PetGuard Pro
Smart Pet Collar for Real-Time Tracking, Geo-fencing & Health Monitoring
Team
- E/21/106 – Bingusari Dissanayaka
- E/21/137 – Chandur Fernando
- E/21/350 – Prashan Samarawickrama
- E/21/428 – Savin Weerasooriya
Table of Contents
- Introduction
- Solution Architecture
- Hardware & Software Designs
- Testing
- Detailed Budget
- Conclusion
- Links
Introduction
Pet owners face significant challenges in ensuring the safety and health of their pets due to limited real-time visibility and delayed responses to emergencies. Pets cannot communicate distress or abnormal conditions, making early detection difficult.
PetGuard Pro addresses this problem through an IoT-based smart pet collar integrated with a cloud backend and a mobile application. The system provides real-time location tracking, geo-fencing alerts, health monitoring, and intelligent notifications, enabling proactive and reliable pet care.
Solution Architecture
High Level Architecture Diagram
Prototype Implementation Diagram
PetGuard Pro follows a device–cloud–mobile architecture:
-
Smart Pet Collar (Edge Device)
Collects location, physiological, and motion data using onboard sensors. -
Cloud Backend (Firebase - Prototype)
Handles real-time data streaming, storage, and synchronization across devices. -
Mobile Application (Flutter)
Provides real-time visualization, alerts, and user interaction.
Communication Flow
- Sensors → ESP32 (data acquisition)
- ESP32 → Firebase (via HTTPS REST APIs over WiFi)
- Firebase → Mobile App (real-time streaming)
- Mobile App → User interface updates & alerts
- Communication uses HTTPS (REST) instead of MQTT for simplicity and native Firebase compatibility.
- Data is transmitted in JSON format with periodic updates (~10 seconds).
Hardware & Software Designs
Hardware Design (Pet Collar Unit)
- Microcontroller: ESP32-WROOM-32
- Positioning: NEO-M8N GPS
- Connectivity: WiFi (prototype), optional cellular for production
- Sensors:
- Temperature: MLX90614 (non-contact IR surface measurement)
- Heart Rate & SpO₂: MAX30102 (PPG-based sensing)
- Motion Tracking: MPU6050 (accelerometer + gyroscope)
- Power System:
- 3.7V Li-Po Battery
- Charging + voltage regulation circuitry
Software Design
Mobile Application
- Framework: Flutter (cross-platform, single codebase)
Core Features
- Real-time GPS tracking with map visualization
- Geo-fencing with boundary breach detection
- Health monitoring (heart rate, estimated body temperature)
- Activity recognition (resting, walking, active states)
- Smart alert notifications and history logs
Data Visualization & UX
- Interactive charts for health trends
- Google Maps integration with route history (polylines)
- Responsive UI using Material Design principles
Performance Optimization
- State Management: Riverpod
- Selective UI Rebuilds for efficiency
- GPS filtering using distance thresholds
- Local caching (Hive) for offline support and reduced latency
Cloud Backend (Prototype Implementation)
- Platform: Firebase
Services Used:
- Realtime Database: Live sensor data streaming
- Cloud Firestore: Structured storage (users, pets, logs)
- Firebase Authentication: Secure user access
- Cloud Functions (optional): Event-based processing
Key Features:
- Real-time synchronization between device and app
- Scalable NoSQL data structure (per-pet organization)
- Efficient separation of live vs historical data
- Offline persistence support
System Limitations & Workarounds
-
GPS inaccuracy (indoors / dense areas)
→ Uses filtered updates and last-known location fallback -
PPG signal noise (motion artifacts)
→ Moving average filtering and stability checks -
Surface temperature vs core body temperature
→ Offset-based calibration to estimate internal temperature
→ Used for trend monitoring rather than medical accuracy -
WiFi instability (ESP32)
→ Local buffering and retry mechanism for reliable transmission
Testing
Testing was conducted using a layered, progressive validation strategy, ensuring reliability at each stage of system development—from individual components to full system integration.
Testing Strategy & Validation Approach
API-Level Testing
Initial validation focused on backend APIs using Postman.
- Verified correct responses for all endpoints
- Ensured consistency of JSON data structures
- Validated error handling and edge-case behavior
Firebase Integration Testing
The system was then integrated with Firebase to validate real-time data handling.
- Manual database updates were performed
- Verified instant synchronization with the mobile application
- Confirmed stability and accuracy of real-time data streaming
Hardware Data Simulation (Pre-Hardware Phase)
Before hardware availability, sensor data was simulated using an MQTT-based approach (HiveMQTT).
- Emulated continuous sensor data streams
- Validated end-to-end flow: device → cloud → mobile app
- Ensured system behavior under real-time update conditions
Hardware Integration & Functional Testing
After assembling the hardware, full system testing was conducted using WiFi-based communication.
- Verified sensor readings:
- Heart rate (MAX30102)
- Temperature (MLX90614)
- Activity (MPU6050)
- Confirmed real-time data updates in the mobile application
- Evaluated system responsiveness during continuous operation
Scenario-based validation included:
- Motion simulation for activity detection
- Body contact for heart rate measurement
- Environmental variation for temperature sensing
Communication & Reliability Testing
- Tested data transmission stability over WiFi
- Validated retry mechanisms under intermittent connectivity
- Measured real-time update latency and consistency
Application Testing
- UI responsiveness and performance validation
- State management behavior using Riverpod
- Firebase integration and real-time updates
- Notification triggering and alert accuracy
Field Testing
- GPS accuracy in indoor and outdoor environments
- Geo-fence boundary detection and alert triggering
- System behavior under real-world usage conditions
Future Testing Plan (Scalability & Network Robustness)
Large-scale testing has not yet been conducted due to the use of Firebase for rapid prototyping.
Planned future work includes:
- Migration to AWS-based scalable infrastructure
- Load testing with multiple devices and users
- Performance evaluation under high data throughput
- Integration of cellular (GSM) communication
- Validation under real network constraints
Detailed Budget
| Item | Quantity | Unit Cost (LKR) | Total (LKR) |
|---|---|---|---|
| ESP32 MCU | 1 | 1500 | 1500 |
| GPS Module (NEO-M8N) | 1 | 3000 | 3000 |
| Health Sensors (PPG, IMU, Temp) | 1 set | 3300 | 3300 |
| Battery & Charging Circuit | 1 | 2200 | 2200 |
| Buck Converter | 1 | 1000 | 1000 |
| LEDs & Buzzer | 1 set | 150 | 150 |
| PCB, Wiring & Connectors | 1 set | 1000 | 1000 |
| Enclosure | 1 | 800 | 800 |
| Total | 12,950 |
Conclusion
PetGuard Pro demonstrates a scalable and efficient IoT-based smart pet monitoring system. By integrating embedded sensing, cloud-based data processing, and a responsive mobile application, the system enables real-time safety monitoring and health insights.
The architecture is designed for extensibility, allowing future enhancements such as cellular connectivity, advanced analytics, and improved power optimization for real-world deployment.