Git Product home page Git Product logo

cambyse / po-lgp Goto Github PK

View Code? Open in Web Editor NEW
1.0 1.0 0.0 443.7 MB

Trajectory-Tree optimization based on KOMO

License: GNU General Public License v3.0

Shell 1.64% Makefile 1.03% C++ 63.50% Gnuplot 0.05% C 13.27% Forth 0.03% TeX 2.26% Python 1.73% GAP 2.39% MATLAB 0.52% HTML 2.40% CSS 0.04% G-code 0.01% Jupyter Notebook 9.31% GDB 0.01% SWIG 0.10% CMake 0.36% PDDL 0.02% Vim Snippet 0.03% Vim Script 1.31%

po-lgp's Introduction

NOTE: if installing on Ubuntu 16.04, please see README.1604.md

MLR Repository

  • Infos for students: README.student.md
  • Infos about the PR2: README.PR2.md

Quick Start on Ubuntu 16.04

Install git:

$ sudo apt-get install git

Add your ssh key to clone the repository.

User->Settings->SSH Keys

Clone the repo:

$ cd ~; mkdir git; cd git
$ git clone [email protected]:mlr-staff/mlr.git

Once the repo has been cloned:

$ cd mlr
$ git checkout roopi
$ ./install/INSTALL_ALL_UBUNTU_PACKAGES.sh
$ ./install/INSTALL_ROS_PACKAGES_KINETIC
$ ./share/bin/createMakefileLinks.sh
$ cp share/build/config.mk.kinetic share/build/config.mk
$ cd share/examples/Roopi/basics/
$ make

Install on Ubuntu 14.04

Install the mlr system by executing:

./install/INSTALL_ALL_UBUNTU_PACKAGES.sh

The 'install/INSTALL' script does some more local installation of packages. But this is usually not necessary!

To use the gof (good old-fashioned) Makefiles,

cd share; bin/createMakefileLinks.sh

Also,

cd share/gofMake; cp config.mk.default config.mk

and edit the latter to change the compile settings, e.g., (un)set dependencies.

Install on Arch

What to do to install MLR on Arch Linux?

First install all dependencies. You need some things from AUR, so first install yaourt or something similar to handle AUR packages.

$ yaourt -S opencv plib ann lapack blas qhull

In order to get a working cblas library we need atlas-lapack, but be warned. This literally will take hours of compile time (something like 5-10 h depending on your system).

$ yaourt -S atlas-lapack

(You can either drink a lot of coffee in the meantime or put LAPACK = 0 in your gofMake/config.mk until it is finished...)

MLR needs a .so-file called libANN, ann provides one, but names it libann so we link it:

$ sudo ln -s /usr/lib/libANN.so /usr/lib/libann.so
$ sudo ldconfig

qhull is newer in arch than in ubuntu and changes some pointer/reference things. You need to put

ARCH_LINUX = 1

in your gofMake/config.mk. This fixes the qhull change.

Everything else is equal to ubuntu.

Install troubleshooting

  • If there is a segmentation fault in c_init (before main) in some boost library, maybe this is caused by linking to PCL in the src/Perception code. Fix: in share/gofMake/config.mk insert a line 'PCL=0', which disables linking to PCL.

QtCreator tips

  • Enable Debugging helpers: ** OLD WAY: .gdbinit -> git/mlr/tools/qt_mlr_types.py ** NEW WAY: in Options -> Debugger -> GDB -> Extra Dumpers add 'git/mlr/tools/qt_mlr_types.py' In this file ther must not be a 'python' and 'end' flags!

  • Enable std::... pretty printing: Options > Debugger > GDB ... UNCHECK the 'Load system GDB pretty printers'

  • set formatting: Options > C++ > Import... git/mlr/tools/qt_coding_style.xml

  • Change signal handling of gdb Open 'Debugger Log' window write gdb commands: handle SIGINT pass handle SIGINT noprint handle SIGINT nostop handle SIGINT ignore

  • append source directory to all ".includes" files: echo "echo '/home/mtoussai/git/mlr/share/src' >> $1" > nogit_append_path chmod a+x nogit_append_path find . -name '..includes' -exec ./nogit_append_path {} ;

Getting started

The best way to learn about this code and understand what is can do it to go through all examples. And take a look at the Doxygen documentation http://sully.informatik.uni-stuttgart.de:8080/job/MLR-share-doxygen/doxygen/

For the examples start with

  • examples/Core
  • examples/Gui
  • examples/Ors

All of them should work. Going through the main should be instructive on how you could use the features. In examples/Ors some demos depend on linking to Physx which is turned off on default.

