ROS navigation package for Pepper

[Jason Young]

Hi,

   Does the navigation package work well on Pepper?

   I tried to compile the Navigation package for Pepper with OpenNao, but couldn't fix the lost of dependent libs, such as PCL, VTK, OpenGL(FreeGlut). When I try to compile FreeGlut, it comes out that:  Could NOT find OpenGL (missing: OPENGL_gl_LIBRARY)

   So, can the navigation package be compiled with OpenNao? 

   And can the package work on Pepper?

   Anyone have tried this package?

[Surya Ambrose]

Hello,

I never tried the navigation package, so I can't tell if it works with Pepper :/

About compilation, when you say "compile with OpenNao" are you referring to virtual machine ? Because ROS packages are not supposed to compile on OpenNao virtual machine, so I am guessing you probably won't be able to fix all the dependency issues. Most likely, the package needs to be compiled using catkin on your machine with your ROS local installation and run on your desktop (with robot IP as a parameter).

Hope this helps,

Surya

--
You received this message because you are subscribed to the Google Groups "ROS Sig Aldebaran" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ros-sig-aldebaran+unsubscribe@googlegroups.com.
To post to this group, send email to ros-sig-aldebaran@googlegroups.com.
Visit this group at https://groups.google.com/group/ros-sig-aldebaran.
To view this discussion on the web visit https://groups.google.com/d/msgid/ros-sig-aldebaran/730f0f83-88f4-4888-b43d-0e1080f12248%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

This email and any attachment thereto are confidential and intended solely for the use of the individual or entity to whom they are addressed.
If you are not the intended recipient, please be advised that disclosing, copying, distributing or taking any action in reliance on the contents of this email is strictly prohibited. In such case, please immediately advise the sender, and delete all copies and attachment from your system.
This email shall not be construed and is not tantamount to an offer, an acceptance of offer, or an agreement by SoftBank Robotics Europe on any discussion or contractual document whatsoever. No employee or agent is authorized to represent or bind SoftBank Robotics Europe to third parties by email, or act on behalf of SoftBank Robotics Europe by email, without express written confirmation by SoftBank Robotics Europe’ duly authorized representatives.

---

Ce message électronique et éventuelles pièces jointes sont confidentiels, et exclusivement destinés à la personne ou l'entité à qui ils sont adressés.
Si vous n'êtes pas le destinataire visé, vous êtes prié de ne pas divulguer, copier, distribuer ou prendre toute décision sur la foi de ce message électronique. Merci d'en aviser immédiatement l'expéditeur et de supprimer toutes les copies et éventuelles pièces jointes de votre système.
Ce message électronique n'équivaut pas à une offre, à une acceptation d’offre, ou à un accord de SoftBank Robotics Europe sur toute discussion ou document contractuel quel qu’il soit, et ne peut être interprété comme tel. Aucun employé ou agent de SoftBank Robotics Europe n'est autorisé à représenter ou à engager la société par email, ou à agir au nom et pour le compte de la société par email, sans qu’une confirmation écrite soit donnée par le représentant légal de SoftBank Robotics Europe ou par toute autre personne ayant reçu délégation de pouvoir appropriée.

[Jason Young]

Hi Surya,

Thanks for you reply. I've tried Moveit package on Ubuntu. But Our goal is to install Ros into Pepper directly.

I have succeed in compiling ROS basic core modules on OpenNao virtual machine and installed into pepper successfully.

But I can't compile Navigation package for  lost dependencies such as OpenGL. 

在 2016年8月22日星期一 UTC+8下午6:17:56，Surya Ambrose写道：

Hello,

I never tried the navigation package, so I can't tell if it works with Pepper :/

About compilation, when you say "compile with OpenNao" are you referring to virtual machine ? Because ROS packages are not supposed to compile on OpenNao virtual machine, so I am guessing you probably won't be able to fix all the dependency issues. Most likely, the package needs to be compiled using catkin on your machine with your ROS local installation and run on your desktop (with robot IP as a parameter).

Hope this helps,

Surya

2016-08-22 10:27 GMT+02:00 Jason Young <yangt...@gmail.com>:

Hi,

   Does the navigation package work well on Pepper?

