Re: answers cntd.

----- Original Message -----
From: "Herman Bruyninckx" <[email protected]>
To: <unlisted-recipients:>; <no To-header on input>
Cc: "Open RObot COntrol Software" <[email protected]>
Sent: Monday, October 01, 2001 1:05 PM
Subject: Re: [Orocos] answers cntd.


> On Mon, 1 Oct 2001, Cezary Zielinski wrote:
>
> [...]
> > What I think you are saying that the same set of sensor readings is
> > used both for servo control and trajectory generation and thus there
> > might be oposing reaction of both levels. If my understanding of the
> > problem is correct, then my answer is that you are probobly right -
> > there will be a conflict. Nevertheless I am not proposing such a
> > thing. My intension was that each level uses its own set of sensors
> > which will provide readings that will not produce conflicting
> > modifications.
>
> Ok, I understand. And that's also what I think defines the borders of
> a `component': _within_ a component, stability (and other proporties)
> are guaranteed (because there is a responsible procedure _in_ the
> component), while using feedback around the component is without
> real-time guarantees. (Unless the internal component is
> deterministic.)
>

We are providing a very powerful and versatile tool. It is capable of many
things, but that also includes the possibility of misusing the tool with
disastrous effects. Our stance was that in a research environment, where the
qualifications of the users are high, it is the responsibility of the
investigator to assemble the system correctly. Nevertheless mistakes are
bound to be made, so on the one hand due caution should be exerted while
experimenting, and on the other hand appropriate error detection and
correction facilities should be provided. That is why I tried to draw your
attantion to this aspects of building such a system within the OROCOS
project.

Cezary

> Orocos mailing list
> [email protected]
> http://mail.mech.kuleuven.ac.be/mailman/listinfo/orocos