This project delivered a drone-powered platform that provides real-time, location-based information for festival sites and crowded areas—such as live weather conditions, congestion level, and parking availability.
We built a real-time data visualization engine that processes key signals collected from drone-mounted sensors (weather, crowd density, and parking capacity) and converts them into intuitive visual outputs. We also designed a map-based UI that lets users tap a specific point to instantly view detailed on-site status.
In addition, we implemented a high-quality live video streaming system fully integrated with drone cameras. Using RTMP and web streaming protocols, the platform supports stable real-time broadcasting. To meet different user needs, we delivered a separated interface structure: an admin console for operating drones and managing data/streams, and a user-facing view focused on fast information lookup and live monitoring.
md
코드 복사
1) Drone-based real-time regional information system
- Collect and analyze on-site data (weather, crowding, parking availability) via drone sensors and cameras.
- Provide continuous, real-time updates that reflect rapidly changing conditions.
- Visualize complex data in an easy-to-understand format (charts/graphics), not raw numbers.
- Support use cases like festivals and large public events so visitors can plan visits efficiently.
2) Real-time live video streaming
- Deliver live footage from drone-mounted high-resolution cameras to show the actual situation on site.
- Consider multi-stream environments where multiple drones/locations can be monitored.
- Ensure stable streaming with minimal lag and interruptions.
- Provide flexible viewing (single stream selection, multi-stream comparison).
3) Separate admin and user modes
- Admin mode: flight operations, monitoring, data registration/management, and stream control.
- User mode: simple, intuitive access to real-time status + live video with minimal learning curve.
1) Real-time data collection + visualization system
- Developed a real-time processing pipeline for sensor-based data (weather, crowd density, parking space).
- Built a visualization engine to convert raw data into charts and graphical indicators.
- Designed a map-based UI: users click/tap a point to view live status in a popup panel.
- Optimized the frontend for low latency by focusing on core functions and lightweight rendering.
- Provided data-type-optimized charts and graphs to maximize readability and decision speed.
2) Live streaming integration (RTMP + web streaming)
- Implemented a high-quality live streaming system linked directly to drone cameras.
- Supported RTMP and web streaming protocols for broader device/network compatibility.
- Added adaptive behavior to maintain stable viewing depending on network conditions.
- Designed a multi-channel structure: each drone assigned its own stream channel for scalable monitoring.
- Enabled flexible watching modes (select one drone, monitor multiple feeds).
3) Admin / user interface separation
- Admin console:
- Drone flight path setup and real-time control
- Monitoring for battery/flight/data collection status
- Data CRUD (create/edit/delete) for collected information
- Stream start/stop controls per drone
- Dashboard-style operational visibility for fast response
- User interface:
- Location-based, map-centered information lookup
- Quick access to live video for selected areas
- Simplified interaction model focused on viewing (not control)
- Reduced friction with lightweight login and minimal steps to access services
