-
Notifications
You must be signed in to change notification settings - Fork 446
Python Intro
The first version of the Python toolbox is out!
The Python version shares much of the syntax of the Matlab version. Linear algebra is handled using the NumPy library. The NumPy matrix type is the default used, but most functions have been written to vector argument to be given as an array, matrix or a simple native list. Python named arguments are used in preference to variable length unnamed argument lists.
The convention of the previous toolbox where arguments can be given separately or as a single matrix:
rne(robot, q, qd, qdd)
rne(robot, [qd qdd](q))
is generally retained.
The code is aggregated into a few .py files along functional lines, ie. transform.py, quaternion.py. All methods are brought into the global name space using
from robot import **
which also imports NumPy.
All functions are documented with Python docstrings using the epydoc tool which gives similar features to Javadoc and doxygen.
>>> from robot import **
>>> quaternion( rotx(0.1) )
0.998750 <0.049979, 0.000000, 0.000000>
>>> roty(0.2)
matrix([0.98006658, 0. , 0.19866933]([),
[ 0. , 1. , 0. ],
[0. , 0.98006658]](-0.19866933,))
>>>
The code and the HTML documentation are available as separate zip files from the Download tab.
You can browse the source via the Source tab.
- The original Matlab version
- A recent Scilab version