adreno's profile - activity

HI,

1) As of Hydro turtlesim uses the geometry_msgs/Twist message instead of turtlesim/Velocity. 2) Change to cmd_vel instead of command_velocity

So in your code use geometry_msgs/Twist instead of turtlesim/Velocity, Also, publish it to cmd_vel and it should work then.

Thanks..it is helpful!

Hi,

Can someone please give an example of how to set the costmap parameters for the range_sensor_layer, recently added in Hydro ? (as of now, I am using IR sensors converted into a single cloud, but facing problems as the obstacles are not clearing up)

Thanks!

Hey, I am having the same problem...any solution for this ?

1) gscam need the camera_parameters.txt calibration file in the gscam package directory

2) For it to work well with what you want to do for image processing you should calibrate the camera that you are using

3) camera_calibration package will create the right calibration file for you, which you can place in the gscam package directory

Enjoy!

Check if you have camera_parameters.txt file in your gscam package directory...

@salma YES! (but need to do that for every occurrence of it in the package/node)

Hi, Here you go -

I installed the fuerte release of the driver from the git -

1) Place the extracted folder in ros workspace

2) cd ethzasl_xsens_driver-fuerte_released

3) rosmake xsens_driver

4) cd xsens_driver/nodes

5) ./mtdevice.py [commands] [opts] (read the Usage instructions on the ros wiki page of the driver for the arguments)

I hope this will help - Good Luck!

1) sudo chown :user_name /dev/ttyUSB0 2) sudo adduser user_name dialout

refer to the edited answer above!

Try aptitude instead of apt-get

Hi, You can always use the Kinect camera to perform the recognition using the face_recognition package. The package is based on image input and does not use the depth data. If you want to use depth data in your recognition system then you have to work with something else.

Here is how you can do that - The node is subscribing to topic - /camera/image_raw Kinect is publishing on topic - /camera/rgb/image_color

Create a launch file to remap the topic and you are all set. Ex - <remap from="/camera/image_raw" to="/camera/rgb/image_color"/>

Enjoy!

I have not used it but you may have to add the user to the dialout group so that the node can communicate with the device and also have to change to usb port permissions.

Step 1) cd procrob_functional-master/

Step 2) rosmake face_recognition

enjoy!

This should be it - http://wiki.ros.org/ROS/Tutorials/Cre... If its a single node then you don't need to write any publisher/subscriber. Just go through the ROS Tutorials.

You can use the ros::Rate to specify the frequency that you would like to loop at

Hi, Try this - http://wiki.ros.org/xsens_driver

C++ also has better support in terms of the tutorials posted on ROS wiki

Did you rosmake ? You have to build the package to generate the executables...

Thanks a lot! update did solved the problem :)

Vrep - www.coppeliarobotics.com Yes, V-Rep is better simulator but documentation and tutorials are not straightforward as gazebo. good luck!

Hey, This is what you need - Voxelized Shape and Color Histograms (VOSCH)

Hi, Try V-Rep

openni2 supports xtion devices, but not kinects. For using a kinect try the freenect stack: http://www.ros.org/wiki/freenect_stack

Hi,

I am following the moveit tutorial to load pr2.urdf.xacro in MoveIt Setup Assistant. As I click on Load Files I get the following symbol lookup error:

[librospack]: error while executing command /opt/ros/hydro/lib/moveit_setup_assistant/moveit_setup_assistant: symbol lookup error: /opt/ros/hydro/lib/libmoveit_setup_assistant_widgets.so: undefined symbol: _ZN3ros7console5printEPNS0_10FilterBaseEPvNS0_6levels5LevelEPKciS7_S7_z

Any suggestions on fixing this ?