   I tried to compile the Navigation package for Pepper with OpenNao, but couldn't fix the lost of dependent libs, such as PCL, VTK, OpenGL(FreeGlut). When I try to compile FreeGlut, it comes out that:  Could NOT find OpenGL (missing: OPENGL_gl_LIBRARY)

   So, can the navigation package be compiled with OpenNao? 

   And can the package work on Pepper?

   Anyone have tried this package?

--
You received this message because you are subscribed to the Google Groups "ROS Sig Aldebaran" group.

To unsubscribe from this group and stop receiving emails from it, send an email to ros-sig-aldeba...@googlegroups.com.
To post to this group, send email to ros-sig-...@googlegroups.com.

Visit this group at https://groups.google.com/group/ros-sig-aldebaran.
To view this discussion on the web visit https://groups.google.com/d/msgid/ros-sig-aldebaran/730f0f83-88f4-4888-b43d-0e1080f12248%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Karsten K]

Hi Jason,

it would most likely be overkill to compile the complete ROS navigation stack on the robot. I would recommend to cleanly install a selection of packages where you can handle the dependencies. Move_base and gmapping should compile without any big overhead.

if it helps, I was testing gmapping on my local machine and remotely subscribed to the laser topic. It was more a proof of concept, but I could create a simple map. This would allow you to not cross compile everything.

To unsubscribe from this group and stop receiving emails from it, send an email to ros-sig-aldebaran+unsubscribe@googlegroups.com.
To post to this group, send email to ros-sig-aldebaran@googlegroups.com.

Visit this group at https://groups.google.com/group/ros-sig-aldebaran.

To view this discussion on the web visit https://groups.google.com/d/msgid/ros-sig-aldebaran/27d3a4d8-bef0-43be-8bb2-e3f0719d6a04%40googlegroups.com.

[Jason Young]

Hi Karsten,

Thanks for your advice.

I have tried gmapping on my remote PC to control Turtle_bot but not implemented with Pepper or Nao.

I think the gmapping module depends on PCL which still need OpenGL. Is there any configure option can be set to compile gmapping without OpenGL? 

在 2016年8月22日星期一 UTC+8下午11:58:27，Karsten Knese写道：

[Mikael Arguedas]

Hi Jason,

I don't think that pcl depends on openGL explicitely (c.f. http://www.pointclouds.org/documentation/tutorials/compiling_pcl_posix.php#dependencies).

Although I can't remember the last time I did try to compile it on a machine not supporting OpenGL... but you should be able to compile it on OpenNao if you're VERY patient. The remote solution suggested by Karsten is a good first step to confirm that the map quality produced is suitable for your application before going through the hassle of cross-compiling everything.

Another SLAM package you may want to consider is Cartographer recently release by Google (ros package here). Main advantages being storing only submaps in RAM and the quality of ray-tracing and resulting maps is much better. But It does depend on PCL too.

Cheers,

Mikael

To unsubscribe from this group and stop receiving emails from it, send an email to ros-sig-aldebaran+unsubscribe@googlegroups.com.
To post to this group, send email to ros-sig-aldebaran@googlegroups.com.

Visit this group at https://groups.google.com/group/ros-sig-aldebaran.

To view this discussion on the web visit https://groups.google.com/d/msgid/ros-sig-aldebaran/d6cad982-d9c9-473b-bfc6-827595e118a2%40googlegroups.com.

[Jason Yang]

Hi Mikael,

Thanks very much. I'll try it.

在 2016年8月23日星期二 UTC+8下午2:36:49，Mikael Arguedas写道：

[Vincent Rabaud]

Hi, I maintain gmapping and it does not depend on PCL.

Navigation does though. But you can use my (not yet merged) PR (and please upvote it too):

https://github.com/ros-planning/navigation/pull/458

so that you can then use the PCL-free navigation:

https://github.com/vrabaud/navigation/commit/28b39815243d12ad5b7499482ecc5b1977cf31cc

To unsubscribe from this group and stop receiving emails from it, send an email to ros-sig-aldebaran+unsubscribe@googlegroups.com.
To post to this group, send email to ros-sig-aldebaran@googlegroups.com.

Visit this group at https://groups.google.com/group/ros-sig-aldebaran.

To view this discussion on the web visit https://groups.google.com/d/msgid/ros-sig-aldebaran/f8f8657e-3dbf-4f62-8a71-2b3b111531a6%40googlegroups.com.

