Luminapath

Inspiration

Bellevue is becoming harder and more dangerous to navigate as a pedestrian. In 2024, the city reached a 10-year high in serious traffic injuries, and 72% of these severe pedestrian crashes occurred at intersections. At the same time, PNW climate change is increasing heavy rainfall, reducing driver visibility and making traditional crosswalk markings ineffective during storms.

During our research, we discovered how Daegu, South Korea, combats distracted walking and low-visibility conditions using ground-level LED crosswalks and AI-assisted school-zone management. This cross-cultural connection inspired us to adapt and enhance their approach for Bellevue’s unique climate and safety challenges. That inspiration became LuminaPath — a smart, self-powered, sensing crosswalk designed to protect pedestrians even when they aren’t paying attention.

What it does

LuminaPath is a smart-sensing pedestrian shield embedded directly into the ground. It improves safety through four connected features:

Dynamic Speed Alerts: LEDs pulse red when a vehicle is approaching too fast for weather or stopping distance, breaking “Smombie” (smartphone zombie) distraction.

Extreme Visibility: Heavy-duty polycarbonate LED strips remain visible through fog, glare, and intense rainfall.

Off-Grid Resilience: Solar panels and LiFePO₄ batteries keep the system active during outages and storms.

Privacy-First Sensing: Radar and LiDAR measure shapes and speeds without capturing faces or license plates.

We also used Gemini to analyze crash data and prioritize which intersections to upgrade first.

How we built it

We combined safety research, crosswalk engineering principles, and Korean smart city case studies into a single integrated system:

Problem modeling – We analyzed Bellevue crash data and rainfall projections to define failure points in current crosswalk infrastructure.

Global benchmarking – We studied Daegu and Seoul’s ground-level LED technology, “Smombie alert” systems, and AI-assisted school zones.

Hardware design – We designed an embedded LED module with solar power, rugged housing, and a low-power controller.

Sensor logic – Using radar/LiDAR detection, we created logic to identify approach speed, pedestrian presence, and unsafe weather conditions.

AI-assisted risk mapping – With Google Gemini, we created a prioritization model to identify high-injury intersections for deployment.

While we could not build a physical prototype in the hackathon timeframe, we produced a full system design, deployment strategy, and demonstration logic.

Challenges we ran into

Balancing innovation with feasibility: Ensuring our idea was creative but also realistic for a city to deploy.

Adapting Korean systems to Bellevue: Weather, traffic patterns, and cultural behaviors differ greatly between regions.

Ensuring privacy: We had to avoid camera-based detection while still providing accurate sensing.

System survivability: Designing for storms, flooding, and outages required off-grid resilience and ruggedization.

AI risk scoring: Modeling high-risk intersections with limited public data was a challenge, but Gemini helped synthesize multiple datasets.

Accomplishments that we're proud of

Creating a fully integrated, climate-resilient safety system instead of just adding lights to a crosswalk.

Designing a solution inspired by real-world Korean innovation and adapting it meaningfully for Bellevue.

Building a privacy-safe sensing system that avoids surveillance concerns.

Using Gemini to produce a risk-based deployment plan grounded in real data.

Producing a solution that feels deployable, impactful, and technically grounded within a single weekend.

What we learned

How drastically climate change is affecting pedestrian safety, especially at intersections.

How Korean smart cities approach walkability and safety with layered, tech-forward methods.

The importance of inclusive design: protecting distracted walkers, children, elders, and transit users.

How to translate global solutions into local, culturally relevant ones.

How AI can assist in urban planning decisions like prioritizing high-risk intersections.

What's next for LuminaPath

Our next steps include:

Building a functional prototype of the LED module and sensing system.

Partnering with Bellevue Transportation to pilot LuminaPath at a high-injury intersection such as NE 8th & Bellevue Way.

Integrating real-time weather data to further refine speed alert thresholds.

Expanding the system to school zones and transit corridors.

Developing a dashboard for city engineers to monitor performance and adapt placement over time.

LuminaPath is built to grow. As rainfall intensifies and distractions increase, we envision a city where pedestrians walk with confidence — because the ground itself protects them.