Singapore Polytechnic Internship

During my internship at Singapore Polytechnic, I developed a remote monitoring dashboard for a fleet of autonomous vehicles. The project addressed the limitations of a local Node-RED dashboard, which could only monitor one vehicle at a time, by enabling centralized oversight as the fleet expanded.

I created a full-stack web application consisting of a JavaScript server and a React-based frontend. The system integrated vehicle sensors, GPS data, and IP camera feeds to provide real-time and historical monitoring capabilities.

• Remote monitoring dashboard for multiple autonomous vehicles
• Live GPS tracking using Google Maps Platform
• Real-time sensor readings (speed, humidity, temperature)
• Historical charts for logged sensor data
• MySQL database for data storage and logging
• MQTT-based data pipeline for efficient sensor data transmission
• IP camera video streaming converted to HLS using ffmpeg
• Threshold-based sensor color coding for quick visualization (e.g., green for normal, red for out-of-threshold)

• Backend: JavaScript, ExpressJS, MQTT, MySQL, ffmpeg
• Frontend: ReactJS, MUI, Google Maps Platform, react-chartjs-2
• Hardware Integration: Node-RED for sensor management, IP camera with onboard router

• Transitioning from local to remote monitoring required building a scalable data pipeline.
• Implementing real-time video streaming involved converting IP camera feeds to HLS format.
• Ensuring data integrity through preprocessing, such as duplicate checks, before database storage.

• Designing and implementing MQTT-based data pipelines for IoT applications.
• Integrating multimedia streaming (video) into web dashboards.
• Using polling mechanisms to maintain server-client synchronization for real-time data.
• Applying popular frameworks like React and Express for production-ready applications.

The dashboard successfully enabled remote monitoring of multiple vehicles, with data logged to a MySQL database hosted on Huawei Cloud. It met all requested features, including live feeds and visualizations, and was valued by supervisors for its use of popular, maintainable technologies.