commonroad-drivability-checker 2022.2

Drivability checker for CommonRoad scenarios.

CommonRoad Drivability Checker

Collision avoidance, kinematic feasibility, and road-compliance must be validated to ensure the drivability of planned motions for autonomous vehicles. The CommonRoad Drivability Checker toolbox unifies these checks in order to simplify the development and validation of motion planning algorithms. It is compatible with the CommonRoad benchmark suite, which additionally facilitates and drastically reduces the effort of the development of motion planning algorithms.

Please post questions, bug reports, etc. related to our tools or website in our forum.

System Requirements

The software is written in Python 3.6/3.7 and C++11 and tested on MacOS and Linux. The usage of the Anaconda Python distribution is strongly recommended. For building the code, the following minimum versions are required:

• GCC and G++: version 9 or above
• CMake: version 3.10 or above.
• Pip: version 21.3 or above

If you are a Mac user, we additionally recommend you to use Homebrew, allowing you to install required dependencies such as Eigen.

Third Party Libraries and Packages

The C++ code depends on the following libraries:

Essential dependencies:

Not included as submodules:

• Eigen3
• Boost
• OpenMP

Included as submodules:

• Box2D
• FCL
• libccd
• gpc
• pybind11
• libs11n

Optional dependencies:

• Triangle (optional library for triangulation, not built by default)
• CGAL (optional library for triangulation, not built by default)
• Pandoc (optional for building documentation)
• Doxygen (optional for building documentation)

Note: Please be aware of the specific licensing conditions when including the optional dependencies Triangle and CGAL. See also notes.txt and the licensing information on the respective package websites for more details.

The Python dependencies are listed in requirements.txt.

Installation

To install the Python package of the drivability checker, please run:

pip install commonroad-drivability-checker

To build the drivability checker package from source, please refer to the installation description in the documentation.

Documentation

A full documentation as well as tutorials to get started with the tool can be found on our toolpage.

Publication

CommonRoad Drivability Checker: Simplifying the Development and Validation of Motion Planning Algorithms

Christian Pek, Vitaliy Rusinov, Stefanie Manzinger, Murat Can Üste, and Matthias Althoff

Abstract— Collision avoidance, kinematic feasibility, and road-compliance must be validated to ensure the drivability of planned motions for autonomous vehicles. Although these tasks are highly repetitive, computationally efficient toolboxes are still unavailable. The CommonRoad Drivability Checker—an open-source toolbox—unifies these mentioned checks. It is compatible with the CommonRoad benchmark suite, which additionally facilitates the development of motion planners. Our toolbox drastically reduces the effort of developing and validating motion planning algorithms. Numerical experiments show that our toolbox is real-time capable and can be used in real test vehicles.

Fulltext is available on mediaTUM.

Bibtex:

@inproceedings{ PekIV20.pdf,
	author = "Christian Pek, Vitaliy Rusinov, Stefanie Manzinger, Murat Can Üste, and Matthias Althoff",
	title = "CommonRoad Drivability Checker: Simplifying the Development and Validation of Motion Planning Algorithms",
	pages = "1-8",
	booktitle = "Proc. of the IEEE Intelligent Vehicles Symposium",
	year = "2020",
	abstract = "Collision avoidance, kinematic feasibility, and road-compliance must be validated to ensure the drivability of planned motions for autonomous vehicles. Although these tasks are highly repetitive, computationally efficient toolboxes are still unavailable. The CommonRoad Drivability Checker— an open-source toolbox—unifies these mentioned checks. It is compatible with the CommonRoad benchmark suite, which additionally facilitates the development of motion planners. Our toolbox drastically reduces the effort of developing and validating motion planning algorithms. Numerical experiments show that our toolbox is real-time capable and can be used in real test vehicles."
}