Program

Develop a 3D digital twin of a section of Bellevue College (or Bellevue) using open GIS data to visualize energy consumption, traffic, or public Wi-Fi usage patterns for better campus(or city) planning decisions. 

Develop a web platform where citizens can propose, vote, and collaborate on civic innovation ideas, integrating gamification to encourage participation. 

Develop a sensor-based system that monitors waste bin levels across the city and optimizes collection routes using real-time data analytics and AI-based predictions. 

 Develop an application that uses live data or simulated sensors to help drivers find available parking spots, estimate availability, and support payments. 

 A system to promote sustainability by monitoring streetlight usage and ambient light levels using IoT sensors. The goal is to dynamically adjust lighting intensity based on time of day, pedestrian movement, and weather conditions to save energy.

A data-driven platform that addresses the problem that standard navigation systems optimize for distance or time, ignoring personal safety. The goal is to enhance public safety by collecting and visualizing crime incident data, identifying hotspots, and providing safe route recommendations that minimize exposure to high-risk areas. The prototype case study is focused on the Greater Seattle–Bellevue Region.

A data-driven platform to address recurring challenges for public transit riders in cities like Bellevue, including unpredictable arrivals, poor coordination, and missed transfers. The goal is to use real-time open transit data and predictive AI to improve commuter experience, reliability, and citywide mobility planning. The prototype phase will focus on Bellevue College and surrounding routes as a controlled testbed. 

This project proposes an Urban Sound Classification System using Audio Machine Learning (ML) and IoT microphone nodes. The problem is that traditional noise meters measure decibel levels but cannot identify the source of the sound (e.g., traffic vs. construction vs. gunshots). The goal is to monitor, classify, and map city noise patterns in real time to provide data-driven insights for urban planners and public safety agencies. The prototype case study is focused on Bellevue, Washington. 

This project proposes a system (SLTLMS) to address the problem of traffic signals going dark during power outages in urban areas like Bellevue, WA. This "all-stop" mode causes congestion, confusion, and increased accident risk. The goal is to allow intersections to maintain limited, safe operation using battery/solar backup and AI-driven local logic. 

Estimate accurate depth information from single camera input Cluster objects around the vehicle based on estimated depth Extract boundaries for clustered objects Calculate and visualize actual minimum distance between objects Demonstrate potential for ADAS functionality expansion

Perception is a vital aspect of autonomous driving, accounting for most system complexity, with around 90% of the challenge focused on environmental sensing and understanding. Key issues include reduced camera obstacle detection accuracy in low light and LiDAR's limited sensing range despite being less affected by lighting. The main challenge is achieving reliable perception across various environments for safe driving. The project aims to test sensor-based autonomous driving on a small-scale vehicle, focusing on lane keeping, traffic signal recognition, obstacle avoidance, and parking execution.

Autonomous driving technology allows machines to navigate without human input using sensors, cameras, radar, and AI. It aims to enhance safety, reduce accidents, and improve mobility and logistics. Looking ahead, it will play a vital role in the smart mobility era for both vehicles and robots. Additionally, mobile and autonomous systems are needed to monitor safety in industrial sites to mitigate risks of fire and intrusion, despite a decline in industrial site fires.