Re: [OpenRAVE-users] PR2 inverse kinematics problem

SourceForge Headquarters 1320 Columbia Street Suite 310 San Diego, CA 92101 +1 (858) 422-6466

I lowered logllthresh to -100. Now a lot of candidate base poses are
generated. However none of them is such that the object to grasp is
reachable (InverseIK returns None when asked for a solution). There must be
something wrong in the code I attached in the original email, but I can't
find it. If I use the same code but instead of using inverse reachability I
generate random base poses around and object then the approach works ok.

Lorenzo

On Tue, Oct 23, 2012 at 12:43 AM, Rosen Diankov [via OpenRAVE Users List] <
ml-...@n3...> wrote:

> if it happens once in a while, then it might be an artefact of the
> discretization when generating the kinematic reachability and inverse
> reachability databases. if it happens 90%+ of the time, then it's a
> bug.
>
> can you try lowering logllthresh to -100?
>
> rosen,
>
> 2012/10/23 lorenzoriano <[hidden email]<http://user/SendEmail.jtp?type=node&node=4025586&i=0>>:
>
>
> > Hi Rosen,
> >
> > thanks for your reply, and keep on with the amazing work!
> >
> > By "wrong" I mean that the few base poses that are generated using the
> > attached code cannot reach the object they were generated for. The only
> > difference between that code and the original show method in the inverse
> > reachability database code is that I am using leftarm instead of
> > leftarm_torso (the latter hadn't even tested all the poses after three
> days
> > of computation).
> >
> > Very often the samplers couldn't even produce a solution (returning
> None).
> >
> > Lowering logllthresh, even down to 0, didn't work.
> >
> > Best,
> >
> > Lorenzo
> >
> > On Mon, Oct 22, 2012 at 8:24 PM, Rosen Diankov [via OpenRAVE Users List]
> > <[hidden email]> wrote:
> >>
> >> by wrong, do you mean no inverse kinematics solutions are not
> generated?
> >>
> >> as for the computeBaseDistribution returning None, have you tried
> >> lowering logllthresh?
> >>
> >> rosen,
> >>
> >> 2012/10/16 lorenzoriano <[hidden email]>:
> >>
> >> > Hi all,
> >> >
> >> > I'm trying to use the inverse kinematics reachability database. I've
> >> > generated the databases for both right and left arms of the PR2 (I
> >> > couldn't
> >> > do it for leftarm_torso, after 48h it had done 27000/131356). The
> code I
> >> > am
> >> > using is attached:  generate_inverse_collision.py
> >> >
> >> > <
> http://openrave-users-list.185357.n3.nabble.com/file/n4025558/generate_inverse_collision.py>
>
> >> >
> >> > However the base poses InverseReachabilityModel produces are always
> >> > wrong,
> >> > and often computeBaseDistribution returns None (no distribution). The
> >> > same
> >> > happens when following the tutorial at
> >> >
> >> >
> http://openrave.org/docs/latest_stable/openravepy/databases.inversereachability/
> >> >
> >> > <
> http://openrave.org/docs/latest_stable/openravepy/databases.inversereachability/>
>
> >>
> >> > .
> >> >
> >> > Does anybody know what I am doing wrong?
> >> >
> >> > Thanks,
> >> >
> >> > Lorenzo
> >> >
> >> >
> >> >
> >> >
> >> >
> >> > --
> >> > View this message in context:
> >> >
> http://openrave-users-list.185357.n3.nabble.com/PR2-inverse-kinematics-problem-tp4025558.html
> >> > Sent from the OpenRAVE Users List mailing list archive at Nabble.com.
> >> >
> >> >
> >> >
> ------------------------------------------------------------------------------
>
> >> > Don't let slow site performance ruin your business. Deploy New Relic
> APM
> >> > Deploy New Relic app performance management and know exactly
> >> > what is happening inside your Ruby, Python, PHP, Java, and .NET app
> >> > Try New Relic at no cost today and get our sweet Data Nerd shirt too!
> >> > http://p.sf.net/sfu/newrelic-dev2dev
