Hi Lukas,
I was following your example of a wrapper and found the following
the function ss2tf has a different interface than your function ss.
The last argument is not the time sample as in your function but an
index of the input vector
http://www.mathworks.com/help/toolbox/signal/ss2tf.html

Is there any of the options in your ss function that can handle this
or should that option be added?

Also, I cannot access tfdata directly nor octave can find it (though
is there in @lti), I am not experienced with objects and may be
something related to that.

Thanks


On Wed, Jul 6, 2011 at 2:03 PM, Juan Pablo Carbajal <[email protected]> wrote:
> Hi Lukas
>
> That definitely sounds like an idea. I will roceed the way you
> suggested and when I got it UNIT running I will send them your
> comments so they update their package.
>
> I will be back with more news.
>
> JPi
>
>
> On Wed, Jul 6, 2011 at 1:02 PM, Lukas Reichlin <[email protected]> 
> wrote:
>> On 05.07.2011, at 14:20, Juan Pablo Carbajal wrote:
>>
>>> On Tue, Jul 5, 2011 at 12:05 PM, Lukas Reichlin
>>> <[email protected]> wrote:
>>>>
>>>> On 05.07.2011, at 08:13, Juan Pablo Carbajal wrote:
>>>>
>>>>> On Tue, Jul 5, 2011 at 7:22 AM, Lukas Reichlin <[email protected]> 
>>>>> wrote:
>>>>>> On 05.07.2011, at 00:04, Juan Pablo Carbajal wrote:
>>>>>>
>>>>>>> Hi all,
>>>>>>>
>>>>>>> I was searching the control package for the functions ss2tf, abcddim.
>>>>>>> If I am not worng they were there some time ago, weren't they? At
>>>>>>> least google seems to support this.
>>>>>>> Where can I find them now?
>>>>>>> I have installed the control package 2.0.2
>>>>>>>
>>>>>>> Thank you very much
>>>>>>>
>>>>>>> --
>>>>>>> M. Sc. Juan Pablo Carbajal
>>>>>>> -----
>>>>>>> PhD Student
>>>>>>> University of Zürich
>>>>>>> www.ailab.ch/carbajal
>>>>>>
>>>>>> Hi Juan
>>>>>>
>>>>>> These two function are part of control-1.0.11, now labeled as 
>>>>>> control-legacy [1].  The control package version 2 is now based on 
>>>>>> SLICOT routines and supports overloaded operators (L = P * C, ...).  The 
>>>>>> new control package is more Matlab-compatible, but m-files using 
>>>>>> functions specific to control-legacy need to be adapted to the newer 
>>>>>> control-2.x.y package:
>>>>>>
>>>>>> Replacement for ss2tf:
>>>>>> sys_tf = tf (sys_ss)
>>>>>>
>>>>>> Replacement for abcddim:
>>>>>> [p, m] = size (sys)
>>>>>>
>>>>>> Regards,
>>>>>> Lukas
>>>>>>
>>>>>> [1]
>>>>>> http://octave.svn.sourceforge.net/viewvc/octave/trunk/octave-forge/extra/control-legacy/
>>>>>>
>>>>>>
>>>>>
>>>>> Thank you very much for the answer.
>>>>> I got to this because of a system identification package available in
>>>>> the internet http://sigpromu.org/idtoolbox/. I do not know whether to
>>>>> get control-legacy or modify the package. What would you suggest?
>>>>>
>>>>>
>>>>> --
>>>>> M. Sc. Juan Pablo Carbajal
>>>>> -----
>>>>> PhD Student
>>>>> University of Zürich
>>>>> www.ailab.ch/carbajal
>>>>
>>>> I'm afraid that UNIT version 3 won't run on Octave because of its 
>>>> graphical user interface.  There's no Matlab-compatible GUI support at the 
>>>> moment (Octave 3.4.2).  If you have enough stamina, you can help improving 
>>>> the control package and/or adapt UNIT (without its GUI) to Octave.
>>>>
>>>> Regards,
>>>> Lukas
>>>>
>>>>
>>>
>>> Thanks for the answer,
>>>
>>> I was thinking on the non-GUI UNIT, of course. I will like to make it
>>> work with Octave (I almost need to do it). Now, you provide two
>>> possible paths, adapt control package to UNIT or vice-versa.
>>> It seems to me that adapting (probably extending with a couple of
>>> wrappers) the control package is more reasonable. What do you think?
>>> So far I have noticed that UNIT needs lots of functions that were in
>>> the old control package but are not there anymore. Is there a
>>> compatibility table between old control and new control packages
>>> available? Where could I look?
>>> If the table doesn't exist I can provide a list of the functions that
>>> UNIT is requesting and see if you can tell which ones are missing and
>>> which ones have other name/interface. What do you think?
>>>
>>> Regards,
>>>
>>> --
>>> M. Sc. Juan Pablo Carbajal
>>> -----
>>> PhD Student
>>> University of Zürich
>>> www.ailab.ch/carbajal
>>
>> Matlab exists since three decades and has many outdated commands which 
>> Mathworks doesn't list on their website any longer, but somehow people keep 
>> using them.
>> Examples are:
>> [num, den] = ss2tf (a, b, c, d)
>> step (num, den)
>> bode (a, b, c, d)
>>
>> More recent editions of the control systems toolbox support "systems", 
>> therefore you don't need to juggle with state-space matrices or polynomial 
>> vectors.  When I started to rewrite the control package, I decided to focus 
>> my efforts on the current set of commands.  Neglecting antique commands also 
>> simplifies the code.
>>
>> Maybe you want to create a package like "control-stone-age-commands" and 
>> provide some wrappers e.g.
>>
>> function [num, den] = ss2tf (varargin)
>>  [num, den] = tfdata (ss (varargin{:}));
>> endfunction
>>
>> Regards,
>> Lukas
>>
>> PS: List of available functions:
>> http://octave.sourceforge.net/control/overview.html
>>
>>
>>
>>
>
>
>
> --
> M. Sc. Juan Pablo Carbajal
> -----
> PhD Student
> University of Zürich
> www.ailab.ch/carbajal
>



-- 
M. Sc. Juan Pablo Carbajal
-----
PhD Student
University of Zürich
www.ailab.ch/carbajal

------------------------------------------------------------------------------
All of the data generated in your IT infrastructure is seriously valuable.
Why? It contains a definitive record of application performance, security 
threats, fraudulent activity, and more. Splunk takes this data and makes 
sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-d2d-c2
_______________________________________________
Octave-dev mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/octave-dev

Reply via email to