<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head>
<meta content="text/html; charset=UTF-8" http-equiv="Content-Type"/>
</head><body style="">
<div>
Hi Sankar,
</div>
<div>
 
</div>
<div>
the RigidBodyState class has a sourceFrame and targetFrame field. So I guess that should be fine as the frame_id is concerned. As for the goal_id, I am not sure what it is for on the ROS side. To collect information from different sources? To handle out of order results? If I don't miss anything it should be ok to ignore in the rock/ros bridge. So, just generate an id on the ROS side.
</div>
<div>
 
</div>
<div>
The problem with introducing new types is that it will make it much more difficult to plug in existing modules.
<br/> And so far, I don't really see a compelling reason (but of course might have missed something).
</div>
<div>
 
</div>
<div>
cheers,
</div>
<div>
 
</div>
<div>
Jakob
</div>
<div>
<br/>On September 15, 2013 at 7:56 PM Sankaranarayanan Natarajan <sankaranarayanan.natarajan@dfki.de> wrote:
</div>
<div style="position: relative;">
<blockquote style="margin-left: 0px; padding-left: 10px; border-left: solid 1px blue;" type="cite">
<div>
Hi Jakob,
</div>
<div>
 
</div>
<div>
frame_id -> so the module knows that the given pose is described w.r.t "frame_id" frame...i.e target pose can given w.r.t base link or laser link/camera link
</div>
<div>
goal_id   -> ROS actionlib needs a unique id to do a action.
</div>
<div>
 
</div>
<div>
Regards
</div>
<div>
-Sankar
<br/>Jakob Schwendner <jakob.schwendner@dfki.de> hat am 15. September 2013 um 18:58 geschrieben:
</div>
<div style="position: relative;">
<blockquote style="margin-left: 0px; padding-left: 10px; border-left: solid 1px blue;" type="cite">
<div>
Hi Sankar,
</div>
<div>
 
</div>
<div>
why do you need the frame_id and the goal_id? Otherwise you could just use base::Joints and RigidBodyState.
</div>
<div>
 
</div>
<div>
cheers,
</div>
<div>
 
</div>
<div>
Jakob
</div>
<div>
<br/>On September 15, 2013 at 6:44 PM Sankaranarayanan Natarajan <sankaranarayanan.natarajan@dfki.de> wrote:
</div>
<div style="position: relative;">
<blockquote style="margin-left: 0px; padding-left: 10px; border-left: solid 1px blue;" type="cite">
<div>
Hi Guys,
</div>
<div>
 
</div>
<div>
I need some datatypes to handle motionplanner input and ouput. I was thinking for atleast for motionplanner input we can use exiting data types by adding two lines to the existing datatypes(z.Bb base/Pose) or a new datatype for motionplanner.. Anyhow a sample struct i have created...
</div>
<div>
 
</div>
<div>
namespace base
<br/>{
<br/>        namespace motionplanner
<br/>        {
<br/>                enum result
<br/>                {
<br/>                        SUCCESSFUL              =  0,  
<br/>                        PLANNER_FAILED          = -1,
<br/>                        COLLISION_DETECTED      = -2,
<br/>                        IK_FAILED               = -3,
<br/>                        PATH_TOLERANCE_VIOLATED = -4,
<br/>                        GOAL_TOLERANCE_VIOLATED = -5
<br/>                };
<br/>
<br/>                struct Pose
<br/>                {
<br/>                        base::Time time;
<br/>                        std::string frame_id;   // the pose is defined w.r.t this frame
<br/>                        std::string goal_id;    // unique id for a task
<br/>                        base::Pose targetpose;
<br/>                };
<br/>
<br/>                struct JointAngle
<br/>                {    
<br/>                        base::Time time;
<br/>                        std::string frame_id;   // the pose is defined w.r.t this frame
<br/>                        std::string goal_id;    // unique id for a task
<br/>                        base::samples::Joints joints;
<br/>                };    
<br/>
<br/>    }   
<br/>}
</div>
<div>
 
</div>
<div>
 
</div>
<div>
Regards
</div>
<div>
-Sankar
</div>
</blockquote>
<br/> 
</div>
<div>
--
<br/>Jakob Schwendner, M.Sc.
<br/>Researcher
<br/>
<br/>DFKI Bremen
<br/>Robotics Innovation Center
<br/>Robert-Hooke-Straße 5
<br/>28359 Bremen, Germany
<br/>
<br/>Phone: +49 (0)421 17845-4120
<br/>Fax: +49 (0)421 17845-4150
<br/>E-Mail: jakob.schwendner@dfki.de
<br/>
<br/>Weitere Informationen: http://www.dfki.de/robotik
<br/>-----------------------------------------------------------------------
<br/>Deutsches Forschungszentrum fuer Kuenstliche Intelligenz GmbH
<br/>Firmensitz: Trippstadter Straße 122, D-67663 Kaiserslautern
<br/>Geschaeftsfuehrung: Prof. Dr. Dr. h.c. mult. Wolfgang Wahlster
<br/>(Vorsitzender) Dr. Walter Olthoff
<br/>Vorsitzender des Aufsichtsrats: Prof. Dr. h.c. Hans A. Aukes
<br/>Amtsgericht Kaiserslautern, HRB 2313
<br/>Sitz der Gesellschaft: Kaiserslautern (HRB 2313)
<br/>USt-Id.Nr.: DE 148646973
<br/>Steuernummer: 19/673/0060/3
<br/>-----------------------------------------------------------------------
</div>
</blockquote>
<br/> 
</div>
</blockquote>
<br/> 
</div>
<div id="ox-signature">
--
<br/>Jakob Schwendner, M.Sc.
<br/>Researcher
<br/>
<br/>DFKI Bremen
<br/>Robotics Innovation Center
<br/>Robert-Hooke-Straße 5
<br/>28359 Bremen, Germany
<br/>
<br/>Phone: +49 (0)421 17845-4120
<br/>Fax: +49 (0)421 17845-4150
<br/>E-Mail: jakob.schwendner@dfki.de
<br/>
<br/>Weitere Informationen: http://www.dfki.de/robotik
<br/>-----------------------------------------------------------------------
<br/>Deutsches Forschungszentrum fuer Kuenstliche Intelligenz GmbH
<br/>Firmensitz: Trippstadter Straße 122, D-67663 Kaiserslautern
<br/>Geschaeftsfuehrung: Prof. Dr. Dr. h.c. mult. Wolfgang Wahlster
<br/>(Vorsitzender) Dr. Walter Olthoff
<br/>Vorsitzender des Aufsichtsrats: Prof. Dr. h.c. Hans A. Aukes
<br/>Amtsgericht Kaiserslautern, HRB 2313
<br/>Sitz der Gesellschaft: Kaiserslautern (HRB 2313)
<br/>USt-Id.Nr.: DE 148646973
<br/>Steuernummer: 19/673/0060/3
<br/>-----------------------------------------------------------------------
</div>
</body></html>