Navan is an autonomous robotic boat built to tackle plastic pollution in waterbodies. Using computer vision and intelligent navigation, Navan detects, locates, and collects plastic waste, helping to keep rivers, lakes, and coastal areas clean.
Plastic detection using YOLOv8, smart navigation, adaptive speed control, and systematic search patterns.
Web dashboard with keyboard controls, real-time camera feed, and variable speed adjustments.
Dual mode operation (Manual & Autonomous), WiFi communication via ESP32, and real-time status monitoring.
Autonomous Mode
1. Camera captures waterbody images
2. YOLOv8 detects plastic in real-time
3. Navigation system targets waste
4. Speed adapts as boat approaches
5. Performs search patterns if no plastic found
Manual Mode
1. Operator controls via web interface
2. Commands sent to Raspberry Pi Flask API
3. API relays to ESP32 motor controller
4. ESP32 executes movement commands
5. Live camera provides real-time feedback
Brain of Navan – AI & control system. Flask REST API, YOLOv8 integration, camera image processing, and navigation logic.
Tech Stack: Python, Flask, OpenCV, YOLOv8
Web dashboard with manual control, real-time camera feed, and mode switching.
Tech Stack: HTML5, CSS3, JavaScript
ESP32 hardware control – WiFi server, brushless motor control, and real-time command execution.
Tech Stack: C++, Arduino, ESP32
YOLOv8 model training for underwater plastic detection. Pre-trained weights available.
Tech Stack: Python, YOLOv8