The main ROSPlan website and documentation is available here.
The ROSPlan framework provides a generic method for task planning in a ROS system. ROSPlan encapsulates both planning and dispatch. It provides with a simple interface, and already includes interfaces to common ROS libraries.
Several demos are available in the rosplan_demos repository. More will be added over time.
ROSPlan supports kinetic and melodic ROS distributions and is developed under Ubuntu 16.04 / 18.04, being linux Ubuntu 18.04 + melodic the recommended setup.
Install dependencies:
sudo apt install flex bison freeglut3-dev libbdd-dev python-catkin-tools ros-$ROS_DISTRO-tf2-bulletSelect or create a catkin workspace:
mkdir -p ROSPlan/src
cd ROSPlan/Get the code:
cd src/
git clone https://github.com/KCL-Planning/rosplanCompile everything:
catkin buildThis demo has been migrated to a separate repo, please follow up this link: rosplan_demos
Get your docker image from the Docker Hub:
docker pull kclplanning/rosplan
Run an interactive bash shell inside the docker, ready to run ROSPlan:
docker run -it --rm kclplanning/rosplan bash
There is a newly developed interface to interact with ROSPlan's Knowledge Base. It was built by Eden Jia, using tkinter in Python. You can find it and use it here.
-
Automatic localisation and docking action interfaces with the Turtlebot 2 (Kobuki base) https://github.com/KCL-Planning/ROSPlan_interface_Turtlebot2
-
Integration with the Component Oriented Layered-base Architecture for Autonomy (COLA2). Developed in the Research Center of Underwater Robotics (CIRS) in the University of Girona (UdG). This architecture is used to control the Autonomous Underwater Vehicles (AUVs) developed in this center. (https://bitbucket.org/udg_cirs/cola2) https://github.com/KCL-Planning/ROSPlan_interface_COLA2
-
Action interfaces for piloting a quadrotor from Jindrich Vodrazka, (takeoff, land, fly_square, and fly_waypoint). https://github.com/fairf4x/ROSPlan_interface_quadrotor