Project
Nephele 2.0 – Interaction Robot
Nephele 2.0 is an intelligent, interactive robot developed under the AWS Cloud Club to enhance conferences and public events. Designed with a custom-built robotic body, precise motor control, real-time face recognition, and cloud-backed voice intelligence, Nephele delivers personalized, engaging interactions for attendees in real-world environments.
Talk with Nephele📸Project Gallery
🎬Project Media
Nephele 2.0 interacting with attendees at a live conference
Nephele 2.0 interacting with attendees at a live conference
👤My Role & Contributions
Project Lead – Hardware, Robotics & AI Systems
Key Responsibilities
- •Served as Project Lead for the Nephele initiative across versions 1.0 to 3.0
- •Designed the complete mechanical and industrial body of the Nephele robot
- •Implemented stepper motor control and motor mapping using Raspberry Pi 5
- •Handled hardware integration including motors, drivers, camera, display, and power systems
- •Developed and optimized machine learning models for real-time face detection and recognition
- •Built computer vision pipelines for human interaction and gesture-based input
- •Worked on 5G-based connectivity concepts for inter-robot (swarm) communication
- •Contributed to cloud engineering by integrating AWS services for speech and AI processing
- •Collaborated with software and cloud team members to ensure end-to-end system stability
- •Managed hardware testing, calibration, and live deployment during events
Technologies Used
🏗️System Architecture
Click to zoom
Problem
Large conferences and public events face challenges such as limited attendee engagement, manual attendance tracking, lack of personalized assistance, and difficulty scaling human-operated help desks across large venues.
Approach
Nephele 2.0 combines embedded hardware control, computer vision, and cloud intelligence. A Raspberry Pi 5 manages real-time perception, motor control, and interaction logic. Stepper motors enable precise navigation, while camera-based vision supports face detection and gesture recognition. Machine learning models handle real-time face recognition. AWS services are integrated for speech recognition, natural language understanding, and speech synthesis, enabling seamless voice-based interaction. High-speed wireless (5G) connectivity enables coordination concepts across multiple robots.
Outcome
Nephele 2.0 successfully demonstrated reliable human–robot interaction in live conference environments. The system delivered personalized greetings, automated attendance workflows, and real-time question answering while maintaining stable performance. The modular design allowed smooth iteration from Nephele 1.0 to later versions, validating the architecture for future expansion.
