Theme
Energy

Location
C-110

Title
Smart Parking Assist System

Team Members
Parshva Mehta, Aman Patel, Abhiram Vemuri, Rajeev Atla

Advisor(s)
Professor Kristin Dana

Abstract
Universities increasingly struggle with parking due to growing demand, dynamic schedules, and limited staffing. Traditional solutions like automated valet systems are often too expensive, complex, and unreliable. As a result, drivers waste time circling crowded lots, increasing emissions and campus congestion. To address this, we developed a scalable Smart Parking Assist System
using edge-based computer vision and a web app for real-time detection and guidance. A YOLO-based model, trained on public and custom Rutgers CoRE building images, runs on a Raspberry Pi with a high-resolution camera to detect parking occupancy on-device. The backend updates a central database and powers a front-end displaying real-time availability via a parking lot map. This end-to-end system offers a low-cost, deployable solution for university campuses and urban areas.

Poster
[LINK]

Video
[LINK]