⚠️ Common Mistakes & How to Avoid Them

Robotics development is full of surprises — some pleasant, many painful.
Here are the most common mistakes made during this project, how they were fixed, and what you can do to avoid them.

---

🛑 Mistake 1 — Building Inside the Source Workspace

What Happened:
In the first week, I ran colcon build inside the robot’s factory source workspace instead of an external one.
This overwrote configs and broke the system.

How It Was Fixed:

• Requested a pre-flashed SD card from the vendor.
• Reinstalled ROS2 + original robot packages.
• Created a separate ~/krish_ws workspace for all custom code.

Lesson Learned

Always develop in a dedicated external workspace — never touch the vendor’s source workspace.

---

🔄 Mistake 2 — Doing Everything in One Node

What Happened:
Early attempts combined detection, navigation, and arm control in a single giant node.
This caused clashes in state handling and made debugging impossible.

How It Was Fixed:

• Split the system into modular packages:
  • object_detector
  • nav_handler (Mission Manager + Pose Setter)
  • mycobot_arm (Pick & Drop)
• Used ROS2 topics and actions for communication.

Lesson: Modular design means one node crash ≠ total system failure.

---

📡 Mistake 3 — Hardware Bottlenecks

What Happened:
The MyCobot arm’s USB control board failed, breaking wired control.

How It Was Fixed:

• Flashed the arm’s firmware for Wi-Fi control.
• Wrote IP auto-detection logic in full_system.launch.py to connect without manual setup.

Lesson: Hardware failures can be solved creatively — and sometimes upgraded in the process.

---

🗺 Mistake 4 — Poor Map & Pose Initialization

What Happened:
Manual AMCL initialization was slow and error-prone, causing bad navigation.

How It Was Fixed:

• Created pose_setter.py to automatically publish the correct initial pose after Nav2 starts.
• Tuned Nav2 parameters to match the working environment.

Lesson: Automating localization removes an entire category of human error.

---

🧮 Mistake 5 — Navigation & Detection Race Conditions

What Happened:
When YOLO detected an object mid-navigation, Nav2 would keep executing the old goal, leading to collisions or missed pickups.

How It Was Fixed:

• Added interrupt logic in mission_manager.py to:
  • Cancel current goals.
  • Stop scanning.
  • Navigate to the detected object.
• Used threading to delay and control transitions.

Lesson: Robust state management is essential for multi-component robotics systems.

---

📝 Tips to Avoid These Mistakes

• Modularize early — split detection, navigation, and manipulation from the start.
• Back up configs before testing major changes.
• Automate setup (pose setting, IP detection) wherever possible.
• Sequence startup with TimerAction to avoid race conditions.
• Test in parts before running the full workflow.

Pro Tip

If something breaks unexpectedly, disable all non-essential nodes and run the system in isolation to find the problem fast.