[Jason Yang]

Hi,

Thanks, I'll try it.

在 2016年8月23日星期二 UTC+8下午3:38:40，Vincent Rabaud写道：

[Karsten K]

I don't remember the exact quality outcome. But I remember you have to be patient and give it some love :)

Technically speaking, you have to be aware that the laser is rather limited with just a few points. So I would recommend you to move pepper in a slow pace and cross check the map outcome in rviz.

Did you encounter any latency issues?

On Wed, Sep 14, 2016 at 3:02 AM, Jason Yang <yangt...@gmail.com> wrote:

Hi Karsten,

I tested gmapping on my local machine and remotely create map with Pepper.

But the result comes out very inaccurate. Did you get satisfied result map?

在 2016年8月22日星期一 UTC+8下午11:58:27，Karsten Knese写道：

Hi Jason,

To view this discussion on the web visit https://groups.google.com/d/msgid/ros-sig-aldebaran/27d3a4d8-bef0-43be-8bb2-e3f0719d6a04%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "ROS Sig Aldebaran" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ros-sig-aldebaran+unsubscribe@googlegroups.com.
To post to this group, send email to ros-sig-aldebaran@googlegroups.com.
Visit this group at https://groups.google.com/group/ros-sig-aldebaran.

To view this discussion on the web visit https://groups.google.com/d/msgid/ros-sig-aldebaran/215eb28b-5dcd-48a2-8ef5-e883974615fd%40googlegroups.com.

[Ronald Cumbal]

Hi everyone,

I've ran slam with Pepper successfully, but the outcome is too noisy. I've tried going at a slow pace, but the map is still not clear enough. Any other recommendations as in how to improve the output map?

[adriano...@icar.cnr.it]

Hi Ronald, 

I have the same issue with mapping. To create a decent map, I teleop pepper with a slow movement and trying to give key points at gmapping package since have a high probability to understand where is the position of the robot on the map and add the new information to build the new part of the map.

I adopt the strategy to follow the walls of the rooms, the corners so if there are odometry errors the system adjust the position in the map with the estimate of the known environment.

[Giovanni Claudio]

Hi Ronald and Adriano, 

What package are you using to create the map? Which sensors are you using?

In my Pepper, I noticed that there is a problem of distortion from the depth sensor. The surface in the following picture should be flat (Pepper here was facing a wall): 

Is this normal? Also Paco has the same problem.

To unsubscribe from this group and stop receiving emails from it, send an email to ros-sig-aldebaran+unsubscribe@googlegroups.com.
To post to this group, send email to ros-sig-aldebaran@googlegroups.com.

Visit this group at https://groups.google.com/group/ros-sig-aldebaran.

To view this discussion on the web visit https://groups.google.com/d/msgid/ros-sig-aldebaran/85e9805c-a5b7-450c-9c51-328de1339132%40googlegroups.com.

[Natalia Lyubova]

Hello Giovanni,

right, the distortion is caused by the lenses but not the sensor... Thus, some researchers remove lenses. Please let us know if it is critical for you.

Best regards,
Natalia

[Giovanni Claudio]

Hi Natalya, 

Thanks for your answer, I thought it was that. Are you referring to the black lens of the eyes right? Is it easy to remove them?

Giovanni

To unsubscribe from this group and stop receiving emails from it, send an email to ros-sig-aldebaran+unsubscribe@googlegroups.com.
To post to this group, send email to ros-sig-aldebaran@googlegroups.com.

Visit this group at https://groups.google.com/group/ros-sig-aldebaran.

To view this discussion on the web visit https://groups.google.com/d/msgid/ros-sig-aldebaran/6a09f3fa-5232-461b-b9d1-7f2086ee962f%40googlegroups.com.

[Natalia Lyubova]

Right, it is caused by the lenses in the eyes. it is possible to remove those lenses and we can do it.
Please, let me know,

Natalia

[adriano...@icar.cnr.it]

HI Giovanni,

I'm using gmapping package with laser sensor. (Topic / pepper_robot / laser from pepper_full.launch).

Adriano

[Giovanni Claudio]

Thanks Natalia.

Right now I am not working with SLAM, so it is not a big problem. I will discuss with my colleagues if we need to remove them for future works.

