Homer Manalo's profile - activity

When using smach with actionlib actions, am I correct in saying that smach acts as the action client for the action server? So I don't need a separate action client anymore?

Does "sudo apt-get remove ros-cturtle-ros" also removes other packages in my home folder? If not, could I just manually delete them or is there a recommended method?

How do you update individual stacks inside the installation directory(/opt/ros/cturtle/stacks)? Say I want to update the geometry stacks because of eigen compatibility with eigen3. I thought updating apt will suffice but it didn't.

Is the ff. possible on the navigation stack:

• Instead of going left or right when an obstacle in front is detected, it just stops and uses the vertical(z) space to move. Is it possible to modify it like this?
• The robot is not wheeled(thrusters instead), but moves by differential drive.
• Is it also possible to replace the laser with a stereo camera instead?

If it is not possible, what are alternatives?

The actionlib accepts only one goal at a time. Does it mean I cannot do this:

#goal
int32[] goal
or
#goal
int32 goal1
int32 goal2

I cannot find any clickable under the posts.

I have some action servers which will run sequentially through smach. I will have a launch file that simultaneously launches these action servers. These action servers will publish a common feedback topic say motor_velocity. If I am not mistaken the topic is published as '/action_server_name/feedback/motor_velocity'. How do I subscribe to that topic without knowing which state is running? The current state that is run by smach should be the server that the subscriber should get the motor_velocity.

• Can I just subscribe to '/feedback/motor_velocity'? Since I launched the servers simultaneously aren't they simultaneously publishing it and may just cause confusion?
• Is it possible to remap the feedback topic?
• What is a better way of doing this?

I just edited out the stereo_nodelet.launch that comes with the uvc_camera package and then added a couple of nodes:

<launch>
  <node pkg="nodelet" type="nodelet" name="camera_process" args="manager" output="screen"/>

  <node pkg="nodelet" type="nodelet" name="uvc_camera_stereo" args="load uvc_camera/Stereo /camera_process" output="screen">
    <param name="width" type="int" value="320" />
    <param name="height" type="int" value="240" />
    <param name="fps" type="int" value="60" />
    <param name="frame" type="string" value="narrow_stereo" />
    <param name="left/device" type="string" value="/dev/video0" />
    <param name="right/device" type="string" value="/dev/video1" />
    <param name="left/camera_info_url" value="file://$(find stereo_pseye)/params/ps_eye_left.yaml" />
    <param name="right/camera_info_url" value="file://$(find stereo_pseye)/params/ps_eye_right.yaml" />
    <remap from="/left/image_raw" to="/stereo/left/image_raw"/>
    <remap from="/right/image_raw" to="/stereo/right/image_raw"/>
    <remap from="/left/camera_info" to="/stereo/left/camera_info"/>
    <remap from="/right/camera_info" to="/stereo/right/camera_info"/>
  </node>

  <node ns="stereo" name="stereo_image_proc" pkg="stereo_image_proc" type="stereo_image_proc">
        <remap from="stereo_image_proc" to="proc" />
  </node>
  <node name="dynamic_reconfigure_load" pkg="dynamic_reconfigure" type="dynparam" args="load /stereo/proc $(find stereo_pseye)/params/ps_eye_stereo.yaml" />
</launch>

And here is the output from launching:

... logging to /home/roboteknik/.ros/log/901b5ebe-600e-11e0-95df-00012e2bb8e1/roslaunch-roboteknik-desktop-29965.log
Checking log directory for disk usage. This may take awhile.
Press Ctrl-C to interrupt
Done checking log file disk usage. Usage is <1GB.

started roslaunch server http://roboteknik-desktop:52183/

SUMMARY
========

PARAMETERS
 * /rosdistro
 * /uvc_camera_stereo/left/device
 * /uvc_camera_stereo/right/camera_info_url
 * /uvc_camera_stereo/left/camera_info_url
 * /rosversion
 * /uvc_camera_stereo/frame
 * /uvc_camera_stereo/right/device
 * /uvc_camera_stereo/width
 * /uvc_camera_stereo/fps
 * /uvc_camera_stereo/height

NODES
  /stereo/
    stereo_image_proc (stereo_image_proc/stereo_image_proc)
  /
    camera_process (nodelet/nodelet)
    uvc_camera_stereo (nodelet/nodelet)
    dynamic_reconfigure_load (dynamic_reconfigure/dynparam)

