Jonathan Millam

• Collaborating with a team of 4 students to optimize the performance of an Atmospheric Water Generator.
• Designing a system for remote data collection, monitoring, and control of the device.
• Evaluating and testing approaches to identify the most efficient use cases under varying environmental conditions.

Project 1: Material Classification with Time-of-Flight Sensors

• Developed neural networks in PyTorch achieving 98%+ accuracy in material identification.
• Collected and processed 64,000+ data samples, designing robust testing environments.
• Designed software to extract physical model parameters from ToF data.
• Contributed to a manuscript in progress for submission to IEEE.

Project 2: Motion Signature Extraction (Ongoing)

• Applying signal processing and machine learning techniques to extract unique motion signatures from sensor data.
• Building data pipelines and experimenting with model architectures for robust feature extraction.
• Designing validation methods to evaluate performance under varying real-world conditions.