🏆 CheckMate — CAT Hackathon 2025 Winning Project

Team HackaPillar

CheckMate is a real-time IoT-powered smart assistant designed to improve safety, productivity, and efficiency in heavy vehicle operations. This award-winning system integrates sensor networks, AI/ML models, and an intelligent chatbot assistant powered by the Gemini API, delivering critical operational insights and support to operators on the field.

---

🚀 Features

• 🔐 RFID-Based Operator Authentication
• 😴 Drowsiness Detection (Teachable Machine)
• 🌡️ Engine Temperature Monitoring (DHT22)
• 🎯 Seatbelt Compliance Detection
• 🚧 Obstacle Detection (HC-SR04)
• 🧠 Engine Fault Detection using ML
• 🏔️📈 Smart Speed Estimation Based on Terrain Slope
  Uses MPU6050 sensor data to estimate terrain slope (inclination) and predict the vehicle’s optimal speed in real-time using ML models.
• ⏱️ Task Completion Time Prediction using regression models
• 📡 Live Dashboard with Real-Time Alerts via WebSockets

---

🧠 Machine Learning Modules

Feature	Model Used	Input Data
Smart Speed Estimation	Regression (Random Forest)	MPU6050 data (accelerometer + gyro) + Slope
Engine Fault Detection	Classification Model	Temp, vibration, obstacle sensors
Task Time Estimation	Regression Model	Vehicle ID + Task Type + Sensor Metrics

➡️ Terrain slope is computed from MPU6050 data and used as a key input in speed estimation, ensuring the system adapts to uphill/downhill conditions.

---

🛠️ Tech Stack

• Hardware: ESP32, RFID Reader, DHT22, HC-SR04, MPU6050
• Embedded: Arduino/C++
• Backend:
  • FastAPI for ML Model Serving
  • Node.js + Express for API Gateway + WebSocket Integration
• Frontend: React.js (Live Dashboard)
• Database: MongoDB (Operator & Task Logs)
• AI Tools: Scikit-learn, Teachable Machine
• Protocols: HTTP, WebSocket

---

🤝 Contributing & Contact

We welcome contributions, feature suggestions, and feedback!

Team HackaPillar
🏆 Winners of CAT Hackathon 2025