URDF and MoveIt configuration (ROS1) files of Gripkit-CR-Plus-L and KRM-iiwa 7. Depending on user's inputs (e.g., end_effector, mobile_base, and controllers), the URDF and MoveIt reconfigure themselves. The repository also provides a simple Gazebo environment and Python classes and functions to experiment object grasping.
The project was done while Hojun Lee was working for Barton Research Group (BRG) at the University of Michigan.
- Repository Structure
- Download Process
- Simulation
- Automated Grasping
- Potential Extensions
- Issues
- ToDo Lists
├── auto_grasp
│ └── src # Python source codes
├── gripkit_cr_plus_l_ad_description
├── gripkit_cr_plus_l_ad_description
│ ├── meshes # STL files
| └── urdf # URDF description
├── gripkit_cr_plus_l_bb_description
├── gripkit_cr_plus_l_cc_description
├── gripkit_cr_plus_l_dd_description
├── iiwa7_description
├── images
├── kmp200_description
├── kmriiwa_description
├── kmriiwa_moveit
│ ├── config # config files
│ └── launch # ROS-launch files
├── models
│ ├── dict # grasp dictionaries
│ ├── marker # waypoint marker
│ └── obj_05 # object folder
└── worlds
Note This repository has been tested on ROS Noetic and Ubuntu 20.04. It also depends on numpy, scipy, and tqdm:
cd ~/catkin_ws/src
git clone https://github.com/kidpaul94/kmr-iiwa-gripkit-cr-plus-l.git
cd kmr-iiwa-gripkit-cr-plus-l/
pip3 install -r requirements.txt
cd ~/catkin_ws
catkin_make
source devel/setup.bash
Note This launch file is only for model visualization purpose. No ROS controllers will get initialized.
roslaunch kmriiwa_description gazebo.launch robot_name:=(choose iiwa or kmriiwa) hardware_interface:=(choose Position, Velocity, or Effort) ee_type:=(choose ad, bb, cc, or dd)
roslaunch kmriiwa_moveit demo.launch robot_name:=(choose iiwa or kmriiwa) hardware_interface:=(choose Position, Velocity, or Effort) ee_type:=(choose ad, bb, cc, or dd)
roslaunch kmriiwa_moveit demo_gazebo.launch robot_name:=(choose iiwa or kmriiwa) hardware_interface:=(choose Position, Velocity, or Effort) ee_type:=(choose ad, bb, cc, or dd)
Note
simulation.py
receives several different arguments. Run the--help
command to see everything it receives.
cd auto_grasp/src
python3 simulation.py --help
The current implementation of the reposity supports top-down grasp, cartesian_space, and joint_space planning. The cartesian_space and joint_space plannings receives list of ROS pose message and joint configuration that a robot model needs to follow. For top-down grasp, it only supports grasping center & direction format at the moment. More details about this format is described below.
Note The representation is defined based on a coordinate system of each object 3D model.
The image above shows the grasping center & direction format that the top-down grasp function uses. In simulation.py
, we provide a list of this representation as a grasp dictionary associated with a specific object model.
To run your own world with other objects, simply put them in the models and worlds folders. You can also swap iiwa 7 with iiwa 14. Let me know if there are additional extensions & features you want to see in this project (e.g., adding more robot models). Contribution is always welcome!
Warning MoveIt trajectory execution sometimes fails without any error or warning. At the moment, the program skips the trial by checking whether the gripper reaches to a final grasping configuration. More stable trajectory planning/execution can be achieved by swaping KDL Kinematics Solver (default) with IKFast Kinematics Solver or TRAC-IK Kinematics Solver.
Model & Controller parameters tuning | |
---|---|
Automation of grasping |