Implementation of object classifier conveyor system in ROS2-Gazebo.

system ros node graph

SETUP

• mkdir -p dev_ws/src

• cd src

• git clone this repo

• install ros humble desktop https://docs.ros.org/en/humble/Installation/Ubuntu-Install-Debians.html

• install ros humble

sudo apt install -y ros-humble-desktop 
sudo apt install python3-colcon-common-extensions

• add this to .bashrc

#underlay ros2 setup
source /opt/ros/humble/setup.bash

#colcon cd
source /usr/share/colcon_cd/function/colcon_cd-argcomplete.bash

#colcon argcomplete
source /usr/share/colcon_argcomplete/hook/colcon-argcomplete.bash

#source local bashrc
# if [ !( ./ -ef ~ ) ]; then
if [ "$(realpath $PWD)" != "$(realpath $HOME)" ]; then
# echo $PWD
source $PWD/.bashrc
source install/setup.bash
fi

• open a new terminal in dev_ws directory

sudo apt install python3-rosdep
sudo pip3 install transforms3d
sudo rosdep init
rosdep update
rosdep install -i --from-path src --rosdistro humble -y

BUILD

• build the workspace from dev_ws directory

colcon build

RUN

• open new terminal from dev_ws or source dev_ws/install/setup.bash

1. To setup the sim world

ros2 launch conveyorbelt_gazebo conveyorbelt.launch.py

2. [To start the belt] again open a new terminal as above to source as mentioned above in a new terminal

ros2 service call /CONVEYORPOWER conveyorbelt_msgs/srv/ConveyorBeltControl "{power: 10, belt_changer_velocity: 0.5}"

3. [To start spawn boxes] again open new terminal as above

ros2 run ros2_conveyorbelt SpawnObject.py

DEMO