Re: [Playerstage-users] have you see this

[Geoffrey Biggs <g.biggs@au...>]

Got to agree with this. I missed it in my initial look, but seeing it
now, it certainly _looks_ pretty... which could mean good for vision
processing simulation. Hard to say what it's like for other sensors.

Hardware physics simulation is certainly very cool. Something Gazebo
should look into, if it's not already.

Geoff

Radu Bogdan Rusu wrote:
> Their visualizer looks neato... gotta give them at least that. :) 
> Looking forward for the Ogre3D integration in Gazebo, then we'll see a 
> nice one-on-one.
> 
> Other than that... scarry. :)
> 
> PhysX support... now that could be something cool.
> 
> I would have to be one crazy researcher to use something like that 
> though. It'll probably be appealing for some highschool students and MS 
> buffs, but any serious university researcher will need an open source 
> alternative.
> 
> I wouldn't worry too much.
> 
> Cheers,
> R.


-- 
Robotics research group, University of Auckland
http://www.ece.auckland.ac.nz/~gbig005/

Re: [Playerstage-users] have you see this

[Radu Bogdan Rusu <rusu@cs...>]

Their visualizer looks neato... gotta give them at least that. :)=20
Looking forward for the Ogre3D integration in Gazebo, then we'll see a=20
nice one-on-one.

Other than that... scarry. :)

PhysX support... now that could be something cool.

I would have to be one crazy researcher to use something like that=20
though. It'll probably be appealing for some highschool students and MS=20
buffs, but any serious university researcher will need an open source=20
alternative.

I wouldn't worry too much.

Cheers,
R.

Victor Manuel J=E1quez Leal wrote:
> I'm not involve any more with mobile robots, just stay in the lists
> for curiosity. But now I'm in shock: Have you see this?
> http://msdn.microsoft.com/robotics/
>  =20
--=20
| Radu Bogdan Rusu | http://rbrusu.com/
| http://www9.cs.tum.edu/people/rusu/
| Intelligent Autonomous Systems
| Technische Universitaet Muenchen

Re: [Playerstage-users] have you see this

[Douglas A Few <Douglas.Few@in...>]

I believe it was Brian Gerkey who said all usernet sites degrade into the 
linux versus windows debate when he was presenting the Player/Stage Segway 
RMP work at one of Doug Gages SDR meetings.  So now it has come to pass.

For the record I am on the linux side.

Douglas A Few
**************************************************
Idaho National Laboratory
Robotic, and Human Systems Group
Principal Research Scientist 
208 526-3077
208 526-4750  robot lab
 http://www.inel.gov/adaptiverobotics
**************************************************



"Fred Labrosse" <ffl@...> 
Sent by: playerstage-users-bounces@...
06/20/2006 04:18 PM
Please respond to
playerstage-users@...


To
playerstage-users@...
cc

Subject
Re: [Playerstage-users] have you see this






On Tue, June 20, 2006 11:12 pm, Jordi wrote:
>
>> > I'm not involve any more with mobile robots, just stay in the lists
>> > for curiosity. But now I'm in shock: Have you see this?
>> > http://msdn.microsoft.com/robotics/
>>
>> I just noticed it following an announcement by robotsoft that it's 
being
>> used by their latest robot.  I'm not sure I want to have anything to do
>> with them anymore... ;-)
>>
>> Fred
>>
>> P.S. More seriously, anybody had a detailed look at it?
>
> I can't, I can't find the "download the source code" link :P  (It may be
> worthy to take a look to the ideas anyway?)

When I went on the website, my browser (konqueror) complained about an
error in a javascript...

