Python Robotics toolbox (python-robotics)

robot kinematics with python-visual visualization

These python modules has been written for use in the course Robot Modeling at the Hague University of Applied Sciences, by Rufus Fraanje (p.r.fraanje@hhs.nl)

In this course, we use the book Robot Modeling and Control by Mark W. Spong, Seth Hutchinson, M. Vidyasagar, Wiley, 2005, some pieces of the code may resemble the notation and derivations from this (excellent) book.

Dependencies

• the great transformations module by Christoph Gohlke
• the easy to use python-visual module (also called vpython) for 3D visualization

Description and examples

• frames.py: construction and visualization of frames in 3D, and performing translations, rotations, homogeneous transformations

Examples:

>>> Rz  = so3(angle=30)   
>>> Ry  = so3(angle=60,axis=[0,1,0])
>>> Rzy = Rz*Ry           # new object, equivalent to Rzy = Rz * Ry
>>> F0 = frame(label='0') # build frame 0 with default settings
>>> F1 = Ry*Rz*F0         # rotation of F0 (note: premultiplication is
                          # rotation about fixed axes
>>> F2.label = '1'        # always set label, because this is not being done
                          # automatically 
>>> F2 = F0*Rz*Ry         # is possible as well, see documentation
                          # under __rmul__()
>>> F2.label = '2'        # always set label, because this is not being done
                          # automatically 

• links.py: construction of joints with attached links and a frame located at the end of the link, specified on the basis of Denavit Hartenberg parameters

from links import * 
from time import sleep

scene = display(title='Test visualization',center=(0,0,0),forward=(0,0,-1),up=(0,1,0),background=visual.color.background,range=2)
scene.fov = 0.01 # mimic orthographic projection

scale = 1;
F0 = frame([0,0,0],label='0, joint 1',scale=scale)

# Two links specified by their Denavit-Hartenberg parameters: 
L1 = link(F0,a=1,alpha=-90,d=0,theta=0,q=45,joint='r',label='1, joint 2',scale=1,unit='deg') 
L2 = link(L1.frame,a=1,alpha=0,d=0,theta=0,q=.5,joint='p',label='2, joint 3',scale=1,unit='deg') 
L1.a = 1.5
print(L1.a)
L1.set_delayed(prop='q',incr=30,delay=2,niter=20):
print(L1.q)

• examples: frames_examples.py and links_examples.py give some examples for using frames.py and links.py, see also ur5.py for a visualization of the Universal Robot UR5 robot, e.g. try

python ur5.py

To do

The current focus has been on 3D visualization of frames, joints, links on the basis of Denavit Hartenberg parameters. Inverse kinematics and (multibody) dynamics haven't been implemented yet. Documentation needs to be completed.

The code has been written with care, but may be improved. Any suggestions are welcome. Feel free to use the code, but its use is without any guarantee.

Acknowledgement

• Matlab Robotic Toolbox by Peter Corke, this is a great toolbox which was the inspiration for this toolbox. But note, python-robotics is not a translation of the Matlab Robotic Toolbox!