Skip to content

ika-rwth-aachen/carla-scenario-runner

 
 

Repository files navigation

Control Actor: carla-scenario-runner

Important

This repository is a minimal fork of the official scenario_runner! All initial and following modifications to the original repository are documented in CARLOS_CHANGELOG.md.

Tip

We recommend to use the carla-scenario-runner as control actor in our open, modular and scalable simulation framework CARLOS.

The carla-scenario-runner is used to enable scenario-based testing and evaluation. It is a powerful engine that follows the OpenSCENARIO standard for scenario definitions.

Note

We set up a Continous Integration (CI) pipeline as GitHub workflow to continously build Docker images for the carla-scenario-runner, publicly available on Docker Hub.



Original README

License: MIT GitHub tag (latest SemVer) Build Status

This repository contains traffic scenario definition and an execution engine for CARLA. It also allows the execution of a simulation of the CARLA Challenge. You can use this system to prepare your agent for the CARLA Challenge.

Scenarios can be defined through a Python interface, and with the newest version the scenario_runner also the upcoming OpenSCENARIO standard and OpenSCENARIO 2.0 standard is supported.

Scenario_Runner for CARLA

Getting the ScenarioRunner

Use git clone or download the project from this page. Note that the master branch contains the latest fixes and features, and may be required to use the latest features from CARLA.

It is important to also consider the release version that has to match the CARLA version.

To use a particular version you can either download the corresponding tarball or simply checkout the version tag associated to the release (e.g. git checkout v0.9.5)

Currently no build is required, as all code is in Python.

Using the ScenarioRunner

Please take a look at our Getting started documentation.

Challenge Evaluation

The CARLA Challenge has moved to the CARLA Autonomous Driving Leaderboard. Please see the leaderboard repository and the getting started guide for more information.

Contributing

Please take a look at our Contribution guidelines.

FAQ

If you run into problems, check our FAQ.

License

ScenarioRunner specific code is distributed under MIT License.

Packages

No packages published

Languages

  • Python 85.7%
  • Java 9.6%
  • XSLT 3.7%
  • Other 1.0%