Fred




_______________________________________________
Playerstage-users mailing list
Playerstage-users@...
https://lists.sourceforge.net/lists/listinfo/playerstage-users

Re: [Playerstage-users] have you see this

[Fred Labrosse <ffl@ab...>]

On Tue, June 20, 2006 11:12 pm, Jordi wrote:
>
>> > I'm not involve any more with mobile robots, just stay in the lists
>> > for curiosity. But now I'm in shock: Have you see this?
>> > http://msdn.microsoft.com/robotics/
>>
>> I just noticed it following an announcement by robotsoft that it's being
>> used by their latest robot.  I'm not sure I want to have anything to do
>> with them anymore... ;-)
>>
>> Fred
>>
>> P.S. More seriously, anybody had a detailed look at it?
>
> I can't, I can't find the "download the source code" link :P  (It may be
> worthy to take a look to the ideas anyway?)

When I went on the website, my browser (konqueror) complained about an
error in a javascript...

Fred

Re: [Playerstage-users] have you see this

[Geoffrey Biggs <g.biggs@au...>]

I haven't had a detailed look yet (haven't had time, but I'll have to
since it's my research area). From the brief look I had last night it's
mainly just an API, kind of a robot version of .NET. More info when I've
had a closer look...

Geoff

Jordi wrote:
>>> I'm not involve any more with mobile robots, just stay in the lists
>>> for curiosity. But now I'm in shock: Have you see this?
>>> http://msdn.microsoft.com/robotics/
>> I just noticed it following an announcement by robotsoft that it's being
>> used by their latest robot.  I'm not sure I want to have anything to do
>> with them anymore... ;-)
>>
>> Fred
>>
>> P.S. More seriously, anybody had a detailed look at it?
> 
> I can't, I can't find the "download the source code" link :P  (It may be 
> worthy to take a look to the ideas anyway?)
> 
> ^_^/

-- 
Robotics research group, University of Auckland
http://www.ece.auckland.ac.nz/~gbig005/

Re: [Playerstage-users] have you see this

[Jordi <mumismo@gm...>]

> > I'm not involve any more with mobile robots, just stay in the lists
> > for curiosity. But now I'm in shock: Have you see this?
> > http://msdn.microsoft.com/robotics/
>
> I just noticed it following an announcement by robotsoft that it's being
> used by their latest robot.  I'm not sure I want to have anything to do
> with them anymore... ;-)
>
> Fred
>
> P.S. More seriously, anybody had a detailed look at it?

I can't, I can't find the "download the source code" link :P  (It may be 
worthy to take a look to the ideas anyway?)

^_^/

Re: [Playerstage-users] have you see this

[Fred Labrosse <ffl@ab...>]

On Tue, June 20, 2006 10:40 pm, Victor Manuel Jáquez Leal wrote:
> I'm not involve any more with mobile robots, just stay in the lists
> for curiosity. But now I'm in shock: Have you see this?
> http://msdn.microsoft.com/robotics/

I just noticed it following an announcement by robotsoft that it's being
used by their latest robot.  I'm not sure I want to have anything to do
with them anymore... ;-)

Fred

P.S. More seriously, anybody had a detailed look at it?

[Playerstage-users] have you see this

[<ceyusa@gm...>]

I'm not involve any more with mobile robots, just stay in the lists
for curiosity. But now I'm in shock: Have you see this?
http://msdn.microsoft.com/robotics/

[Playerstage-users] Problems with gazebo, playernav, and gzbuilder

[Glenn Laguna <galagun@sa...>]

Hi All,

For reference, I've tried this with gazebo_cvs, gazebo-0.6, player_cvs, and player-2.02.

First the problem with gzbuilder. It seems to create only a partial map if the
input file is over 10,000 pixels. From looking at the archives, I think that
Solly saw this also, but I didn't see a solution posted. Anyone have any
thoughts?


Concerning gazebo and playernav, gzbuilder always creates a sligtly larger map
than the input file. How should I set scale factors, origins, orientations,
etc, so that, for example, the lower left corner of the gazebo map corresponds
to the lower left corner of the playernav map?

Playernav seems to put the initial pose in the lower left corner, no matter
where the robot is in gazebo, and than the path in playernav seems to have
only a very rough relationship to the path the robot follows in gazebo.

