ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange |
2022-05-12 15:31:07 -0500 | received badge | ● Good Question (source) |
2018-04-23 02:47:28 -0500 | received badge | ● Nice Question (source) |
2017-03-18 09:11:43 -0500 | received badge | ● Famous Question (source) |
2017-03-01 05:38:48 -0500 | received badge | ● Famous Question (source) |
2017-02-08 10:29:05 -0500 | received badge | ● Enthusiast |
2017-02-07 19:52:25 -0500 | received badge | ● Notable Question (source) |
2017-02-07 17:52:55 -0500 | answered a question | ROS not calling callbacks fast enough Based on Dirk's suggestion, I tried the noTcpDelay transport hint, and that seems to have fixed the issue. Thanks! |
2017-02-07 11:47:37 -0500 | received badge | ● Popular Question (source) |
2017-02-07 10:28:51 -0500 | commented question | ROS not calling callbacks fast enough However, they are communicating on separate topics, so why would the simulation interfere? My CPU usage is still around 50%. |
2017-02-07 10:28:13 -0500 | commented question | ROS not calling callbacks fast enough I made a simple talker/listener package to test this. When I just run the the talker and listener nodes with nothing else running, both run fine at 1kHz, and even at 10kHz. If I also have my simulation running in parallel, then I get msg drop. |
2017-02-06 19:01:51 -0500 | asked a question | ROS not calling callbacks fast enough EDIT: Here's a basic talker/listener example that reproduces the problem, if anyone else wants to try it. The talker is sending msgs of type Original Post: There is something wrong with how ROS is calling subscriber callbacks at a much slower rate than msgs are being published. Here's a minimal example. The topic "gazebo/link_states" is being published to at 500 Hz (using Resource monitor shows cpu usage at less than 50%. Even setting high priority for this process changes nothing, so I don't think its a bottleneck in the hardware. I'm on Ubuntu 16.04 LTS, with ROS Kinetic. Anyone have any idea at what I should look at next to help troubleshoot? Thanks for your help. |
2017-01-27 15:53:51 -0500 | received badge | ● Nice Question (source) |
2017-01-24 06:45:03 -0500 | received badge | ● Notable Question (source) |
2017-01-10 03:51:16 -0500 | received badge | ● Popular Question (source) |
2016-11-29 14:18:45 -0500 | asked a question | Subscriber callback missing messages Summary: I have a node publishing messages at ~300hz, but a callback subscribing to the topic in another node only gets called at ~25hz. The spinOnce in the subscriber node is being called at ~700hz, so I don't know why it's missing messages. Publisher node: Subscriber node: (more) |
2016-11-29 13:00:54 -0500 | received badge | ● Famous Question (source) |
2016-11-29 13:00:54 -0500 | received badge | ● Notable Question (source) |
2016-11-29 13:00:54 -0500 | received badge | ● Popular Question (source) |
2016-01-27 12:19:05 -0500 | asked a question | How to include opencv headers using short paths While trying to compile old code, I run into the error: If I change the include line in the code to then it compiles fine. Is there something that has changed in my environment? How do I get the first version to compile consistently? |
2016-01-15 09:30:06 -0500 | received badge | ● Student (source) |
2015-06-25 15:20:56 -0500 | received badge | ● Famous Question (source) |
2015-06-02 03:29:02 -0500 | received badge | ● Notable Question (source) |
2015-04-22 07:05:10 -0500 | received badge | ● Popular Question (source) |
2015-04-12 18:08:17 -0500 | asked a question | rospy tf waitForTransform not waiting I keep getting lookup extrapolation errors when trying to use tf, even when using listener.waitForTransform. For example I get: Exception: Lookup would require extrapolation into the future. Requested time 1428875486.097693682 but the latest data is at time 1428875486.097676516, when looking up transform from frame [ugv0] to frame [image] The difference in time is less than a millisecond, and even if my timeout duration is very long, I still get an extrapolation error. My code has: tfl.waitForTransform('image',obj_name,timeStamp,rospy.Duration(10.0)) The tf call is in a callback on a topic that has data published to it at a very fast rate, approximately 360hz, and the relevant transforms are also all being published at ~360hz except for a static transform, which is at 10hz. I would attach the pdf of the view frames output, but I dont have enough points to attach files. What is wrong? |
2013-08-24 02:50:34 -0500 | received badge | ● Famous Question (source) |
2013-06-20 02:50:16 -0500 | received badge | ● Notable Question (source) |
2013-06-10 06:05:43 -0500 | marked best answer | How to pass arguments to python node I would like pass arguments to a node like so: Is this possible? How do I define my python function? Currently it is: but when I use the command line and run rosrun I get errors saying the my function takes two arguments and 0 were passed. Using topics to pass in information seems over complicated and I don't want to use parameters because I don't need to make myArg1 and myArg2 available to any other nodes. Any help would be appreciated. Edit: Running fuerte |