Affiliation of Primary Author

Faculty

Publication Date

2025

Neural network curriculum development and student research in Electronics Engineering Technology

Share

COinS