An OpenAI gym wrapper for CARLA simulator
An OpenAI gym third party environment for CARLA simulator.
$ conda create -n env_name python=3.6
$ conda activate env_name
$ git clone https://github.com/cjy1992/gym-carla.git
$ pip install -r requirements.txt
$ pip install -e .
PYTHONPATH
environment variable:$ export PYTHONPATH=$PYTHONPATH:$YourFolder$/CARLA_0.9.6/PythonAPI/carla/dist/carla-0.9.6-py3.5-linux-x86_64.egg
$ ./CarlaUE4.sh -windowed -carla-port=2000
You can use Alt+F1
to get back your mouse control.
Or you can run in non-display mode by:
$ DISPLAY= ./CarlaUE4.sh -opengl -carla-port=2000
$ python test.py
See details of test.py
about how to use the CARLA gym wrapper.
The termination condition is either the ego vehicle collides, runs out of lane, reaches a destination, or reaches the maximum episode timesteps. Users may modify function _terminal in carla_env.py to enable customized termination condition.
The reward is a weighted combination of longitudinal speed and penalties for collision, exceeding maximum speed, out of lane, large steering and large lateral accleration. Users may modify function _get_reward in carla_env.py to enable customized reward function.