Students from the NC State University Department of Mechanical and Aerospace Engineering debuted Friday their projects on new applications of Artificial Intelligence (AI) that they developed in a first-of-its-kind AI course offered at the department.
The brand new course, MAE 495 – 011 and MAE 589 – 011: Artificial Intelligence for Engineering Applications, is taught by MAE Assistant Research Professor Veeraraghava Raju Hasti, who joined the department in 2023 and specializes in interdisciplinary research at the intersection of engineering, computer science and economics – focusing on the development of transformational digital tools and technologies using physics-based and data-driven approaches.
“Our MAE students have harnessed the power of AI, showcasing innovative applications in engineering through their short-term course projects,” Hasti said about this semester’s projects. He continued that the cutting-edge work these students are doing in the emerging field of AI applications in engineering is comparable to that of top AI research institutions like the University of Michigan and Harvard.
Power of Ignited Minds
Friday’s Poster Session, titled “Power of Ignited Minds: Unveiling the Future of AI in Engineering,” allowed students to debut their work to students, faculty and other curious minds and highlight what they believe to be the future of Artificial Intelligence in the field of engineering.
Ten posters debuted at the event, highlighting a number of different innovative applications of AI technology from AI-Based Manufacturing Systems that optimize production and cost reduction, to AI-generated Airfoil shapes for mitigating dust accumulation on Solar panels, and everything in between.
One such project from Shang-Ru Li, one of the students in the cutting-edge AI course, focuses on using AI to detect potential COVID-19 Patients. According to the project’s poster presentation, the process is comprised of three core steps using Artificial or Residual Neural Networks (ResNet) to execute classification tasks on Chest X-ray images.
“A general convolutional neural network model is first utilized to ensure the feasibility of performing image classification tasks on a given dataset with three labels,” The poster’s abstract states. “ResNet is later used on the same dataset with two categories only, and the training weights are saved for the following transfer learning purpose. Finally, ResNet with the pre-trained weights is exploited on the ternary classification task. The result proves the usefulness of using transfer learning on an unseen category.”
Another Student, Martha Crisp, presented at the poster session her research into using similar neural networks to predict air quality index.
“Popular prediction methods including linear and polynomial regression have been used to forecast an abundance of different datasets, but the nature of pollution levels reflected in Air Quality Index (AQI) datasets are too complex to be accurately modeled by these methods,” Crisp’s abstract states. “This research uses the Long Short-Term Memory (LSTM) model, a variation of the Recurrent Neural Network, to predict future AQI levels in Delhi, India. Upon evaluation, the R squared value of the model resulted in a 0.954 value. This close prediction is visualized against when the true and predicted values are graphed.”
Hasti plans to continue to expand the AI-curriculum at MAE, and further assist students in making projects liek these a reality so that researchers can further make use of the power of Artificial Intelligence.