CR5_Project
                                
                                
                                
                                    CR5_Project copied to clipboard
                            
                            
                            
                        Dobot CR5 with Intel Realsense D435i for object detection and pose estimation (bin-picking application).
CR5_Project 
Dobot CR5 with Intel Realsense D435i for object detection and pose estimation (bin-picking application).
This project is present the robot arm application especially bin-picking base on Dobot CR5 using ROS which detect an object and 6d pose estimation with intel realsense d435i depth camera by detection find-object app.
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/IntelRealSense/realsense-ros.git
git clone https://github.com/introlab/find-object.git
git clone https://github.com/monkeyrom/CR5_Project.git
cd $HOME/catkin_ws
building
catkin_make
set the dobot type
echo "export DOBOT_TYPE=cr5" >> ~/.bashrc
source ~/.bashrc
source $HOME/catkin_ws/devel/setup.bash
1. Launch Project
- Connect the robotic arm with following command, and 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

find object GUI

2. Add object image for detection
- Using find object gui for adding image
- 
edit
 - 
add object from scene
 - 
take picture
 - 
crop object
 
 - 
 

tf synchronize

3. Run a terminal for controlling robot
    rosrun CR5_Project service_call.launch
- this command will run 2 nodes and spawn new terminal for commanding
- service_call
 - main_control
 
 
new terminal for input command

Real Robotic Arm
Dobot CR5

Intel Realsense D435i

References
- CR_ROS: https://github.com/Dobot-Arm/CR_ROS
 - find-object: https://github.com/introlab/find-object