Ask Your Question

philglau's profile - activity

2019-05-20 02:21:47 -0500 received badge  Enlightened (source)
2019-05-20 02:21:47 -0500 received badge  Guru (source)
2019-04-05 20:37:09 -0500 received badge  Great Answer (source)
2018-02-01 15:44:28 -0500 received badge  Notable Question (source)
2018-01-31 04:26:31 -0500 received badge  Good Answer (source)
2017-12-26 06:41:53 -0500 received badge  Famous Question (source)
2017-12-26 06:41:53 -0500 received badge  Notable Question (source)
2017-10-07 00:56:57 -0500 received badge  Nice Answer (source)
2017-09-24 16:00:33 -0500 received badge  Popular Question (source)
2017-09-02 08:03:53 -0500 received badge  Nice Question (source)
2017-03-29 20:53:03 -0500 received badge  Famous Question (source)
2017-03-27 23:47:31 -0500 asked a question Debugging tf problems Tutorial - Typos/errors

There seem to be some errors in the 'debugging tf' tutorial, some of which references the previous tf intro tutorials. My guess is this is a hangover from an earlier version of the debugging_tf tutorials. I'm using a 'kinetic' install

1)roslaunch turtle_tf start_debug_demo.launch is launched to demonstrate the problem, but on my kinetic install, this 'start_debug_demo' file seems to reference python files rather than C++. Additionally, later in the tutorial, it's called with roslaunch learning_tf start_debug_demo.launch but the file doesn't exist on the path learning_tf.

Assuming that we should be using the C++ version of the launch file and that the user has just finished the previous learning_tf tutorials in C++, I would recommend copying the existing 'start_demo.launch' file to 'start_debug_demo.launch' and adding some steps in Section 1 for adapting it by indicating that:

<node name="turtle_pointer" pkg="turtle_tf" type="turtle_tf_listener" respawn="false" output="screen" >

should be changed to:

<node name="turtle_pointer" pkg="turtle_tf" type="turtle_tf_listener_debug" respawn="false" output="screen" >

2) Likewise, the CMake file needs to be adjusted in the learning_tf directory and have the following added. This step isn't mentioned and for people new to C++ might be confusing if omitted:

add_executable(turtle_tf_listener_debug src/turtle_tf_listener_debug.cpp)
target_link_libraries(turtle_tf_listener_debug ${catkin_LIBRARIES})

3) on Step 4, it indicates:

$ make
$ roslaunch learning_tf start_debug_demo.launch

I found that this did not work. I believe it should be (as appears to be correctly spelled out in Step 5):

Re-build your package at the top folder of your catkin workspace
$ catkin_make
$ roslaunch learning_tf start_debug_demo.launch

Also, roslaunch learning_tf start_debug_demo.launch will not work unless some steps are taken (as suggested in #1 above) to make sure that a file 'start_debug_demo' exists in the learning_tf directory.

Overall, a nice tutorial. Thank you to those who worked to make it in the first place. Please don't take my proofreading as criticism, it is just to help the next person who comes along the tf tutorial path.

2017-03-20 19:14:48 -0500 received badge  Notable Question (source)
2017-03-20 03:09:57 -0500 received badge  Popular Question (source)
2017-03-20 02:25:41 -0500 received badge  Self-Learner (source)
2017-03-20 02:25:41 -0500 received badge  Teacher (source)
2017-03-19 21:40:08 -0500 received badge  Popular Question (source)
2017-03-19 18:34:08 -0500 received badge  Student (source)
2017-03-19 17:49:17 -0500 answered a question python module empy missing - Tutorials

I think this has to do with a conflict between the python module 'em' and 'empy' both of which use the same name.

If you have 'em' installed, it seems to supersede 'empy' which is what is needed by ROS.

See this link for details

I ended up compiling from source before I figured this out. To resolve it on my self-compiled kinetic, I did:

pip uninstall em
pip install empy

And it seems to have fixed the problem. Kind of a frustrating error as the apt-get install covered in the tutorial doesn't automatically detect this conflict.

