Projects

UAV Swarm-Based Antenna Design for Intelligent Edge (iEDGE)

The proposed UAV swarm-based antenna design integrates seamlessly with Intelligent Edge (iEDGE) systems to enhance wireless communication and sensing capabilities in dynamic and distributed environments. By leveraging a coordinated UAV swarm to form reconfigurable antenna arrays, such as Yagi-Uda configurations, this approach enables high-gain, directional communication tailored to the needs of iEDGE systems. The mobility and adaptability of the UAV swarm allow the iEDGE to dynamically adjust its communication links in response to real-time environmental and operational changes. This project aims to research and develop an efficient, scalable, and cost-effective UAV swarm-based antenna solution to support the high-performance requirements of iEDGE systems in applications like edge computing, autonomous networks, and mission-critical IoT deployments.

Edge Computing with O-RAN for Integrated Sensing and Communication

This project will develop a low-cost, open-source, and reproducible method for integrated sensing and communication (ISAC) leveraging the Open Radio Access Network (O-RAN) architecture, which enables disaggregated RAN deployment and artificial intelligence enhanced network, resource, and service management. The work will proceed from simulations to a working ISAC system, quantifying tradeoffs between sensing and communication performance. The project will design and prototype a real-time disaggregated application (d-App) deployed at the O-RAN edge to demonstrate how localized, real-time processing can dynamically balance resources between sensing and communication services based on current conditions, prioritizing UAV sensing when threats are likely while maximizing communication throughput when the airspace is clear. Initial results will leverage NVIDIA’s Aerial platform to accelerate development and validate the edge computing approach. Experimental data will be collected, labeled, and stored for enabling use in future research.

Edge AI for Unitree Go1 EDU Autonomy: Natural Language to Motion

This project aims to turn plain-English instructions into safe, executable behaviors on a quadruped robot. Students will build a perception-to-action stack where an LLM interprets user commands, an object-detection module grounds those commands in the scene, and a safety controller enforces constraints before publishing motion goals to ROS for the Go1. Students will implement prompt/skill libraries (e.g., “follow the red backpack,” “inspect the window”), integrate vision models for real-time grounding, and design watchdogs for collision avoidance. By the end, participants will demo robust “talk-to-walk” autonomy, voice or text in, verified trajectories out, along with reproducible code, datasets, and a short paper documenting system design, experiments, and ethical considerations for human-robot interaction.

Privacy-Preserving Classroom Engagement Analytics via On-Device Edge Computing

This project investigates the design and evaluation of a privacy-preserving classroom analytics system using intelligent edge computing. The system estimates aggregate student engagement metrics (e.g., head orientation, activity level) by performing all sensing, feature extraction, and inference locally on embedded hardware, without storing or transmitting identifiable data. Multiple technical approaches, including classical computer vision pipelines and lightweight deep learning models, are explored to compare accuracy, latency, power consumption, and interpretability under real-time constraints. The project emphasizes privacy-by-design principles, integrating technical safeguards and ethical analysis alongside system optimization. Experimental evaluation under varied conditions demonstrates how edge-based intelligence can support educational insight while respecting student privacy and resource limitations.

ML-Optimized Optically Transparent Antennas for Intelligent Edge (iEDGE) Connected Vehicles and Smart Cities

This project focuses on the design and optimization of optically transparent antennas to enable seamless wireless connectivity for connected vehicles and smart city technology within Intelligent Edge (iEDGE) environments. Optically transparent antennas can be integrated into windshields, windows, and other transparent urban infrastructure, providing robust communication while preserving aesthetic and functionality. This project will leverage machine learning-based optimization techniques, such as multi-objective particle swarm optimization (MOPSO), to assist antenna designers that balance antenna performance metrics (bandwidth, gain, reflection coefficient, etc.) with the transparency constraints of transparent conductive materials. Using such methods to assist antenna design, researchers can employ more application-specific requirements for iEDGE nodes. The goal is to develop scalable, data-driven methodologies that enhance low-latency wireless links within the smart connected vehicle and smart city IoT fields.

Environment-Aware Mobile Edge Computing

This project enables students to practice cutting-edge edge computing techniques on Jetson edge devices and Hiwonder robotic cars, focusing on how environmental factors like temperature and wind speed affect mobile edge devices. These mobile edge devices experience fluctuating computing capacity due to environmental conditions, with GPU temperatures rising during compute-intensive tasks, causing performance degradation and increased latency. Students will evaluate these effects and develop an environment-aware iEDGE system that dynamically adjusts computing power based on real-time environmental data to minimize inference latency. This hands-on experience prepares students for real-world challenges in edge computing while advancing adaptive, sustainable mobile edge systems.