Spatial Math for ROS.
Intergration library between rospy and spatialmath-python.
Currently this lib just contains conversion functionality.
Expect the conversion modules to work in any ROS1 version.
pip install spatialmath-rospy# These classes extend their spatialmath-python counterparts
# to provide ROS functionality
from spatialmath_rospy import SE3, SO3, UnitQuaternion
pose_msg = SE3(1, 2, 3).to_ros()
print(type(pose_msg), '\n', pose_msg)
"""
<class 'geometry_msgs.msg._Pose.Pose'>
position:
x: 1.0
y: 2.0
z: 3.0
orientation:
x: 0.0
y: 0.0
z: 0.0
w: 1.0
"""
se3: SE3 = SE3.from_ros(pose_msg)
print(se3)
"""
1 0 0 1
0 1 0 2
0 0 1 3
0 0 0 1
"""For those who prefer a functional style or don't want to use the extended classes
import spatialmath as sm
from spatialmath_rospy import to_spatialmath, to_ros
pose_msg = to_ros(sm.SE3(1, 2, 3))
se3: sm.SE3 = to_spatialmath(pose_msg)[Not Recommended]
You may prefer to use this option if wanting to add the .from_ros() and .to_ros() methods to the original SE3, SO3 and UnitQuaternion classes via a monkey-patch. This may be useful for integrating legacy code. Not recommended as static type analysis tools like PyLance will not work.
import spatialmath as sm
from spatialmath_rospy import monkey_patch_spatialmath
# Invoke this at any point before calling conversion functions
monkey_patch_spatialmath()
pose_msg = sm.SE3(1, 2, 3).to_ros()The to_ros() function returns a Pose msg by default.
A Transform msg can be returned instead with to_ros(as_tf=True).
from spatialmath_rospy import SE3
tf_msg = SE3(1, 2, 3).to_ros(as_tf=True)
print(type(tf_msg), '\n', tf_msg)
"""
<class 'geometry_msgs.msg._Transform.Transform'>
translation:
x: 1.0
y: 2.0
z: 3.0
rotation:
x: 0.0
y: 0.0
z: 0.0
w: 1.0
"""Quaternion msgs convert to UnitQuaternion objects and vice versa:
from spatialmath_rospy import UnitQuaternion
quat_msg = UnitQuaternion(1, [0, 0, 0]).to_ros()
print(type(quat_msg), '\n', quat_msg)
"""
<class 'geometry_msgs.msg._Quaternion.Quaternion'>
x: 0.0
y: 0.0
z: 0.0
w: 1.0
"""
unit_quat = UnitQuaternion.from_ros(quat_msg)
print(unit_quat)
" 1.0000 << 0.0000, 0.0000, 0.0000 >>UnitQuaternion can also be converted to a Transform msg with to_ros(as_tf=True):
from spatialmath_rospy import UnitQuaternion, SE3
quat = UnitQuaternion(1, [0, 0, 0])
tf_msg = quat.to_ros(as_tf=True)
print(tf_msg)
"""
translation:
x: 0
y: 0
z: 0
rotation:
x: 0.0
y: 0.0
z: 0.0
w: 1.0
"""This Transform will have always zero translation.
Just pass a std_msgs.msg.Header in to to_ros() to construct stamped objects:
from std_msgs.msg import Header
from spatialmath_rospy import SE3
pose_stamped_msg = SE3(1, 2, 3).to_ros(
Header(frame_id="world")
)
print(type(pose_stamped_msg), '\n', pose_stamped_msg)
"""
<class 'geometry_msgs.msg._PoseStamped.PoseStamped'>
header:
seq: 0
stamp:
secs: 0
nsecs: 0
frame_id: "world"
pose:
position:
x: 1.0
y: 2.0
z: 3.0
orientation:
x: 0.0
y: 0.0
z: 0.0
w: 1.0
"""This works for all supported ros msg types:
-
Pose/PoseStamped -
Transform/TransformStamped -
Quaternion/QuaternionStamped
Thanks goes to these wonderful people (emoji key):
Cal Hays π€ π π |
This project follows the all-contributors specification. Contributions of any kind welcome!
This package was created with Cookiecutter and the browniebroke/cookiecutter-pypackage project template.