2017-03-19 16:28:28 -0500 received badge  Scholar (source)
2017-03-19 16:28:22 -0500 commented answer /opt/ros/kinetic/setup.bash file missing after reinstall of kinetic

I should have been clearer above I had actually run sudo apt autoremove prior to running the rm -fr commands. After running sudo apt autoremove, I was still left with a /opt/ros/kenetic/ directory, that was mostly empty except for folders, thus the rm -fr to get rid of final remnants.

2017-03-19 16:23:11 -0500 commented answer /opt/ros/kinetic/setup.bash file missing after reinstall of kinetic

Thank you for the advice regarding using remove and purge. I will use that going forward. I ran sudo apt-get install --reinstall ros-kinetic-catkin and it does appear to have placed the missing files into /opt/ros/kinetic/. Thank you.

2017-03-18 19:39:10 -0500 asked a question /opt/ros/kinetic/setup.bash file missing after reinstall of kinetic

I'm using Ubuntu 16.04 I had ROS Kinetic installed following the tutorial and it was working. I was experiencing errors with a tutorial, covered here, so I removed ROS in order to do a complete fresh install. I did so by using:

sudo apt-get remove ros-kinetic-desktop-full 
sudo apt-get remove python-rosinstall
sudo rm -fr /etc/ros
sudo rm -fr /opt/ros
sudo rm -fr ~/.ros
sudo apt-get autoremove

To purge my system of my first ROS install.

I then reapplied all the directions given in the installation tutorial.

It seems however the the 'setup.bash' file is not installed the second time around. When I ls /opt/ros/kinetic I get only the following:

philglau@phil-glau-ubuntu-16:/opt/ros/kinetic$ ls
bin  etc  include  lib  share

The 'setup.bash' script is missing !! (using du -hs indicates there are 145MB of other files in this /kinetic folder, so it's not like it didn't install most of the stuff.) Furthermore, when I installed the second time, it gave absolutely no errors or warning to indicate that something was missing or wrong.

This is from dpkg -l

philglau@phil-glau-ubuntu-16:~$ dpkg -l | grep ros-kinetic-catkin
ii  ros-kinetic-catkin                           0.7.6-0xenial-20170217-212821-0800                    amd64        Low-level build system macros and infrastructure for RO

Can't figure out why this is. Did I do something incorrect when uninstalling the original version like forget some other cache or directory??

Is it possible to just manually install the missing setup.bash script? If so where would I get it?

Thank you in advance for your help.

2017-03-18 17:22:26 -0500 asked a question python module empy missing - Tutorials

I'm walking through the tutorials and am on Introduction to msg and srv

I'm using Ubuntu 16.04 and Kinetic. My python is /usr/bin/python My pip is /usr/local/bin/pip

I'm running the command catkin_make install on Step 6 and I get the following 'em' not found error:

[ 69%] Generating Javascript code from beginner_tutorials/Num.msg
[ 76%] Generating C++ code from beginner_tutorials/Num.msg
[ 84%] Generating Javascript code from beginner_tutorials/AddTwoInts.srv
Traceback (most recent call last):
  File "/opt/ros/kinetic/share/gencpp/cmake/../../../lib/gencpp/", line 41, in <module>
    import genmsg.template_tools
  File "/opt/ros/kinetic/lib/python2.7/dist-packages/genmsg/", line 39, in <module>
    import em
ImportError: No module named em
beginner_tutorials/CMakeFiles/beginner_tutorials_generate_messages_cpp.dir/build.make:63: recipe for target '/home/philglau/catkin_ws/devel/include/beginner_tutorials/Num.h' failed
make[2]: *** [/home/philglau/catkin_ws/devel/include/beginner_tutorials/Num.h] Error 1
make[2]: *** Waiting for unfinished jobs....
Traceback (most recent call last):
  File "/opt/ros/kinetic/share/gencpp/cmake/../../../lib/gencpp/", line 41, in <module>
    import genmsg.template_tools
  File "/opt/ros/kinetic/lib/python2.7/dist-packages/genmsg/", line 39, in <module>
    import em
ImportError: No module named em
beginner_tutorials/CMakeFiles/beginner_tutorials_generate_messages_cpp.dir/build.make:70: recipe for target '/home/philglau/catkin_ws/devel/include/beginner_tutorials/AddTwoInts.h' failed