Giovanni

[Ronald Cumbal]

Hi Adriano and Giovani!

Thanks for the suggestions. I'm also using gmapping package with laser sensor /pepper_robot/laser from the C++ bring-up.

I have built a map with Pepper, but I had to fix some details manually:

These are the parameters I'm using in the launch file of Gmapping:

<launch>
    <node pkg="gmapping" type="slam_gmapping" name="slam_gmapping" output="screen">
        <remap from="/scan" to="/pepper_robot/laser"/>
        <param name="map_update_interval" value="0.1"/>
        <param name="maxRange" value="1.2"/>
        <param name="maxUrange" value="1.4"/>
        <param name="minimumScore" value="100.0"/>
        <param name="linearUpdate" value="0.1"/>
        <param name="angularUpdate" value="0.5"/>
        <param name="temporalUpdate" value="0.1"/>
        <param name="resampleThreshold" value="0.5"/>
        <param name="particles" value="100"/>
        <param name="xmin" value="-25.0"/>
        <param name="ymin" value="-25.0"/>
        <param name="xmax" value="25.0"/>
        <param name="ymax" value="25.0"/>
        <param name="delta" value="0.02"/>
    </node>

</launch>

I also checked odometry and laser readings with the suggestions from http://wiki.ros.org/navigation/Tutorials/Navigation%20Tuning%20Guide. I found that translation odometry is correct, but odometry for orientation has a problem with the laser on the left. This is the image I got after performing the test, see that the laser points do not match consistently in the left of the image.

I also tried running AMCL and move Pepper around the map by using 2D Navigation Goal, but I encounter two problems:
- Pepper starts moving, but stops before reaching the goal without any error reported anywhere.
- Pepper moves toward the goal, but it reaches a different part of the real hallway than the virtual map.

So I'm stuck with the following doubts:

I fixed manually some problem of the map, would this cause problems while navigating?
Is there any way to correct/calibrate laser readings?
The biggest problem I encounter is the lack of features inside the hallway for localization. Any suggestion to improve localization?
My next step is to navigate around the map with different planning algorithms. Any suggestions where  can start?

Thanks for all the help again.

--
You received this message because you are subscribed to a topic in the Google Groups "ROS Sig Aldebaran" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/ros-sig-aldebaran/FEtO4Gf9erc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to ros-sig-aldebaran+unsubscribe@googlegroups.com.

To post to this group, send email to ros-sig-aldebaran@googlegroups.com.
Visit this group at https://groups.google.com/group/ros-sig-aldebaran.

To view this discussion on the web visit https://groups.google.com/d/msgid/ros-sig-aldebaran/CABLAssmVtKUW7i-dmQ%2B3tmVn99Uz%2BJZZNd-h%3Df9MzWVrXAX42A%40mail.gmail.com.

For more options, visit https://groups.google.com/d/optout.

--

MSc Candidate in Mechanical Engineering - System and Control 2015-2017

Faculty Mechanical, Maritime and Materials Engineering

Delft University of Technology

Tel: +31 (0)651475643

Email: ronald.c...@gmail.com

           ronald....@hotmail.com

[Paco Dupont]

Hi Ronald,

The robot stop because of the front sonar that is flickering. I've the problem on two Pepper v1.6 here, the sonar regularly return a low range whereas there is no obstacle in front of the robot. The lower module managing velocity cmd on NaoQi then stop the robot to avoid collision and then refuse any incoming forward cmd.

This create a lots of mess with navigation programming ...

To view this discussion on the web visit https://groups.google.com/d/msgid/ros-sig-aldebaran/85e9805c-a5b7-450c-9c51-328de1339132%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to a topic in the Google Groups "ROS Sig Aldebaran" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/ros-sig-aldebaran/FEtO4Gf9erc/unsubscribe.

To unsubscribe from this group and all its topics, send an email to ros-sig-aldeba...@googlegroups.com.

To post to this group, send email to ros-sig-...@googlegroups.com.
Visit this group at https://groups.google.com/group/ros-sig-aldebaran.

To view this discussion on the web visit https://groups.google.com/d/msgid/ros-sig-aldebaran/CABLAssmVtKUW7i-dmQ%2B3tmVn99Uz%2BJZZNd-h%3Df9MzWVrXAX42A%40mail.gmail.com.

