Urban Sentil

• 

  Urban Sentil

Inspiration

UrbanSentinel is a decentralized, privacy-preserving safety net designed to detect urban emergencies such as gunshots, falls, and medical distress. Unlike traditional cloud-based acoustic monitoring systems, UrbanSentinel relies entirely on Edge AI and Federated Learning. The system processes all acoustic data locally on edge devices (zero cloud storage of PII) and transmits only critical metadata anomalies and encrypted model weight updates, ensuring complete citizen privacy while maximizing dispatch efficiency.

What it does

The system processes all acoustic data locally on edge devices (zero cloud storage of PII) and transmits only critical metadata anomalies and encrypted model weight updates, ensuring complete citizen privacy while maximizing dispatch efficiency.

How we built it

The architecture is divided into three distinct geographic and logical layers: the Edge Layer, the Aggregation Layer, and the Presentation Layer.

Challenges we ran into

It is hard to demo it

Accomplishments that we're proud of

n/a

What we learned

AI tool

What's next for Urban Sentil

try to make it work

Built With

• javascript
• python
• typescript