rct-python

Robotics Coordinate Transform (Python)

Dependencies

1. You will need the tf2_py implementation which was split from the roscode. You can find this lib here

https://github.com/ros/geometry_experimental/tree/indigo-devel/tf2_py

export prefix="/tmp/rct"
export prefix_tf2_py="${prefix}/tf2_py"
cd /tmp
git clone https://github.com/ros/geometry_experimental.git
cd geometry_experimental/
mkdir build && cd build
cmake .. -DCMAKE_INSTALL_PREFIX=$prefix_tf2_py
make install -j4

2. RSB See documentation at:

http://docs.cor-lab.de//rsb-manual/0.11/html/index.html

3. RST See documentation at:

http://docs.cor-lab.de//rst-manual/0.11/html/index.html

Installation

Once you got the environment setup, you can install rct-python easily.

Source your environment and set up your (target) $prefix and install:

export prefix="/tmp/rct"
export prefix_rct_python="${prefix}/rct-python/"
python setup.py install --prefix=$prefix

Usage

There are two examples in the examples folder within the repo (will not be installed). See these for a detailed test. They will also work with their C++ counterpart.

import time
import rct
import logging
logging.basicConfig(level=logging.WARNING)
logging.getLogger('rct').setLevel(logging.INFO)

a = rct.TransformerFactory()

p = a.create_transform_publisher("ExamplePublisher")

t_s = rct.Transform("blub", "parentoruu", "childoruu", time.time())
p.send_transform(t_s, rct.TransformType.STATIC)

t_d = rct.Transform("bla", "leParent", "leChild", time.time())
p.send_transform(t_d, rct.TransformType.DYNAMIC)

r = a.create_transform_receiver()

Example Usage

You can simply start the cross-lib-provider.py file, which will provide a static and a dynamic transformation. There is a similar example in the java implementation of rct, which does the with a different predix.

Afterwards, simply running the cross-lib-tester.py will perform a test on the by python and java provided transformation.

Alternatively, you can paste this in an ipython session within a setup environment to test one single transformation and apply it.

import logging
logging.getLogger('rsb').setLevel(logging.DEBUG)
logging.getLogger('rct').setLevel(logging.DEBUG)

import time

import numpy as np
np.set_printoptions(precision=4)

import rct
from pyrr import Matrix33, Matrix44, Vector4

tf2_subscriber = rct.TransformerFactory().create_transform_receiver()

print "Available Transforms:\n\t".join(("\t" + tf2_subscriber.all_frames_as_string()).split('\n'))

when = time.time()
time.sleep(0.2)

tf = tf2_subscriber.lookup_transform("python_static", "base", when)

p = Vector4([1.0, 1.0, 1.0, 1.0])
p_transformed = tf.transformation.apply_transformation(p)

print p, "-->", p_transformed

Notes

Please note that the future pattern has not yet been implemented in this version. Only past transformations can be obtained.