🖥️ System Components Overview

See the source code for implementation details.

Your LIMO Pro + MyCobot project is made up of modular ROS2 nodes, connected by topics, actions, and services.
This architecture allows for flexibility, fault isolation, and clear debugging.

---

🏗️ System Architecture

---

🧩 Component Relationships

Component	Role	Key Dependencies
Object Detector	Detects objects, outputs 3D poses	Camera, depth data, YOLO weights
SLAM Node	Builds/updates map, localizes robot	LiDAR, odometry, camera
Navigation Stack	Plans paths & controls movement	Map from SLAM, goals from Mission
Pose Setter	Sets initial robot pose after startup	Map coordinates
Mission Manager	Main logic — coordinates detection, navigation, pick/drop	All other nodes
Pick Node	Picks object using arm & gripper	MyCobot arm, target pose
Drop Node	Drops object at target location	MyCobot arm, predefined drop pose

---

🔄 Communication Flow

1. SLAM Node

   • Publishes /map, /odom, /tf
   • Used by Navigation for localization

2. Object Detector Node

   • Subscribes to /camera/color/image_raw and /camera/depth/image_raw
   • Publishes /target_pose (geometry_msgs/PoseStamped)

3. Mission Manager Node

   • Subscribes to /target_pose
   • Sends goals to Navigation via /navigate_to_pose action
   • Calls Pick/Drop nodes when needed

4. Pick & Drop Nodes

   • Receive triggers from Mission Manager
   • Send movement/gripper commands to MyCobot

5. Navigation Stack (Nav2)

   • Executes goals from Mission Manager
   • Publishes feedback and result back to Mission Manager

---

🔌 Integration Points

Integration Point	Connected Components	Purpose
/target_pose	Detector → Mission Manager	Trigger navigation to detected object
/navigate_to_pose action	Mission Manager → Navigation Stack	Autonomous movement to location
Arm control service/action	Pick/Drop → MyCobot Arm	Physical manipulation
/map & /odom topics	SLAM → Navigation Stack	Position tracking & path planning
/cmd_vel	Navigation Stack → LIMO Base	Direct velocity control

---

💡 Best Practices

• Launch nodes in correct order (full_system.launch.py handles this)
• Keep node responsibilities separate for easier debugging
• Use RViz to visualize map, poses, goals, and detections in real time
• Monitor topics with:

  ros2 topic list
  ros2 topic echo /target_pose
  ros2 action list

---

🎯 Next Steps

• System Integration
• SLAM
• Waypoint Navigation
• Object Detection Pipeline