“Amateur bike riders are typically unaware of how fast their components wear down,” said team member Harris Nesteruk. “We wanted to provide a solution for predictive bike maintenance while keeping the cost down as much as possible, especially given that we primarily sought to target college students with our product.” 

The team’s solution uses an embedded hardware system to predict maintenance needs, ensuring riders aren’t stranded by unexpected failures. Cost-effective, easy to install, and fully on-device, Bike Buddy removes barriers that often keep technology out of reach. 

The six-member computer engineering team includes Aniket Garg, Jason Hsiao, Nesteruk, Leandro Odorici, Keshav Parthasarathy, and Gate Tangchartsiri, and is advised by Professor Linda Milor. 

The Bike Biddy system combines sensors for wheel rotation, temperature, humidity, and brake force with an ESP32-S3 microcontroller. Data is processed using classical machine learning and signal analysis, then sent via Bluetooth to a custom iOS app so cyclists receive actionable insights delivered straight to their phone.

“This project was an application of all the skills we have learned in the past few years at Tech, ranging from embedded systems to mathematical modeling,” Parthasarathy said. “All of us had previously taken ECE4180, and we found that the experience was invaluable throughout the development cycle. Most of us also had prior robotics experience that was valuable for sensor and system integration.”