Getting Started with OpenRAVE environment. Guidance could be followed at Stéphane Caron's blog.

Error may be frequently encountered during compile. Alternatively, docker image is recommended.

Check personalrobotics/ros-openrave

sudo docker pull personalrobotics/ros-openrave:latest

Since the image is pre-built, no need to docker build with Dockerfile. Just run it. Mirror the /root/data to your current workspace.

sudo docker run -it --rm -v /home/dingyi/workspace/dingyi-code/side-project/openrave-test:/root/data --network host ada6b60e98b7 bash

Where image id ada6b60e98b7 could be obtained using sudo docker images.

Now go to /root/data in the docker image will direct to the same workspace.

xmlfile like ur5.robot.xml represents the robot model. Use ikfastpy to generate the ikfast solver.

python `openrave-config --python-dir`/openravepy/_openravepy_/ikfast.py --robot=ur5.robot.xml --iktype=transform6d --baselink=0 --eelink=6 --savefile=ikfast61.cpp --maxcasedepth 1

It takes several minutes to generate the closed-form ikfast solver, which is a cpp file named by you. Then ikfast61.cpp is shown in your workspace.

Now you can exit the docker image.

Compile the ikfast solver using gcc(need -llapack).

gcc -fPIC -lstdc   -DIKFAST_NO_MAIN -DIKFAST_CLIBRARY -shared -Wl,-soname,libik.so -o libik.so ikfast61.cpp

libik.so comes out in your workspace.

Compile the executable demo for testing purpose.

g   ikfastdemo.cpp -lstdc   -llapack -o compute -lrt

Run compute for easy tests for fk and ik.

Verify your results with the real robot.

For more information and to explore other interesting projects, please refer to the following research papers:

@phdthesis{diankov_thesis,
  author = "Rosen Diankov",
  title = "Automated Construction of Robotic Manipulation Programs",
  school = "Carnegie Mellon University, Robotics Institute",
  month = "August",
  year = "2010",
  number= "CMU-RI-TR-10-29",
  url={http://www.programmingvision.com/rosen_diankov_thesis.pdf},
}

@inproceedings{zeng2018learning,
title={Learning synergies between pushing and grasping with self-supervised deep reinforcement learning},
author={Zeng, Andy and Song, Shuran and Welker, Stefan and Lee, Johnny and Rodriguez, Alberto and Funkhouser, Thomas},
booktitle={2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
pages={4238--4245},
year={2018},
organization={IEEE}
}