No subject

Tue Mar 2 09:14:18 PST 2010

"There is a consensus to collapse the reference point and reference fr=
ame into<br>
a single coordinate frame.&quot;<br>
Mathematically, this seems ok to me too. However, will this involve a<br>
overhead in the implementation, such as adding new frames to the tf tree, m=
the code messy etc, if I want to find the velocity of a large number<br>
of different<br>
points in the frame of the gripper?<br>
Providing a reference point (in reference frame coordinates) does not seem =
too much additional information to me.<br>
<font color=3D"#888888"><br>
</font><div><div></div><div class=3D"h5"><br>
On Tue, Jun 8, 2010 at 1:52 AM, Tully Foote &lt;<a href=3D"mailto:tfoote at wi=">tfoote at</a>&gt; wrote:<br>
&gt; Hi Advait,<br>
&gt; There has been a bit of discussion of how to deal with twists well.=A0=
&gt; how to transform them has many different meanings implied by how they =
&gt; represented.=A0 In particular we found there were 4 different conventi=
&gt; without a clear way to distinguish between them.=A0 There is the resul=
ts of an<br>
&gt; API review at <a href=3D"
_API_Review" target=3D"_blank">
&gt; The biggest problem is that there is a lot of semantic information car=
&gt; in an instance of a Twist.=A0 Without forcing a lot of conventions on =
users we<br>
&gt; could not find a way to transform Twists without requiring a lot of ex=
&gt; information.<br>
&gt; There is an API to get the twist between two frames in the cturtle ver=
&gt; of tf.<br>
&gt; And kdl provides may twist manipulation methods which are recommended =
&gt; computing things involving twists.<br>
&gt; If anyone else has input we can come back to this.=A0 I would like to =
add this<br>
&gt; capability, but building a consensus on representations will be requir=
&gt; Tully<br>
&gt; On Mon, Jun 7, 2010 at 5:00 PM, Advait Jain &lt;<a href=3D"mailto:adva=
it at">advait at</a>&gt; wrote:<br>
&gt;&gt; I just realized that the wrench and twist messages currently<br>
&gt;&gt; do not explicitly store the point at which the wrench or twist<br>
&gt;&gt; is expressed.<br>
&gt;&gt; Is there is support in TF or somewhere else for getting the<br>
&gt;&gt; equivalent twist or wrench at a different point in space?<br>
&gt;&gt; e.g. to get the equivalent wrench at point B given a wrench<br>
&gt;&gt; acting at point A, I would use:<br>
&gt;&gt; Force at point A =3D Force at point B<br>
&gt;&gt; Torque at point A =A0=3D vector from A to B X Force + Torque at po=
int B<br>
&gt;&gt; It would be nice to have easy ways to be able to say, for<br>
&gt;&gt; example, that given a force that the robot is applying at the<br>
&gt;&gt; door handle, what the torque is at the hinge of the door.<br>
&gt;&gt; Another example could be if the we want a robot to perform<br>
&gt;&gt; a bi-manual and need to compute the motion of two points<br>
&gt;&gt; on the same rigid body.<br>
&gt;&gt; I think it would be really cool if TF could also shift twists,<br>
&gt;&gt; wrenches correctly in addition to changing coordinate frames.<br>
&gt;&gt; What are other people&#39;s thoughts about this?<br>
&gt;&gt; Advait<br>
&gt;&gt; _______________________________________________<br>
&gt;&gt; ros-users mailing list<br>
&gt;&gt; <a href=3D"mailto:ros-users at">ros-users at</=
&gt;&gt; <a href=3D"" target=
&gt; --<br>
&gt; Tully Foote<br>
&gt; Systems Engineer<br>
&gt; Willow Garage, Inc.<br>
&gt; <a href=3D"mailto:tfoote at">tfoote at</a>=
&gt; (650) 475-2827<br>
&gt; _______________________________________________<br>
&gt; ros-users mailing list<br>
&gt; <a href=3D"mailto:ros-users at">ros-users at</a><b=
&gt; <a href=3D"" target=3D"=
ros-users mailing list<br>
<a href=3D"mailto:ros-users at">ros-users at</a><br>
<a href=3D"" target=3D"_blan=
</div></div></blockquote></div><br><br clear=3D"all"><br>-- <br>Tully Foote=
<br>Systems Engineer<br>Willow Garage, Inc.<br><a href=3D"mailto:tfoote at wil=">tfoote at</a><br>(650) 475-2827<br>


More information about the ros-users mailing list