Skip to main content

💡 Tips & Tricks

Whether you’re just starting with the LIMO Pro + MyCobot + YOLO system or already deep into development, these tips will help you work smarter, debug faster, and optimize performance.

🚀 Quick Wins for New Users​

  • Always develop in a separate colcon workspace — never in the robot’s source workspace.
  • Keep two SD cards: one in use, one as a bootable backup.
  • Store your map files, YOLO weights, and config YAMLs in a Git repo for easy restoration.

⚡ Performance Optimization​

  • YOLO Speed:
    • Reduce image resolution if latency is high.
    • Increase conf threshold in object_detector.py to filter noise.
  • Navigation Smoothness:
    • Tune Nav2 parameters in nav2_params.yaml — especially acceleration and max velocity.
    • Avoid overloading the CPU by running visualization tools (rviz2, rqt_image_view) on a remote PC.
  • Arm Control:
    • Wi-Fi control avoids USB bottlenecks and improves reliability.
    • Use IP auto-detection logic from full_system.launch.py to remove connection delays.

🔄 Workflow Improvements​

Use Modular Testing

Run and verify each subsystem individually before full integration:

  1. YOLO Detection — run object_detector alone, check /yolo/annotated.
  2. Navigation — run Nav2 with pose_setter.py, verify waypoint traversal.
  3. Arm Control — test pick_node.py and drop_node.py without navigation.
  • Use TimerAction in launch files to stagger startup and avoid race conditions.
  • Keep a "safe" default map and a "development" map — switch depending on your testing environment.

⏱ Time-Saving Techniques​

  • Parameterize everything — IPs, file paths, thresholds — so you can change them via CLI instead of editing code.
  • Use ros2 topic echo and ros2 topic hz to quickly check if a topic is alive and healthy.
  • For fast restarts, kill all nodes: 
    pkill -9 -f ros2
on August 8, 2025 by KRISH RAY