ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange |
2014-01-28 17:21:47 -0500 | marked best answer | Behavior of JointTrajectoryActionController with a single goal If I send the JointTrajectoryActionController a single goal using the SimpleActionClient with a velocity and acceleration of 0 each, will the commanded motor torques simply be like a PID controller with a fixed set point? Will the r_arm_controller (which I am assuming is a JointTrajectoryActionController) start behaving like a JointPositionController in this case? Thanks. |
2014-01-28 17:21:43 -0500 | marked best answer | Planning with OMPL using a subset of joints I would like to be able to treat the PR2 as a planar three link manipulator and test out some built in planners (openrave or OMPL). Is there a way to restrict openrave and/or OMPL to use only the shoulder pan, elbow flex, and wrist flex joint of the PR2? |
2014-01-28 17:21:42 -0500 | marked best answer | How to contributing python versions of tutorials? e.g. Tutorials for arm_navigation I sometimes find myself converting the C++ versions of the tutorials into python. e.g. http://code.google.com/p/gt-ros-pkg/s... What, if any, would be a good way to add links to python sample code to the tutorials? |
2013-10-21 15:13:51 -0500 | answered a question | [turtlemimic.launch] is neither a launch file in package [test_launch] nor is [test_launch] a launch file name? Is there a space between the package name in your package.xml? I was having the same error in a different package and removing spaces inside <name></name> solved the problem for me. Please see this issue for a possible cause of your problem. |
2013-04-15 07:37:23 -0500 | received badge | ● Famous Question (source) |
2013-02-07 10:11:18 -0500 | commented question | rospy publisher hanging when subscriber drops connection because of wifi - is this normal? |
2013-02-07 10:11:13 -0500 | commented question | rospy publisher hanging when subscriber drops connection because of wifi - is this normal? I have opened a ticket for this: https://github.com/ros/ros_comm/issues/169 |
2013-01-07 12:16:19 -0500 | received badge | ● Notable Question (source) |
2013-01-06 12:09:50 -0500 | commented answer | avoiding having to call catkin_make for changes to python modules? Thanks Tully and Thibault. additional details: after adding a setup.py and adding catkin_python_setup() to CMakeLists, once we run catkin_make, it creates an __init__.py in <catkin_workspace>/devel/lib/python2.7/dist-package/<your_package> This __init__.py redirects back to the source space. |
2013-01-06 12:04:34 -0500 | received badge | ● Popular Question (source) |
2013-01-04 15:49:36 -0500 | asked a question | avoiding having to call catkin_make for changes to python modules? While coding in Python (and using rosbuild) I find it very convenient to not to have to 'make' before testing my code: e.g. I change a module (which lives in its own ROS package) that my Python script uses and without any 'make' I kill the script, run it again and have it use the updated module. With catkin, am I correct in understanding that I will need to do a catkin_make each time so that the updated module gets copied into the devel space and is thus available to my script? |
2012-10-15 01:58:22 -0500 | received badge | ● Famous Question (source) |
2012-10-15 01:58:22 -0500 | received badge | ● Popular Question (source) |
2012-10-15 01:58:22 -0500 | received badge | ● Notable Question (source) |
2012-10-09 11:39:32 -0500 | commented answer | openni_launch not working in fuerte + ubuntu precise 12.04 Removing these two kernel modules worked for me. Thanks sleonard! |
2012-09-07 10:49:08 -0500 | received badge | ● Famous Question (source) |
2012-09-07 10:49:08 -0500 | received badge | ● Notable Question (source) |
2012-08-24 21:42:38 -0500 | received badge | ● Popular Question (source) |
2012-08-24 21:42:38 -0500 | received badge | ● Notable Question (source) |
2012-08-24 21:42:38 -0500 | received badge | ● Famous Question (source) |
2012-08-15 11:34:48 -0500 | received badge | ● Popular Question (source) |
2012-08-15 11:34:48 -0500 | received badge | ● Notable Question (source) |
2012-08-15 11:34:48 -0500 | received badge | ● Famous Question (source) |
2012-06-13 19:59:59 -0500 | received badge | ● Popular Question (source) |
2011-06-04 11:51:56 -0500 | marked best answer | laser_driver stack missing in pr2-desktop install See REP 108 for which stacks are included in which variants. Note that, as specified there, laser_drivers is not included in pr2-desktop or any of the variants that it extends. |
2011-06-03 04:20:53 -0500 | asked a question | laser_driver stack missing in pr2-desktop install It looks like the laser_drivers stack is missing from the pr2-desktop dimaondback install. Is this intentional? Here is what I did -
When I tried to roscd to hokuyo_node, I got an error.
Installing the laser_drivers stack fixed the problem:
|
2011-03-28 00:47:16 -0500 | received badge | ● Student (source) |
2011-03-15 08:23:26 -0500 | marked best answer | Behavior of JointTrajectoryActionController with a single goal The desired position trajectory will be a fifth-order polynomial, with start and end boundary conditions of 0 velocities and accelerations. Also called a "minimum jerk" movement in some communities. If you omit accelerations, you'll get a cubic spline. If you also omit velocities, you get linear interpolation. This is some documentation I found about this: http://www.ros.org/wiki/robot_mechanism_controllers/JointSplineTrajectoryController |
2011-03-15 08:23:25 -0500 | commented answer | Behavior of JointTrajectoryActionController with a single goal thanks a lot Mrinal, this was really useful! |