Skip to content

Automate bin-picking tasks using the Dobot CR5 robotic arm. The system combines YOLOv5 for object detection and FAST/BRISK for feature detection and matching. With a depth sensor camera, we achieve real-time object detection and precise pose estimation, improving industrial automation and operational efficiency.

License

monkeyrom/3D_Object_Detection_and_Pose_Estimation_for_Automated_Bin-Picking_Application

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

46 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

3D Object Detection and Pose Estimation for Automated Bin-Picking Application

(The paper is in the process of being published)

Abstract

Robotic arms have gained popularity in various industries due to their accuracy and efficiency in completing tasks. In this study, we propose a method for automating bin-picking tasks using the Dobot CR5 robotic arm, combining the state-of-the-art YOLOv5 CNN model for object detection with traditional feature detectors, descriptors, and matching techniques. Specifically, we employ the FAST and BRISK algorithms for robust and efficient feature detectors, descriptors, and matching. By integrating these techniques and utilizing a depth sensor camera to capture depth and color images, our system achieves real-time object detection and precise pose estimation, enabling the robotic arm to pick objects accurately. This integration of small-scale camera technology with advanced algorithms contributes to the advancement of industrial robotics, opening up new possibilities for automating challenging tasks and enhancing overall operational efficiency.

Keywords: robotic arm, bin-picking, YOLOv5 CNN model, depth sensor camera, object detection and pose estimation

Hybrid Technique

Requirement

  • ubuntu 20.04
  • ROS noetic

Building

Use git to clone the source code

cd $HOME/catkin_ws/src
git clone https://github.com/Dobot-Arm/CR_ROS.git
git clone https://github.com/introlab/find-object.git
git clone https://github.com/monkeyrom/3D_Object_Detection_and_Pose_Estimation_for_Automated_Bin-Picking_Application.git

Installing Realsense-ROS

You need to install realsense-ros to use realsense2_camera package. The steps to installing realsense-ros have shown here.

building

cd $HOME/catkin_ws/src/Bin-Picking
catkin build

Set the robot type

echo "export DOBOT_TYPE=cr5" >> ~/.bashrc
source ~/.bashrc
source $HOME/catkin_ws/devel/setup.bash

1. Launch Project

  • Connect the robotic arm with the following command, and the default robot_ip is 192.168.1.6
    roslaunch CR5_Project CR5_with_realsense.launch
  • This command will launch
    • dobot_bringup
    • realsense camera pointcloud
    • find object 2d
    • tf synchronisation

rviz display

rviz display

find object

2. FAST and BRISK Traditional object detection for pose estimation

FAST Technique

3. Run a terminal for running Yolo node

    rosrun CR5_Project yolo_order.py
  • This command will run the node
    • yolo_listener

YOLOv5

4. Run a terminal for controlling the robot

    rosrun CR5_Project service_call
  • This command will run 2 nodes and spawn a new terminal for commanding
    • listener
    • main_order

Robot Pose Calculation

Real Robotic Arm

Dobot CR5

Intel Realsense D435i

Intel Realsense D435i

References

About

Automate bin-picking tasks using the Dobot CR5 robotic arm. The system combines YOLOv5 for object detection and FAST/BRISK for feature detection and matching. With a depth sensor camera, we achieve real-time object detection and precise pose estimation, improving industrial automation and operational efficiency.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published