/kinodyn

Primary LanguagePython

python packages needed:

  * python-control http://sourceforge.net/apps/mediawiki/python-control/index.php?title=Main_Page
    wget http://sourceforge.net/projects/python-control/files/control-0.5b.tar.gz
    tar -xvf control-0.5b.tar.gz
    cd control-0.5b
    python setup.py install

    * python-control depends on slycot (Python bindings to slicot. slicot requires a fortran compiler)
      git clone git://github.com/avventi/Slycot.git
      cd Slycot
      python setup.py install

  * NetworkX http://networkx.lanl.gov/
    apt-get install python-networkx

  * shapely (can be easily avoided) https://github.com/sgillies/shapely
    apt-get install python-shapely
    * have installed libgeos for speedup
      apt-get install libgeos-dev
    * sudo pip install descartes

  * cvxopt 
    Won't install with pip or easy_install since setup.py is in the wrong place. 
    Still, use pip to download the build files, navigate to python/build/cvxopt/src and run python setup.py install
    (or try apt-get install python-cvxopt)

  * SciPy/NumPy/Matplotlib
    apt-get install python-matplotlib

  * scikit-learn
    pip install -U scikit-learn (only use for nearest-neighbor queries in Euclidean distance -- for drawing Voronoi bias in RRT example

not needed, but this is what I use to run python
  * IPython
    apt-get install ipython
    
    (to use IPython's graphical console if you have problems)
    Installing PyQt: http://problemssol.blogspot.com/2010/12/compile-and-install-pyqt4-for-python27.html
    need libpyside-dev package to install for python2.7 in 10.04:
    
    https://launchpad.net/~pyside/+archive/ppa 
    (there is a nicer way to do this. just follow the instuctions on the ppa website)
    * add the sources to /etc/apt/source.list
    * sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys 073700C1
    * sudo apt-get update
    
    (still not sure how to get libpyside-python2.7.so)

to run an example:
git clone git://github.com/goretkin/kinodyn.git

bash> cd kinodyn
bash> ipython --pylab
ipython> %run examplets/rrt_2d_example.py
ipython> plt.show()

click on the plot to force extensions