For more options, visit https://groups.google.com/d/optout.

[Paco Dupont]

Hi Natalia,

Can we remove the lenses without warranty void, or we must return Pepper to Aldebaran ?

Best regards,
Paco.

[Natalia Lyubova]

Hello Paco,

it is better to ask our support before : https://account.aldebaran.com/support/
I am not sure what happens with warranty then, however we have removed them for some of our partners on their request.


Best regards,
Natalia

[Paco Dupont]

Hi Adriano and Ronaldo,

can you share the configuration you used for Pepper with gmapping ? I'm experiencing some strange behavior between the map and the odom frame during the mapping process. My odom frame got some drifting will moving Pepper without gmapping, but nothing unusual.

But when I run gmapping the map and the odom frame are moving from each other (this is due to the odometry correction from gmapping I think but I've some configuration issue so). On this screen the robot run for 3-4 meters and the wall on the left of the robot is the same as the one near the map referencial.

Best regards,
Paco.

[Ronald Cumbal]

Hey Paco,

This is the configuration of my launch file:

<launch>
    <node pkg="gmapping" type="slam_gmapping" name="slam_gmapping" output="screen">
        <remap from="/scan" to="/pepper_robot/laser"/>

        <param name="base_frame"            value="base_footprint" />
        <param name="map_frame"             value="map"/>
        <param name="odom_frame"            value="odom"/>

        <param name="map_update_interval"   value="0.1"/>

        <param name="minimumScore"          value="100"/>

        <param name="linearUpdate"          value="0.1"/>

        <param name="angularUpdate"         value="0.1"/>
        <param name="temporalUpdate"        value="-1.0"/>

        <param name="resampleThreshold"     value="0.5"/>
        <param name="particles"             value="100"/>
        <param name="xmin"                  value="-25.0"/>
        <param name="ymin"                  value="-25.0"/>
        <param name="xmax"                  value="25.0"/>
        <param name="ymax"                  value="25.0"/>
        <param name="delta"                 value="0.02"/>

        <param name="maxRange"              value="1.4"/>
        <param name="maxUrange"             value="1.2"/>
    </node>

</launch>

I have notice a big problem with the Rotational Odometry of Pepper. A clear offset error is accumulated with every rotation. Anyone would know how to calibrate this error?

--
You received this message because you are subscribed to a topic in the Google Groups "ROS Sig Aldebaran" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/ros-sig-aldebaran/FEtO4Gf9erc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to ros-sig-aldebaran+unsubscribe@googlegroups.com.
To post to this group, send email to ros-sig-aldebaran@googlegroups.com.

Visit this group at https://groups.google.com/group/ros-sig-aldebaran.

To view this discussion on the web visit https://groups.google.com/d/msgid/ros-sig-aldebaran/75d517f6-f167-48de-ba3c-64416117d6b3%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--

[Paco Dupont]

Thanks a lot Ronald I've got better results with you parameters, I think my minimumScore was too low and my angularUpdate too high.

I'll give a try with AMCL and move_base now.

Thanks again,
Paco.

To unsubscribe from this group and all its topics, send an email to ros-sig-aldeba...@googlegroups.com.

To post to this group, send email to ros-sig-...@googlegroups.com.
Visit this group at https://groups.google.com/group/ros-sig-aldebaran.

To view this discussion on the web visit https://groups.google.com/d/msgid/ros-sig-aldebaran/75d517f6-f167-48de-ba3c-64416117d6b3%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[adriano...@icar.cnr.it]

Hi Paco and Ronald, 

My configuration parameters are similar, I've been conducted several experiments with different parameters, but I had some problem to understand what are the best parameter set up. 

I wonder if I increment the number of particles (i.e. 300+) I will have better results. 

I would like to have a detailed map because the environment is composed by hallways and rooms similar each other, and the details give a lot of information for the localisation accuracy.

Adriano

To unsubscribe from this group and all its topics, send an email to ros-sig-aldeba...@googlegroups.com.

To post to this group, send email to ros-sig-...@googlegroups.com.
Visit this group at https://groups.google.com/group/ros-sig-aldebaran.