Here's my gazebo world file: it's almost pioneer2at.world from the gazebo
examples:

<?xml version="1.0"?>
<gz:world xmlns:gz="http://playerstage.sourceforge.net/gazebo/xmlschema/#gz"; xmlns:model="http://playerstage.sourceforge.net/gazebo/xmlschema/#model"; xmlns:window="http://playerstage.sourceforge.net/gazebo/xmlschema/#window"; xmlns:param="http://playerstage.sourceforge.net/gazebo/xmlschema/#param">;

  <model:ObserverCam>
    <id>userCam0</id>
    <xyz>-0.389 -0.898 1.046</xyz>
    <rpy>0 37 31</rpy>
    <imageSize>640 480</imageSize>
    <updateRate>10</updateRate>

    <nearClip>0.2</nearClip>
    <farClip>100</farClip>

    <displayRays>false</displayRays>
    <displayAxes>false</displayAxes>
    <displayCoM>false</displayCoM>
    <displayTextures>true</displayTextures>
  </model:ObserverCam>

  <model:LightSource>
    <id>light0</id>
    <xyz>-5.000 -5.000 10.000</xyz>
  </model:LightSource>


  <model:Terrain>
    <terrainFile>mymap.gzb</terrainFile>
    <xyz>0.0 0.0 0.0</xyz>
    <color>0.0 1.0 0.0</color>
    <textureFile>grid.ppm</textureFile>
  </model:Terrain>


  <model:Pioneer2AT>
    <id>robot1</id>
    <xyz>1.0 1.0 0.0</xyz>
    <model:SickLMS200>
      <id>laser1</id>
      <xyz>0.100 0.000 0.30</xyz>
      <rpy>0 0 0</rpy>
      <scanRate>75</scanRate>
      <model:MonoCam>
        <id>camera1</id>
        <xyz>0 0 0.20</xyz>
        <nearClip>0.5</nearClip>
        <farClip>3</farClip>
      </model:MonoCam>
    </model:SickLMS200>
  </model:Pioneer2AT>

</gz:world>


Here's my player config file. It's the one posted by Nathan a few weeks ago,
with some minor differences.

driver
 (
   name "gazebo"
   provides ["simulation:0"]
   plugin "/usr/local/lib/libgazeboplugin"
   server_id "default"
   alwayson 1
 )
 
driver
 (
   name "gazebo"
   provides ["odometry:::position2d:0"]
   gz_id "robot1"
 )
 
driver
 (
   name "vfh"
   provides ["position2d:1"]
   requires ["odometry:::position2d:0" "laser:0"]
   safety_dist_0ms 0.1
   safety_dist_1ms 0.1
   distance_epsilon 0.3
   angle_epsilon 5
 )
 
driver
 (
   name "gazebo"
   provides ["laser:0"]
   gz_id "laser1"
 )
 
driver
 (
   name "mapfile"
   provides ["map:0"]
   filename "mymap.pgm"
   resolution 0.1
   origin [-1.13 -1.13]
 )

driver
(
  name "amcl"
  provides ["localize:0" "position2d:2"]
  requires ["odometry:::position2d:1" "laser:0" "laser:::map:0"]
  odom_drift[0] [0.3 0.0 0.0]
  odom_drift[1] [0.0 0.2 0.0]
  odom_drift[2] [0.2 0.0 0.2]
)
 

driver
 (
  name "wavefront"
  provides ["planner:0"]
  requires ["output:::position2d:1" "input:::position2d:0" "map:0"]
  alwayson 1
 )
 

I'd appreciate any help, since I apparently missed something.

Regards,

-- 
 ===================================================================
| Glenn Laguna, Ph.D.                    e-mail: galagun@... |
| Sandia National Labs -- Dept 6634      Phone:  (505) 844-2308     |
| MS-1003                                                           |
| Albuquerque, NM  87185-5800            Fax:    (505) 844-6161     |
 ===================================================================

Re: [Playerstage-users] rFlex rs232 settings

[Toby Collett <tcollett+player@pl...>]

I have never had to set these externally to player.

