AI traffic management on Google Cloud
Camera feeds to adaptive signal timing, with a human override.
- Status
- PrototypeComplete proof of concept with deployment scripts
- Period
- Sep 2025
- Role
- Architecture and full implementation
- Brand
- Tuchopini
Serverless functions extract frames from traffic cameras into a message queue; a streaming pipeline runs them through cloud vehicle detection into a data warehouse; a signal controller adapts timing per approach with emergency-vehicle and manual override; an API service feeds a live dashboard. Nine subsystems including a simulation harness and automated cloud setup.
What it does
- Queued ingestion that absorbs camera and network failures
- Cloud vision detection and classification
- Adaptive signal timing with emergency override
- Warehouse analytics, monitoring, simulation
An AI that cannot be overridden at a real junction is a liability, not a feature. The override path is designed in, not bolted on.
More ai, vision and iot
Built ▶ 1 AI, vision and IoT
AI traffic survey and vehicle counting
Survey-grade counts from video: 14 categories, by direction, in 15-minute slots.
Sep 2025 → present
Prototype ▶ 1 AI, vision and IoT
Smart traffic lights (ESP8266 + Raspberry Pi)
Two coordinated intersections on commodity hardware, and a vision-driven signal demo.
Sep 2025
AI, vision and IoT
Smart parking occupancy
Two Raspberry Pis, two sensors, one live bay count you can trust.
Sep 2025