Motion planning consists of finding a path from a state A to a goal state B while avoiding obstacles.
There is a wide range of motion planning libraries that focus mostly on motion planning problems formulated in the configuration space OMPL. This approach is usually based on inverse kinematics to transform real-world goals into suitable configurations.
This repository formulates a generic motion planning scene, including both moving and static obstacles and a generic formulation of goals.
Obstacles are split into dynamic and static obstacles. Dynamic obstacles are further split into movable and fixed obstacles. Movable obstacles can be pushed by robots while fixed ones act as walls or static parts of the environment.
Goals for motion planning should not depend on the robot's structure, neither should they involve orientation that are generally hard to obtain and hardly human understandable.
I have decided to change the name of the package on pypi to mpscenes
as it
seems more convenient a name. Early versions were published under the name
motion_planning_scenes
.
pip3 install .
or
pip3 install mpscenes
This package is build using poetry. Poetry generates a virtual environment automatically. If you have poetry installed you can test it through:
poetry install
poetry shell
Then you are in a virtual environment in which you can test this specific package. Read more about poetry on the documentation website.