The MRPT project

1. Introduction

Mobile Robot Programming Toolkit (MRPT) includes libraries and tools aimed to help researchers in the areas of mobile robotics and computer vision in the development of efficient implementations with reusability of code as a priority. It features classes for easily managing 3D(6D) geometry, probability density functions (pdfs) over many predefined variables (points,landmarks,poses,maps,...), Bayesian inference (Kalman filters, particle filters), image processing, obstacle avoidance, etc.

2. Resources

• Main website http://www.mrpt.org/

• C++ API reference: http://reference.mrpt.org/.

• Google group for questions: http://www.mrpt.org/forum/

• Download the latest unstable code with:

  • Git:

      git clone https://github.com/jlblancoc/mrpt.git mrpt-git
    

  • Subversion:

      svn co https://github.com/jlblancoc/mrpt/trunk/ mrpt-svn
    

• Example configuration files for MRPT applications can be found at: MRPT/share/mrpt/config_files

• Some sample datasets are stored in: MRPT/share/mrpt/datasets. A more complete dataset repository is available online.

3. Compiling

3.1 FIVE SECONDS INSTRUCTIONS

1. Invoke cmake-gui and select:

   • Where source is --> MRPT source root directory
   • Where to build binaries --> Pick a new, empty directory.

   If your platform doesn't support cmake-gui, open a console, chdir to a new empty directory and execute:

   $ ccmake <PATH_TO_MRPT_SOURCES>
   

2. Within cmake-gui (or ccmake), set all the build options as you desire, then press "Configure" and "Generate". To build the examples, enable "BUILD_SAMPLES".

3. Build with the IDE / compiler you selected from CMake (Visual Studio, GNU Make,...)

3.2. Further details (RECOMMENDED!)

Read the compiling instructions.

4. License

MRPT is released under the new BSD license.