To view this discussion on the web visit https://groups.google.com/d/msgid/ros-sig-aldebaran/75d517f6-f167-48de-ba3c-64416117d6b3%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[Vincent Rabaud]

Have you guys given a try to Catographer ? https://google-cartographer.readthedocs.io/en/latest/

I just don't want to maintain gmapping anymore :)

To unsubscribe from this group and stop receiving emails from it, send an email to ros-sig-aldebaran+unsubscribe@googlegroups.com.
To post to this group, send email to ros-sig-aldebaran@googlegroups.com.

Visit this group at https://groups.google.com/group/ros-sig-aldebaran.

To view this discussion on the web visit https://groups.google.com/d/msgid/ros-sig-aldebaran/6b011847-043d-42be-b945-a537a0abef5a%40googlegroups.com.

[Paco Dupont]

Hi Vincent,
I tried cartographer last mounth but I was stuck with this Error :

https://groups.google.com/forum/?nomobile=true#!topic/ros-sig-aldebaran/4IfKl35wGX8

> Best regards,
> Paco.
>
>
>
> Le jeudi 30 mars 2017 11:53:55 UTC+2, adriano...@icar.cnr.it a écrit :
> HI Giovanni,
> I'm using gmapping package with laser sensor. (Topic / pepper_robot / laser from pepper_full.launch).
>
>
>
>
>
> Adriano
>
>
>
>
> Il giorno giovedì 30 marzo 2017 11:03:12 UTC+2, Giovanni Claudio ha scritto:
> Hi Ronald and Adriano, 
>
>
> What package are you using to create the map? Which sensors are you using?
>
>
> In my Pepper, I noticed that there is a problem of distortion from the depth sensor. The surface in the following picture should be flat (Pepper here was facing a wall): 
>
>
>
>
>
>

> Is this normal? Also Paco has the same problem.
>
>
>
>
> 2017-03-30 10:54 GMT+02:00 <adriano...@icar.cnr.it>:
>
> Hi Ronald, 
>
>
>
> I have the same issue with mapping. To create a decent map, I teleop pepper with a slow movement and trying to give key points at gmapping package since have a high probability to understand where is the position of the robot on the map and add the new information to build the new part of the map.
> I adopt the strategy to follow the walls of the rooms, the corners so if there are odometry errors the system adjust the position in the map with the estimate of the known environment.
>
>
>
>
>
>
>
>
> Il giorno martedì 28 marzo 2017 16:33:52 UTC+2, Ronald Cumbal ha scritto:
> Hi everyone,
>
> I've ran slam with Pepper successfully, but the outcome is too noisy. I've tried going at a slow pace, but the map is still not clear enough. Any other recommendations as in how to improve the output map?
>
> On Wednesday, September 21, 2016 at 8:20:48 AM UTC+2, Karsten Knese wrote:
> I don't remember the exact quality outcome. But I remember you have to be patient and give it some love :)
> Technically speaking, you have to be aware that the laser is rather limited with just a few points. So I would recommend you to move pepper in a slow pace and cross check the map outcome in rviz.
>
>
> Did you encounter any latency issues?
>
>
> On Wed, Sep 14, 2016 at 3:02 AM, Jason Yang <yangt...@gmail.com> wrote:
>
> Hi Karsten,
> I tested gmapping on my local machine and remotely create map with Pepper.
> But the result comes out very inaccurate. Did you get satisfied result map?
>
> 在 2016年8月22日星期一 UTC+8下午11:58:27，Karsten Knese写道：
> Hi Jason,
>
>
> it would most likely be overkill to compile the complete ROS navigation stack on the robot. I would recommend to cleanly install a selection of packages where you can handle the dependencies. Move_base and gmapping should compile without any big overhead.
>
>
> if it helps, I was testing gmapping on my local machine and remotely subscribed to the laser topic. It was more a proof of concept, but I could create a simple map. This would allow you to not cross compile everything.
>
>
>
>
> On Mon, Aug 22, 2016 at 5:12 AM, Jason Young <yangt...@gmail.com> wrote:
>
>
>
>

> Hi Surya,Thanks for you reply. I've tried Moveit package on Ubuntu. But Our goal is to install Ros into Pepper directly.