Ive just had a look at the source and confirmed that the driver sets all 
of this to appropriate values.

   strncpy( rdev.ttyport, device_name, MAX_NAME_LENGTH);
   rdev.baud           = 115200;
   rdev.databits       = 8;
   rdev.parity         = N;
   rdev.stopbits       = 1;
   rdev.hwf            = 0;
   rdev.swf            = 0;


Toby

Eric Innis wrote:
> 1.)  Does the rflex driver handle the details of the connection such
> as baud rate and flow control?
> Or do I need to configure that myself via 'setserial' or similar?
> 
> 2.)  Does anybody know the parameters for the rFlex board e.g. 9600-8-N-1 ?
> 
> 
> Thanks once again,
> 

Re: [Playerstage-users] Playernav and AMCL

[Toby Collett <tcollett+player@pl...>]

Please make your posts to the playerstage-users mailing list as this way 
they are archived and other can answer your posts.

As far as my testing went AMCL is functioning, however I was using it 
purely for localisation and never used it in combination with the player 
navigation tools (VFH/player nav etc)

It sounds like your issue may be with interactions between these, it is 
quite possible that there are a couple of bugs in the works, but at this 
stage they are unknown bugs so please let us know what you find.

I assume you read the message about the position2d interface to AMCl not 
working in some versions of 2.0.X, it might be a good Idea to try CVS 
and see if there is any difference.

Toby

chandan datta wrote:
> Hi toby,
> thanks for the hint to enable the AMCL debug gui.
> I just want to check if the AMCL driver is working properly in player 
> 2.0.2.So ,is their any link issue with playernav and AMCL ??
> 
> Plz suggest any other way to test the driver.
> 
> I modified the wavefront.cfg file in stage and the simple.world 
> file.They are attached.On running these i can see the particles in the 
> AMCL gui but on setting the pose and goal, the robot rotates in its own 
> position without going to the goal.
> 
> Is there any particular tweaking we need in the amcl.cc file ??
> 
> I don't know if vfh is working properly on stage now.
> 
> regards,
> Chandan
> 
> / )  / _   _/_    /  )__/_/_
> (__/)(//)(/(//) /(_/(// /(/
> 
> _________________________________________________________________
> How good are you in a Formula One car? Play now 
> http://server1.msn.co.in/sp05/tataracing/onlinegame.asp

[Playerstage-users] rFlex rs232 settings

[Eric Innis <eric.innis@gm...>]

1.)  Does the rflex driver handle the details of the connection such
as baud rate and flow control?
Or do I need to configure that myself via 'setserial' or similar?

2.)  Does anybody know the parameters for the rFlex board e.g. 9600-8-N-1 ?


Thanks once again,

-- 
Eric Innis
Georgia Institute of Technology
Electrical and Computer Engineering: GRITS Lab.
Aerospace Engineering: Computational Combustion Lab.
404.312.0950

Re: [Playerstage-users] amcl problem in player 2.0.2

[Brad Kratochvil <bkratochvil@et...>]

chandan datta wrote:
> Hi,
> I am not getting the same results with the amcl driver in the latest player 
> release as Player 1.6.5 .
> I subscribe to the localize and planner devices by using wavefront.cfg from 
> stage-2.0.2.
> Then view the map using playernav and give the robot a pose and a goal.The 
> robot neither moves towards the goal nor can i see the particle filters in 
> playernav GUI.
> 
> Has anyone tested it out ??

In the early 2.0.XX versions of Player the AMCL driver was not 
outputting any data on the position2d interface.  It still worked, but 
you would always have to get the hypothesis.  A week or two ago, I fixed 
this and now AMCL should be outputting the current best hypothesis on 
the position2d interface.  Since then, we've been using it for a couple 
of weeks without problems.  I'm not sure if this change made it into 
2.0.2, but CVS should work.

Best regards,

Brad