Setup

This package is being tested with:

• Python 3.10 in Ubuntu 22.04

• Optionally with ROS 2 Humble and Iron

pip install (Limited)

You can quickly install pyrobosim through PyPi. However, note that this will not include any of the ROS 2 or Task and Motion Planning functionality.

pip install pyrobosim

Local Setup

If using ROS 2, clone this repo in a valid colcon workspace. Otherwise, if running standalone, clone it wherever you would like.

To set up your Python virtual environment, configure and run

./setup/create_python_env.bash

By default, this will create a Python virtual environment in ~/python-virtualenvs/pyrobosim.

If you want to use PDDLStream for task and motion planning, you should also run:

./setup/setup_pddlstream.bash

To then source this virtual environment, run

source ./setup/source_pyrobosim.bash

As documented in the above script, we recommend making a bash function in your ~/.bashrc script so you can easily just call pyrobosim from your Terminal to get started.

pyrobosim() {
   source /path/to/pyrobosim/setup/source_pyrobosim.bash
}

Additional ROS 2 Setup

After you have installed pyrobosim and activated your Python virtual environment, you must build your colcon workspace to install the pyrobosim_msgs and pyrobosim_ros packages. For example, if you have cloned this repo to ~/pyrobosim_ws/src/pyrobosim, you can do:

cd ~/pyrobosim_ws
colcon build
. install/local_setup.bash

For ROS 2 workflows, you can also make bash function to get set up like this:

pyrobosim_ros() {
   source /path/to/pyrobosim/setup/source_pyrobosim.bash humble
}

The additional humble argument will make sure that ROS 2 Humble and your built colcon workspace are sourced in addition to activating the Python virtual environment.

Docker Setup

We also provide Docker images compatible with ROS 2 releases.

If you already have sourced ROS 2 in your system (e.g., source /opt/ros/humble/setup.bash), then you should have a ROS_DISTRO environment variable set. Otherwise,

export ROS_DISTRO=humble
docker compose build

Then, you can start any of the existing demos. To see all the available services, refer to the docker-compose.yaml file or use Tab completion.

docker compose run demo
docker compose run demo_ros

Alternatively, you can start a Docker container.

docker compose run base
# python3 src/pyrobosim/examples/demo.py

You can also start a new Terminal and go into the same container as follows.

docker exec -it <pyrobosim-base-1> bash
# ros2 launch pyrobosim_ros demo.py

The source code on your host machine is mounted as a volume, so you can make modifications on your host and rebuild the Colcon workspace inside the container.