A multi-modal IoT + Affective Computing system for real-time emotional well-being support.
GUARDIAN ANGEL is an IoT-based emotion-aware monitoring system designed to provide real-time emotional feedback using physiological biosignals.
The system continuously monitors Heart Rate Variability (HRV) and Electrodermal Activity (EDA/GSR), classifies the user's emotional state using Machine Learning, and delivers personalized ambient feedback through lighting and audio.
The solution integrates ESP32 sensing, AWS cloud services, and Raspberry Pi automation to create a complete real-time emotion monitoring system.
- Real-time HRV and EDA/GSR monitoring using ESP32
- MQTT-based communication with AWS IoT Core
- Emotion classification using an LSTM model
- Serverless prediction with AWS Lambda
- Raspberry Pi-based ambient feedback system
- RGB LED and audio control based on detected emotion
- User stop button for manual interruption
- End-to-end IoT architecture
- Reads physiological sensor data
- Publishes data securely using MQTT
- AWS IoT Core receives sensor data
- AWS Lambda performs emotion prediction using the trained model
- Retrieves prediction results
- Controls RGB LEDs
- Plays emotion-specific audio
- Supports manual stop functionality
- Heart Rate Variability (HRV)
- Electrodermal Activity (EDA/GSR)
These physiological indicators are used to estimate the user's emotional state.
- ESP32 Development Board
- MAX30102 Pulse Sensor
- GSR/EDA Sensor
- Raspberry Pi
- RGB LED
- Speaker
- Push Button
- AWS IoT Core
- AWS Lambda
- MQTT Protocol
- ThingSpeak
The project uses an LSTM-based neural network for emotion classification.
| Class | Emotion |
|---|---|
| 0 | Neutral |
| 1 | Happy |
| 2 | Sad |
| 3 | Stress |
| Emotion | LED | Audio |
|---|---|---|
| Neutral | Blue | Calm Music |
| Happy | Green | Happy Music |
| Sad | Cyan | Relaxing Music |
| Stress | Red | Stress Relief Audio |
- ESP32 collects HRV and GSR data.
- Sensor data is transmitted to AWS IoT Core.
- AWS Lambda predicts the emotional state.
- Raspberry Pi receives the prediction.
- LEDs and audio are adjusted according to the detected emotion.
- The user can stop the feedback using the physical button.
Emotion-Aware-Rooms
│
├── ESP32-controller_code.ino
├── data_preprocessing.py
├── dataset_lstm.py
├── lstm_train.py
├── lambda_function.py
├── raspby_room_control.py
├── NODEMCU-CIRCUIT.png
├── RASPBERRY-CIRCUIT.png
├── README.md
- Python
- Arduino (ESP32)
- Raspberry Pi
- AWS IoT Core
- AWS Lambda
- MQTT
- ThingSpeak
- TensorFlow / Keras
- LSTM
- Machine Learning
- Smart Homes
- Healthcare Monitoring
- Elderly Care
- Student Wellness
- Emotion-Aware Automation
- Ambient Assisted Living
- Mobile application
- Web dashboard
- Real-time analytics
- Cloud database integration
- Multiple emotion detection models
- Voice emotion recognition
- Wearable device integration
GUARDIAN ANGEL demonstrates how IoT, cloud computing, and machine learning can work together to create an intelligent emotion-aware monitoring system. By combining physiological sensing with automated ambient feedback, the system provides a scalable foundation for smart wellness applications.