Autonomous Vision-Based Line Following Robot

This was our final embedded systems project.

📝 Abstract

This project focuses on the design and development of an autonomous line-following robot in compliance with the RoboCupJunior Rescue Line rules.


⚙️ Platform & Tools

  • Hardware: NVIDIA JetBot platform
  • Software: OpenCV-based vision pipeline

🚦 Project Goal

The robot’s task was to follow a black line on a white floor while dynamically adapting to:

  • Turns
  • Roundabouts where a direction must be taken
  • Gaps and interruptions in the track

🧠 Implementation

  • Line detection & navigation: Classical computer vision with OpenCV to preprocess inputs, find the line’s centroid, and compute steering corrections
  • Intersection handling: Detected green markers → triggered predefined actions (turns, U-turns)
  • Obstacle recovery: Integrated detection & fallback strategies to maintain continuous operation

🎯 Key Takeaway

This project emphasizes real-world adaptability under competition constraints (unpredictable layouts, environmental variations).
By relying on efficient OpenCV algorithms, we demonstrated that classical vision techniques can successfully solve autonomous navigation challenges in a competitive setting.

🎥 Project Demo

Watch the Robot in Action

📄 Project Report

Read the Full Report

Explore the methodology, experiments, and results in detail below: