ikfast-pybind-tmp 0.1.2

ikfast_pybind is a python binding generation library for the analytic kinematics engine ikfast.

ikfast_pybind is a python binding generation library for the analytic kinematics engine IKfast. The python bindings are generated via pybind11 a CMake-based build system.

Note: You need the ikfast .h and .cpp ready to generate the python bindings. This URDF-to-cpp generation part needs to be done with openrave and IS NOT done by this repo, please see this tutorial for details.

The assembly sequence and motion planning framework pychoreo relies on this library to generate compatible IK modules for robots across brands, scales, and dofs.

Prerequisites

ikfast_pybind depends on the following dependencies, which come from pybind11 for building the python bindings.

On Unix (Linux, OS X)

• A compiler with C++11 support

• CMake >= 2.8.12

On Windows

• Visual Studio 2015 (required for all Python versions, see notes below)

• CMake >= 3.1

It is recommended (especially for Windows users) to test the environment with the cmake_example for pybind11 before proceeding to build conmech.

Installation

git clone --recursive https://github.com/yijiangh/ikfast_pybind
cd ikfast_pybind
pip install .
# try with '--user' if you encountered a sudo problem

For developers:

git clone --recursive https://github.com/yijiangh/ikfast_pybind
cd ikfast_pybind
python setup.py sdist
pip install --verbose dist/*.tar.gz

With the setup.py file included in the base folder, the pip install command will invoke CMake and build the pybind11 module as specified in CMakeLists.txt.

References

Citation

If you find IKFast useful, please cite OpenRave:

@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},
}

Related links

tutorial on ikfast cpp generation: See this tutorial for a detailed instruction on how to generate the ikfast cpp code from an URDF.

Testing ikfast modules with a pick-n-place demo in pybullet

pybind11

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog and this project adheres to Semantic Versioning.

0.1.1

Added - Added support for kawasaki_rs010n robot

Changed - Reorganize test files to having a test file for each robot type.

Updated - pybind11 set to track master, commit e08a58111, which should fix pip installation issue.

0.1.0

Available robots - kuka_kr6_r900 (tested) - ur3 - ur5 - abb_irb4600_40_255 - franka_panda (tested) - eth_rfl (tested)

Added Modules for franka_panda, eth_rfl robots.

Add ifkast modules for ur5, kuka_kr6_r900, abb_irb4600. abb_irb4600 test fails some time randomly - need to regenerate its IKfast cpp files (might be the floating point truncation issue).

Include the upstreamed ur_kinematics commit 6734142 July 2 2019 but it seems that the old one works more stably… I will do more tests on this.