ROS_MASTER_URI=http://localhost:11311

core service [/rosout] found
process[camera_process-1]: started with pid [29983]
process[uvc_camera_stereo-2]: started with pid [29989]
process[stereo/stereo_image_proc-3]: started with pid [29991]
process[dynamic_reconfigure_load-4]: started with pid [30015]
[ INFO] [1302067659.004025404]: Loading nodelet /uvc_camera_stereo of type uvc_camera/Stereo to manager /camera_process with the following remappings:
[ INFO] [1302067659.004331790]: /left/camera_info -> /stereo/left/camera_info
[ INFO] [1302067659.004419777]: /left/image_raw -> /stereo/left/image_raw
[ INFO] [1302067659.004486763]: /right/camera_info -> /stereo/right/camera_info
[ INFO] [1302067659.004560530]: /right/image_raw -> /stereo/right/image_raw
[ERROR] [1302067659.048475968]: Failed to load nodelet [/uvc_camera_stereo] of type [uvc_camera/Stereo]: According to the loaded plugin descriptions the class uvc_camera/Stereo with base class type nodelet::Nodelet does not exist. Declared types are  image_proc/debayer image_proc/rectify image_view/disparity image_view/image pcl/BAGReader pcl/BoundaryEstimation pcl/ConvexHull2D pcl/EuclideanClusterExtraction pcl/ExtractIndices pcl/ExtractPolygonalPrismData pcl/FPFHEstimation pcl/FPFHEstimationOMP pcl/MomentInvariantsEstimation pcl/MovingLeastSquares pcl/NodeletDEMUX pcl/NodeletMUX pcl/NormalEstimation pcl/NormalEstimationOMP pcl/NormalEstimationTBB pcl/PCDReader pcl/PCDWriter pcl/PFHEstimation pcl/PassThrough pcl/PointCloudConcatenateDataSynchronizer pcl/PointCloudConcatenateFieldsSynchronizer pcl/PrincipalCurvaturesEstimation pcl/ProjectInliers pcl/SACSegmentation pcl/SACSegmentationFromNormals pcl/SegmentDifferences pcl/StatisticalOutlierRemoval pcl/TestListener pcl/TestPingPong pcl/TestTalker pcl/VFHEstimation pcl/VoxelGrid stereo_image_proc/disparity stereo_image_proc/point_cloud stereo_image_proc/point_cloud2 test_nodelet/ConsoleTest test_nodelet/Plus uvc_camera/CameraNodelet uvc_camera/StereoNodelet
Waiting for service /stereo/proc/set_parameters...
[dynamic_reconfigure_load-4] process has finished cleanly.
log file: /home/roboteknik/.ros/log/901b5ebe-600e-11e0-95df-00012e2bb8e1/dynamic_reconfigure_load-4*.log
[uvc_camera_stereo-2] process has finished cleanly.
log file: /home/roboteknik/.ros/log/901b5ebe-600e-11e0-95df-00012e2bb8e1/uvc_camera_stereo-2*.log

Is there a way to automatically run a node using/inside a C++/Python program? Say some event happened and I want to run this node, then another event happened and I want to close the node. Or perhaps a launch file?

When tying to run rviz I get a bunch of this message:

Xlib:  extension "GLX" missing on display ":0.0".

What is causing this? The rviz splash screen just flash in a sec. Using diamondback on Ubuntu 10.10.

Why can't I run the camera calibration? I am using diamondback in Ubuntu 10.10. I get the ff:

Waiting for service /stereo/left/set_camera_info ...
Service not found
Waiting for service /stereo/right/set_camera_info ...
Service not found

I have some questions about using SimpleActionState with Smach:

• How does one define/set the action as succeeded, preempted, or aborted? Or is this set in the actionlib server using set_succeeded, etc.?
• If this is set using the actionlib server can I still have a custom outcome? If yes, how?
• Do I still need to define a new class or function to handle the states or do I just have to add the states (using StateMachine.add)?
• Do I need to manually run actionlib servers or is this automatically run by smach?

The recommended(8x6) calibration board is so big they won't print on a normal printer. How do you make yours? I'm using a white illustration board then pasting cut-outs of black art papers. In the long run they tend to be curved and my cuts are not that precise which is not ideal in calibration. I'm looking for a better way to have a longer lasting calibration board.

How do you read a PointCloud2 in C++? I'm interested in getting the depth(z point). Maybe an equivalent of this: reading pointcloud2 in python.

How big is the full installation? I currently have <5GB of free space on my root partition (the cturtle on my system is about 3.9GB). Is it safe to mount the /opt folder to another partition?

Is there a package that provides visual odometry aside from vslam? I found visodo but it has a missing/deleted dependency(vop).