> To unsubscribe from this group and stop receiving emails from it, send an email to ros-sig-aldeba...@googlegroups.com.
>
> To post to this group, send email to ros-sig-...@googlegroups.com.
>
> Visit this group at https://groups.google.com/group/ros-sig-aldebaran.
>

[Syed Ali Raza]

Hi,

Has anyone solved the odometry issue?

I am using gmapping with the same configuration like others and using pepper's laserscan.

I moved the robot very carefully and avoided unnecessary rotations. Still, the resulting map is very bad.

Any suggestions? Thanks!

Cheers

Ali

[HaSu]

My gmapping attempts were rather poor either so far. I just tried the parameters suggested here and we are finally getting somewhere.

I think it helps to merge all the laser scanners together onto one topic. there are several tools for this, if anyone is interested I can post a link to the one we currently use.

Cartographer was never tested with such little amount of points apparently so I'm not sure if it would work at all - at least that's what I've been told. 

One thing one can do to improve gmapping a bit is to record a bag file and play it back at half the speed, while running gmapping. It usually provides a better map.

Question: What are the experiences so far concerning the range of the laser scanners? I remember reading somewhere that they are usable up to 5m but I am not sure? Does anyone know? Or, with what range are they still usable for navigation?

cheers!

HaSu

[luka.fua2]

Back there, when worked with gmapping. Used 3D camera as sensor for building map. Ther is a package for transforming data from camera to point cloud data and you should change in launch file input topic to gmapping.

Luka K.

Poslano s pametnog telefona Samsung Galaxy.

-------- Izvorna poruka --------

Šalje: HaSu <haw...@informatik.uni-bremen.de>

Datum: 20.09.2018. 19:18 (GMT+01:00)

To: ROS Sig Aldebaran <ros-sig-...@googlegroups.com>

Naslov: [aldebaran] Re: ROS navigation package for Pepper

--
You received this message because you are subscribed to the Google Groups "ROS Sig Aldebaran" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ros-sig-aldeba...@googlegroups.com.
To post to this group, send email to ros-sig-...@googlegroups.com.
Visit this group at https://groups.google.com/group/ros-sig-aldebaran.

To view this discussion on the web visit https://groups.google.com/d/msgid/ros-sig-aldebaran/8633e248-5a1a-4962-8342-fb9ad376ecbe%40googlegroups.com.

[Ronald Cumbal]

I remember that my lasers would work well up to 2 meters. Nevertheless, you can modify this values in the ROS Bridge reading these lasers. The choice given by Luka K. is your best option, although I recall head-motion would introduce a lot of noise. I also used the 3D camera for localization with AMCL and it worked much better. 

These are the parameters I used at the end:

<!--- Start Gmappign node -->
	<node pkg="gmapping" type="slam_gmapping" name="slam_gmapping" output="screen" launch-prefix="xterm -e">

	<remap from="/scan" to="/pepper_robot/laser"/>
	<param name="base_frame" value="base_footprint" />
	<param name="map_frame" value="map"/>
	<param name="odom_frame" value="odom"/>
	<param name="map_update_interval" value="0.1"/>
	<param name="minimumScore" value="100"/>
	<param name="linearUpdate" value="0.1"/>
	<param name="angularUpdate" value="0.1"/>
	<param name="temporalUpdate" value="-1.0"/>
	<param name="resampleThreshold" value="0.5"/>

	<param name="particles" value="1000"/>
	<param name="xmin" value="-5.0"/>
	<param name="ymin" value="-5.0"/>
	<param name="xmax" value="5.0"/>
	<param name="ymax" value="5.0"/>

	<param name="delta" value="0.02"/>
	<param name="maxRange" value="1.4"/>

	<param name="maxUrange" value="1.5"/>
	</node>

 

Ronald C.

--
You received this message because you are subscribed to a topic in the Google Groups "ROS Sig Aldebaran" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/ros-sig-aldebaran/FEtO4Gf9erc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to ros-sig-aldeba...@googlegroups.com.

To post to this group, send email to ros-sig-...@googlegroups.com.
Visit this group at https://groups.google.com/group/ros-sig-aldebaran.
To view this discussion on the web visit https://groups.google.com/d/msgid/ros-sig-aldebaran/8633e248-5a1a-4962-8342-fb9ad376ecbe%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Mechanical - Control Engineer

Tel: +593 (0)960443409

Email: ronald.c...@gmail.com

           ronald....@hotmail.com