ubuntuslave's profile - activity

I think I figured it out after looking at the arguments/options differences between catkin_tools and colcon. Here is a

Is there a ROS 2 bag GUI? I understand that the former ROS bags are now in the form of an sqlite database. For now, I ha

Is there any rqt_bag alternative for visualizing/playing ROS 2 bags?

How to use Eclipse IDE for developing Ros2 packages? In the past with ROS1, a simple command like catkin build MY_ROS1_P

@dinesh: Where did you get the mesh files (e.g., blade.dae and body.dae) for this helicopter model? I'm trying to test y

@dinesh: Where did you get the mesh files (e.g., blade.dae and body.dae) for this helicopter model? I'm trying to test y

@dinnesh: Where did you get the mesh files (e.g., blade.dae and body.dae) for this helicopter model? I'm trying to test

I think the misleading part could be the use of seconds in this case because I believe the time-stamp is given in the co

You need to have your .bashrc environment properly sourced. As stated in the documentation, you should load your .bashrc

I would like to note that this slow performance happens even when the service client is set to be persistent.

My message size is an array (i.e. float32[] of about 648*480*300 or 93312000 elements or 355 MB size). I noticed that writing/reading the array to/from disk takes less time than passing it via a ROS Service. Anyways to speed up the message passing? It takes 5 seconds now to receive the msg.

You should instead create a service node

Does anyone have experience with the following scenario:

A Python node that needs to pass a large amount of float values (an array of images) as a multidimensional numpy array to a CPP node.

The dilemma: What is better (in terms of memory or speed):

a) To pass the data as a serialized float32[] message array between these nodes?

b) To write the data to a binary file and just communicate the file's location as a string msg?

I know that using a Nodelet would be more memory efficient, but it's not supported for nodes written in Python. Do you believe to that writing the point cloud data to disk (as a binary file) and just passing a simple message flag for the receiver node to read such a file is any better (speed/mem)?

I also got a Chameleon Point Gray camera to work in OS X via libdc1394. The problem is that I can't figure out how to set the camera mode for it to give me color in OpenCV.

I've been having issues when running simultaneously both the Kinect's camera and a secondary (external) USB camera. I have successfully run several cameras together (different drivers and flavors). I am also able to run a kinect without RGB depth registration (so the RGB camera is off) plus other external cameras. However, I can't seem to be able to run both the Kinect RGB camera and an additional USB camera.

Has anyone been successful with this setup: Kinect with RGB registration + Secondary USB camera?

I'm trying to use a nodelet with an external library. But when running the nodelet, it obviously complaints that it can't find some symbol which is in this external library.

I can link the regular node the library without problems. However, when I use the nodelet version, it complains. I'm certain that I'm not linking the nodelet to the library. I don't know how to.

If you know how to link a nodelet to an external library, please share your know-how-to.

The proposed SOLUTION in CMakeLists.txt relevant piece looks like the following:

set(MVIMPACT_LIBRARIES   mvDeviceManager)

# Nodelet:  
rosbuild_add_library(mv_bluefox_driver_nodelet src/nodelets.cpp src/camera.cpp) 
# Linking the nodelet to the mvDeviceManager library
target_link_libraries(mv_bluefox_driver_nodelet ${MVIMPACT_LIBRARIES})  

# Regular Node:
rosbuild_add_executable(camera_node src/camera_node.cpp src/camera.cpp)
target_link_libraries(camera_node ${MVIMPACT_LIBRARIES})

I'm trying to get the calibration files (yaml) for a panoramic image. The image has been already rectified (unwrapped) using other the Taylor expansion method (OcamCalib Toolbox in MATLAB), but I'm not sure if in order to process it further I need to obtain the calibration parameters given by camera_calibration.

In other words, given a 360 degree Field-of-view lens, is the pinhole model used by camera_calibration applicable?

Obviously they are not the same. Which one should a developer be using?

I see that the camera_drivers stack has support for Willow Garage's wge100 Global Shutter Ethernet Cameras that come on the PR2's arms. Does any one know which cameras are these (hardware-wise)? or How can one buy them?

When creating a package whose header files will be included by others, we have to add the following to the manifest.xml as per to specify the package's include path.

<export>
   <cpp cflags="-I${prefix}/include" />
</export>

Should this be done automatically as it's done for libraries?

What is the difference between make and rosmake? Which one is recommended?

Why am I getting this after pressing "Ctrl+C"?

Warning: recursive print statement has occurred. Throwing out recursive print.

Warning: recursive print statement has occurred. Throwing out recursive print.

forever

I can't stop it!

Whenever I launch gmapping (even without the option output="screen"), I still get the annoying messages, such as:

Registering Scans:Done

Laser Pose= 0.886656 -0.155651 2.77905

Registering Scans:Done

Laser Pose= 0.886656 -0.155651 2.77905

This is a bit annoying since it fills up the screen with that info. Does anyone know how to suppress these messages?