So obviously, it thinks em is missing. So I ran both of the following to try to resolve it:

philglau@phil-glau-ubuntu-16:~$ sudo -H pip install empy
Requirement already satisfied: empy in /usr/lib/python2.7/dist-packages

And also tried:

philglau@phil-glau-ubuntu-16:~$ sudo apt-get install python-empy
[sudo] password for philglau: 
Reading package lists... Done
Building dependency tree       
Reading state information... Done
python-empy is already the newest version (3.3.2-1build1).
python-empy set to manually installed.
The following packages were automatically installed and are no longer required:
  linux-headers-4.4.0-59 linux-headers-4.4.0-59-generic linux-image-4.4.0-59-generic linux-image-extra-4.4.0-59-generic
Use 'sudo apt autoremove' to remove them.
0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.

And if I manually load python from the CLI, I can import em without it barking

philglau@phil-glau-ubuntu-16:~/catkin_ws$ python
Python 2.7.12 (default, Nov 19 2016, 06:48:10) 
[GCC 5.4.0 20160609] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import em

Likewise if I directly run from the CLI it doesn't bark either:

python /opt/ros/kinetic/lib/python2.7/dist-packages/genmsg/

I also 'reinstalled' ROS and it said 'nothing was updated.' I have not yet attempted to completely uninstall and then re-install, but hopefully that won't be necessary.

After all my attempts to fix this problem, I'm still getting the package not found error when running catkin_make as indicated in the tutorial. Any thoughts??

Thank you in advance for any help. - Phil

2017-03-17 01:11:19 -0500 commented answer rospack depends1 error, rosdep version?

Ditto. I'm not interested in giving up anaconda as a "fix". First I used "which pip" to make sure my pip was the anaconda version and then using 'pip install rosdep' resolved this for me as well.

2014-09-08 23:48:01 -0500 received badge  Supporter (source)
2013-11-06 12:39:47 -0500 received badge  Famous Question (source)
2013-09-10 20:02:27 -0500 received badge  Notable Question (source)
2013-09-04 06:28:37 -0500 received badge  Popular Question (source)
2013-08-29 20:35:14 -0500 asked a question Hokuyo Node Tutorial adapted to Groovy


The HokuyoNode tutorial seems to have been written for Fuerte, I'm running Groovy (and started learning with it, so have no knowledge of Fuerte)

(I don't have enough karma to publish the link, so I abbreviated it.)

The tutorial seems to work fine until 1.7 where the following fuerte command (??) is given:

rosrun rviz rviz -d `rospack find hokuyo_node`/hokuyo_test.vcg

This results in an error:

ERROR: the config file '/opt/ros/groovy/stacks/laser_drivers/hokuyo_node/hokuyo_test.vcg' is a vcg file, which is the old rviz config format.
New config files have a .rviz extension and use YAML formatting. The format changed between Fuerte and Groovy. There is not (yet) an automated conversion program.

What is the Groovy command equivalent of what is given in the tutorial that I should run on Groovy in order to get the laser scan to show up in rviz as shown in the tutorial ?

PS. I was able to get the Hokuyo Scanner up and it appears to be streaming. I ran

rosrun hokuyo_node hokuyo_node

It connected and indicates "Streaming Data"

Thanks for your time.

2013-08-29 20:32:40 -0500 asked a question can't post question?

I can't seem to post a question. This is just a test.