Wildcard Week¶

🧠 Idea¶

In this week, I explored autonomous navigation for my delivery robot using Jetson Nano, Raspberry Pi, YDLIDAR, YOLOv5, and ROS2. The goal was to integrate machine vision and SLAM (Simultaneous Localization and Mapping) for a smart robot that can see, detect, and map its environment.

Tools & Technologies¶

• 📦 Jetson Nano – for real-time AI inference
• 📷 Webcam + Pi Camera – for computer vision
• 📡 YDLIDAR X3 Pro – for 360° mapping
• 🐍 Python + YOLOv5 – object detection
• 🤖 ROS2 Foxy – for SLAM and robot control
• 🛠️ Raspberry Pi 5 – secondary controller
• 🧭 RViz + SLAM – visualization and localization

--

🖼️ Photos from the Process¶

🔌 Jetson Nano Setup¶

🧭 LIDAR Connection and Testing¶

🧑‍💻 Object Detection with YOLOv5¶

⚙️ ROS2 & SLAM Setup¶

📷 Camera Integration for Live Vision¶

💻 ROS Node Debugging and Mapping¶

🔧 Terminal Output and Sensor Logs¶

🎯 AprilTag Localization¶

💬 Object and Person Detection with Jetson¶

🔍 Summary¶

This week's work allowed me to: - Set up a real-time object detection system using Jetson Nano - Use YDLIDAR for generating 2D maps and navigate autonomously - Learn how to work with ROS2, RViz, and SLAM - Combine machine learning and robotic sensing in one pipeline

📂 Files & Resources¶

• YOLOv5 model (trained weights)
• LIDAR SDK and ROS2 package
• Launch files and ROS2 nodes for mapping and navigation
• Python scripts for vision and data fusion

🧠 Category¶

Machine Vision + Embedded Programming + Robotics (SLAM)
This fits into Wildcard Week because it combines embedded AI, ROS2, SLAM, and LIDAR — technologies not covered in other weeks.