- Notifications
You must be signed in to change notification settings - Fork4
Python binding for Grasp Pose Generator (pyGPG)
NotificationsYou must be signed in to change notification settings
lianghongzhuo/pygpg
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
Grasp Pose Generator is a cpp package that creat grasp candidates for 3D point clouds.This package binding it with python.
conda install -c conda-forge pcl cmake boost./build_pygpg.sh
- Note, if the conda failed to find the correct package to install, you can try installingmamba and using mamba to install the dependencies:
mamba install pcl cmake boost
importnumpyasnpimportpygpgpoints=np.random.rand(3000,3)# put your point cloud here, should be a nX3 numpy array, here is an example random arraynum_samples=10000show_grasp=Falsegripper_config_file="PATH_TO_GRIPPER_CONFIG_FILE"# see gripper_params.cfg for an examplegrasps=pygpg.generate_grasps(points,num_samples,show_grasp,gripper_config_file)# grasps is a list of grasp objet, to construct a Transformation matrix from each grasp object, use:pose_list= []forgraspingrasps:pose=np.eye(4)pose[:3,0]=grasp.get_grasp_approach()pose[:3,1]=grasp.get_grasp_binormal()pose[:3,2]=grasp.get_grasp_axis()pose[:3,3]=grasp.get_grasp_bottom()pose_list.append(pose)
python example/example.py
PressR to see the result:
If you found pyGPG useful in your research, please consider citing:
@software{pygpg, author = {Hongzhuo Liang}, title = {Python binding for Grasp Pose Generator (pyGPG)}, month = Aug, year = 2021, doi = {10.5281/zenodo.5247189}, url = {https://doi.org/10.5281/zenodo.5247189}}About
Python binding for Grasp Pose Generator (pyGPG)
Topics
Resources
Stars
Watchers
Forks
Packages0
No packages published