Project Overview: Robotic Order Fulfillment System

Introduction

• Objective: To modernize the factory order fulfillment process with an automated robotic system.
• Technologies Employed: Advanced robotics, ROS (Robot Operating System), intelligent algorithms, and user-centric design principles.
• Primary Tools: MOVEIT for robotic motion planning, RViz for visualization, Gazebo for simulation.
• Skills Demonstrated: Robotic programming, automated system integration, dynamic path planning, and real-time processing.

System Description

• Components: A fleet of advanced robotic arms and mobile robots optimized for precise and efficient order handling.
• Operations:
  • Automated retrieval and packaging of items using robotic arms.
  • Autonomous delivery robots ensuring accurate and timely order delivery.
• Design Focus: Enhanced interaction between human operators and robotic systems for streamlined operations.

ROS Architecture Comprehensive Overview

Detailed Node Functions

• Arm1 ROS Node

  • Role: Central processing unit for order management, handling the specifics of order retrieval and prioritization.
  • Priority-Based Management: Utilizes a sophisticated priority queue to manage and execute orders based on urgency and importance.
  • ROS Services:
    • AddOrderService: Enables addition and prioritization of new orders into the system's queue.
    • GetOrderService: Provides real-time updates on order status and initiates order execution.
  • MOVEIT Integration: Employs MOVEIT for advanced motion planning, ensuring optimal and safe robotic movements.
  • Inter-node Communication:
    • Acts as a service client to Arm2, facilitating precise object placement based on order priority.

• Arm2 ROS Node

  • Functionality: Executes actions related to object pickup and placement, directly interacting with the conveyor belt operations.
  • Services and Actions:
    • Receives specific action commands from the Object Detector Node to adjust positioning or initiate pickups.
    • Provides a service for determining and communicating the correct placement box based on object priority, enhancing order accuracy.

• Object Detector Node

  • Capabilities: Detects and classifies objects moving on the conveyor using advanced computer vision techniques.
  • Data Handling:
    • Subscribes to /conveyor/camera1/image_raw for real-time image processing.
    • Publishes to image_output for monitoring and debugging of the detection process.
  • Role: Specifies actions for Arm2 based on object detection and classification results.

Additional Services and Topics

• Respawn_objects Service:

  • Utilized by Arm1 to manage the respawn of objects on the boxes, ensuring continuous operation without manual resets.
  • Arm1 sends service requests to the Spawn_Objects node, which processes these requests to maintain system readiness and object availability.

• gripper/control Service:

  • Allows Arm1 to manipulate the vacuum gripper for secure object handling during the order fulfillment process.

• ROS Topics for Visual Feedback and Debugging:

  • ariac/arm1/move_group/display_planned_path: Used by Arm1 to display the planned motion paths on RViz, providing a visual tool for monitoring and adjusting robotic movements.

System Limitations and Future Directions

• Current Shortcomings:
  • Reduced efficiency with high-speed conveyor operations.
  • Limited capability in detecting and handling multiple objects simultaneously.
• Proposed Enhancements:
  • Implement a feedback loop between arms to improve synchronization and error handling.
  • Integrate speed sensors to adapt robotic actions to varying conveyor speeds.

Additional Resources

• Project Demonstration Video: Watch Here