> >> > _______________________________________________
> >> > Openrave-users mailing list
> >> > [hidden email]
> >> > https://lists.sourceforge.net/lists/listinfo/openrave-users
> >>
> >>
> >>
> ------------------------------------------------------------------------------
>
> >> Everyone hates slow websites. So do we.
> >> Make your web apps faster with AppDynamics
> >> Download AppDynamics Lite for free today:
> >> http://p.sf.net/sfu/appdyn_sfd2d_oct
> >> _______________________________________________
> >> Openrave-users mailing list
> >> [hidden email]
> >> https://lists.sourceforge.net/lists/listinfo/openrave-users
> >>
> >>
> >> ________________________________
> >> If you reply to this email, your message will be added to the
> discussion
> >> below:
> >>
> >>
> http://openrave-users-list.185357.n3.nabble.com/PR2-inverse-kinematics-problem-tp4025558p4025582.html
> >> To unsubscribe from PR2 inverse kinematics problem, click here.
> >> NAML
> >
> >
> >
> >
> > --
> > Lorenzo Riano, PhD
> > Research Scientist
> > EECS
> > University of California Berkeley
> > Berkeley, CA
> > 94720
> >
> > web: http://www.eecs.berkeley.edu/~riano/
> > email: [hidden email]
> >
> > skype: lorenzo.riano
> >
> > “I have come to believe that the whole world is an enigma, a harmless
> enigma
> > that is made terrible by our own mad attempt to interpret it as though
> it
> > had an underlying truth.”
> > Umberto Eco, The name of the Rose
> >
> >
> > ________________________________
> > View this message in context: Re: PR2 inverse kinematics problem
> >
> > Sent from the OpenRAVE Users List mailing list archive at Nabble.com.
> >
> >
> ------------------------------------------------------------------------------
>
> > Everyone hates slow websites. So do we.
> > Make your web apps faster with AppDynamics
> > Download AppDynamics Lite for free today:
> > http://p.sf.net/sfu/appdyn_sfd2d_oct
> > _______________________________________________
> > Openrave-users mailing list
> > [hidden email] <http://user/SendEmail.jtp?type=node&node=4025586&i=1>
> > https://lists.sourceforge.net/lists/listinfo/openrave-users
> >
>
> ------------------------------------------------------------------------------
>
> Everyone hates slow websites. So do we.
> Make your web apps faster with AppDynamics
> Download AppDynamics Lite for free today:
> http://p.sf.net/sfu/appdyn_sfd2d_oct
> _______________________________________________
> Openrave-users mailing list
> [hidden email] <http://user/SendEmail.jtp?type=node&node=4025586&i=2>
> https://lists.sourceforge.net/lists/listinfo/openrave-users
>
>
> ------------------------------
>  If you reply to this email, your message will be added to the discussion
> below:
>
> http://openrave-users-list.185357.n3.nabble.com/PR2-inverse-kinematics-problem-tp4025558p4025586.html
>  To unsubscribe from PR2 inverse kinematics problem, click here<http://openrave-users-list.185357.n3.nabble.com/template/NamlServlet.jtp?macro=unsubscribe_by_code&node=4025558&code=bG9yZW56by5yaWFub0BnbWFpbC5jb218NDAyNTU1OHwtMTkzOTAwOTI0Nw==>
> .
> NAML<http://openrave-users-list.185357.n3.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml>
>

-- 
Lorenzo Riano, PhD
Research Scientist
EECS
University of California Berkeley
Berkeley, CA
94720

web: http://www.eecs.berkeley.edu/~riano/
email: lor...@be...
skype: lorenzo.riano

“I have come to believe that the whole world is an enigma, a harmless
enigma that is made terrible by our own mad attempt to interpret it as
though it had an underlying truth.”
Umberto Eco, The name of the Rose

--
View this message in context: http://openrave-users-list.185357.n3.nabble.com/PR2-inverse-kinematics-problem-tp4025558p4025587.html
Sent from the OpenRAVE Users List mailing list archive at Nabble.com.