Continue with more specialized packages:

  • examples/Optim
  • examples/Motion
  • examples/Algo
  • examples/Infer

which implement algorithmic methods.

Finally examples/Hardware contains drivers/recorders for Kinect, Cams, G4.

examples/System is code for designing parallized modular systems (using Core/module.h) - which I think works robustly but needs more tools for others to be usable.

examples/relational is Tobias' relational RL code

Documentation

Create the doxygen documentation with:

cd share
make doc

WINDOWS

Large parts of the code compile with MSVC. Try this:

download https://ipvs.informatik.uni-stuttgart.de/mlr/marc/source-code/libRoboticsCourse.13b.msvc.tgz

goto http://www.microsoft.com/en-us/download/developer-tools.aspx

search for Visual C++ 2010 Express

download and install

open the 'solution' msvc/RoboticsCourse.sln

Creating a new project is tricky! You could avoid it by copying new code into an existing project, replacing the main.cpp (e.g. in example\Ors\ors). But in case you need to:

When creating new projects: -- set the MT_MSVC preprocessor flag (perhaps also MT_FREEGLUT, MT_ANN, MT_GL, MT_QHULL) -- set the Debugger work directory to ....\share\example\Ors\ors or alike -- set an Additional Includepath to ....\share\src -- perhaps set the VC++ Directory 'libpath' to ....\msvc\freeglut\lib -- add the created lib-files in msvc\Debug to the project (drag and shift-drop from others) -- understand, that programs are by default run in msvc\Debug -- perhaps you need to copy dlls around - (don't know how to set the path globally)

Run on Virtual Machines

Setting up this software to run on a virtual machine is an option for whoever can't make it work on his OS. The biggest issue of this approach is getting the OpenGL rendering to work.

Instructions:

  1. Install the most recent version of virtualbox for your OS. (don't use the package repository, use https://www.virtualbox.org/wiki/Downloads).

  2. Download the ISO for the desktop-amd64 version of Ubuntu (12.04 is recommended, more recent versions may also work) and install it on a new VM.

  3. run: sudo apt-get update sudo apt-get upgrade

  4. [Optional] If the above software updates are slow, close the VM and change the network setting "Attached to: Bridged Adapter".

  5. [Optional] Step 6 is said to require the linux-headers of the current version. Nonetheless, I've made everything work already before without this step.

run: sudo apt-get install build-essential linux-headers-$(uname -r)

  1. Select Devices -> Insert Guest Additions CD Image. Then run the autorun script in the mounted CD drive. This installs the virtualbox-guest-addition software specific for your virtualbox version (don't use package repository version). Check that everything is installed correctly.

  2. If there were no problems with the installation, unmount the CD drive and reboot the VM.

  3. Proceed installing the MLR code base.

Tips)

Take frequent snapshots of the VM, whenever you're in a state where it is working correctly (this also helps during the above setup procedure). If things break at some point, transfer the code you've written (either using some versioning tool, or a USB stick) to the most recent snapshot of the VM and proceed working.

MISC

  • Installing portaudio19-dev: Currently has broken links in ubuntu. Solution: first install libjack-jackd2-dev

po-lgp's People

Contributors

hildensia avatar sotte avatar robert-lieck avatar cambyse avatar roboloni avatar abaisero avatar etpr avatar cinhcet avatar vienngo avatar controltrees avatar stfnsky avatar ricoj avatar boehmrn avatar aorthey avatar buschbapti avatar cmapyopc avatar thomas-gruber avatar matthew-bernstein avatar

Stargazers

 avatar

Watchers

 avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.