Gazebo Math

API Reference

8.0.0~pre1
Python Get Started

Previous Tutorial: C++ Get Started

Overview

This tutorial describes how to get started using Gazebo Math with Python.

NOTE: If you have compiled Gazebo Math from source, you should export your PYTHONPATH.

export PYTHONPATH=$PYTHONPATH:<path to your workspace>/install/lib/python

We will run through an example that determines the distance between two points in 3D space. Start by creating a bare-bones main file using the editor of your choice.

def main():
pass
if __name__ == "__main__":
main()

The easiest way to include Gazebo Math is through import gz.math8.

At this point your main file should look like

import gz.math8
def main():
pass
if __name__ == "__main__":
main()

Now let's create two 3D points with arbitrary values. We will use the gz.math.Vector3 class to represent these points. Gazebo Math provides some Vector3 types which are: Vector3d (Vector3 using doubles), Vector3f (Vector3 using floats) and Vector3i (Vector3 using integers). The result of this addition will be a main file similar to the following.

from gz.math8 import Vector3d
def main():
point1 = Vector3d(1, 3, 5)
point2 = Vector3d(2, 4, 6)
if __name__ == "__main__":
main()

Finally, we can compute the distance between point1 and point2 using the gz.math.Vector3.distance() function and output the distance value.

from gz.math8 import Vector3d
def main():
point1 = Vector3d(1, 3, 5)
point2 = Vector3d(2, 4, 6)
distance = point1.distance(point2);
print("Distance from {} to {} is {}".format(point1, point2, distance))
if __name__ == "__main__":
main()

Running this program should result in the following output:

Distance from 1 3 5 to 2 4 6 is 1.7320508075688772