Senior Design Projects - Summer 2025 Showcase

This project introduces a novel radio frequency (RF)-based object detection and identification system intended to counter the shortcomings of traditional computer vision systems in environments with poor visibility, occlusions, or adverse weather. The system utilizes the ADALM-PLUTO software-defined radio (SDR) to transmit RF signals and scan the reflected signals from nearby objects. These measurements are used to calculate Radar cross-section (RCS) values, which are unique to each object based on size, shape, and material characteristics. These values are then input into machine learning algorithms to classify and identify objects in real-time accurately. The system is designed to be used in the ISM band, compliant with regulatory needs, and enjoys the advantage of a compact, portable package for ease of deployment. Its primary design drivers are precision, flexibility to support multiple object types, ease of use through a simple interface and setup, and reliability of operation in deployment conditions representative of actual deployment. Its target markets include autonomous vehicles, defense and security networks, smart factories, and city infrastructures. A thorough feasibility study evaluated technical, resource, economic, schedule, legal, cultural, and marketing factors. The result was a high overall feasibility score, confirming the project's viability. Electromagnetic interference, low microcontroller resource availability, and SDR integration challenges were identified and addressed with corresponding mitigation methods. About traditional optical and infrared systems, the RF-based solution is advantageous in terms of being able to operate in the absence of line-of-sight and obscured or low-light conditions. These capabilities make it a suitable replacement in cases where visual systems cannot perform. The system generally presents a cost-effective, reliable, and scalable solution for future object detection and classification technologies that satisfy the demands of today's automation-dependent industries and environmental adaptability. 

High-rise window cleaning involves significant safety risks, as workers must perform tasks while suspended at great heights. The reliance on manual labor makes the process slow, costly, and susceptible to human error. Frequent accidents and equipment failures highlight the urgent need for safer alternatives. Developing an autonomous cleaning system would minimize these risks and streamline maintenance operations. 

StudyBuddy is a new and innovative web application designed to facilitate effective academic collaboration among students. Students have common difficulties when preparing to succeed academically, or form study groups, that’s why StudyBuddy makes use of an advanced matching algorithm. This algorithm intelligently pairs students based on their coursework, interests, availability, and individual study preferences. The platform features seamless integration with real-time communication capabilities, like texting, scheduling tools, built-in AI tutor, and robust security measures to protect user privacy. Developed using modern technologies including React, Node.js, and Firebase, StudyBuddy provides a scalable and efficient platform with a user-friendly environment tailored to enhance academic productivity and foster meaningful social interactions. The project indicates a commitment to sustainability through optimized code efficiency and effective resource use. By addressing both technical and social challenges of the development, StudyBuddy offers a modern solution that supports academic success through intelligent connectivity. 

Spoilers is a platform for discovering, discussing, and reviewing movies and TV shows. Blending the best of a movie database and social media, it allows users to interact through posts, reviews, and spoiler-free conversations. Powered by multiple APIs, Spoilers integrates trusted sources like TMDB to provide rich metadata and discussions. Designed for an immersive, user-friendly experience, the platform features a modern interface with optional dark mode for comfortable viewing. Spoiler detection is also used, with the help of OpenAI, to ensure consistent user engagement. Users can dive into TV and movie discussions, save content for later through favorites and watchlists, and enjoy a spoiler-protected environment. Users can add spoiler tags to posts, while AI moderation automatically detects and manages unmarked spoilers, keeping discussions safe and enjoyable. With intuitive navigation and a vibrant community, Spoilers makes it easier than ever for fans to connect and explore entertainment without the worry of unwanted spoilers.