jbinney's profile - activity

@ruffsl i haven't tried 20.04 yet - I'll update this issue when i do (might be a while)

Ah! It was (at least for me) a problem with the default apparmor profile blocking that dbus call. Adding "--security-opt

Update to the update: i was wrong... that didn't fix it.

Update: i managed to workaround the problem by downloading the newest docker seccomp config from https://raw.githubuserc

@ruffsl - did you ever figure out why --privileged was needed in this case? We use --net=host and we see gazebo, rviz, a

One alternative to shallow clones might be to have bloom use the "--reference" option when cloning rosdistro. Then there

Seems to be a github issue. I just reproduced it by cloning using an xauth token - it ran very slow. Then tried again, a

Interestingly i can clone the the rosdistro repo in 30 seconds. Not sure why "Pulling latest rosdistro branch" takes so

Bloom takes ~15 minutes to pull down rosdistro I've noticed this when releasing packages for noetic. It looks like it is

When I use catkin_make inside of a tmux session, and then source the devel/setup.sh, I sometimes end up with the CMAKE_PREFIX_PATH and the ROS_PACKAGE_PATH being backwards. This causes my overlay to be an underlay, so catkin finds system ROS packages before my source packages. It doesn't always happen, and I don't have a reliable way to reproduce it. Its possible that it isn't actually related to tmux, but I've only seen the problem when inside a tmux session.

I'm posting here in the hope that someone has some insight into why this happens.

Here's the issue for this on github: https://github.com/ros/catkin/issues/455

The way I get around this is by adding the following to my .gitconfig:

# Always use ssh for github, even if the remote URL uses https or git                                                                                                 
[url "git@github.com:"]                                                                                        
  insteadOf = git://github.com/                                                                                
[url "git@github.com:"]                                                                                        
  insteadOf = https://github.com/

Then just make sure that you've added the your ssh key from the computer you are on to your github profile.

I had to re-open the question in order to add my comment as an answer, but you can close it again now.

Looks like you may be missing some dependencies. For example, one of the errors appears to fail to find something in the effort_controllers package. You can install dependencies using rosdep: rosdep install -i --from-path .

Great! I'll add that as an answer so to this question so it is easier for people to see. Out of curiosity, what tutorial are you following to use the segway in simulation?

Looks like you may be missing some dependencies. For example, one of the errors appears to fail to find something in the effort_controllers package. You can install dependencies using rosdep. Could you try running "rosdep install -i --from-path ." in your source directory?

In your setup.py you list 'src' as the package_dir but I don't see any 'src' directory in your package directories.

I've gotten this build error when compiling a version of moveit_ros against a different version of moveit_core. Can you run "git pull" in both the moveit_ros and moveit_core directories and then try building again?

Recompiling robot_state_publisher from source didn't work for me either, but recompiling both robot_state_publisher and the robot_model repo from source did work. I'm yet sure that this is just a bug within libpcre

There might be simpler first projects than self driving cars :-)

I'd recommend starting by reading the journal papers written by the DARPA Urban Challenge competitors. A good one to start with is the BOSS paper

Then just work backwards from there. A small matter of learning state of the art planning, controls, and perception.

The actionName won't be a path.. I think it will be just the root name of the action type; in this place "Control", or if that isn't unique then the package_name/action_name (control/Control). Then rosbridge will have to find that action. So make sure that the control package is built, and you've sourced the devel/setup.bash in the workspace where you built it before you run rosbridge. If you need help building your ros workspace, see this tutorial: http://wiki.ros.org/catkin/Tutorials/...

Could you describe what other nodes you have running? I'm not able to reproduce the memory leak by just running the code posted here - do you have subscribers on the chatter_pub topic?

In the past I've usually manually done the "reloading" of the urdf by restarting the entire robot. Automatically reloading the URDF while running would be neat, but also could be tricky. You would need to make sure that every ROS node is designed to watch for and use a modified robot description.

You're using a deprecated ROS version (electric) on a version of ubuntu that is no longer supported (11.10) with a camera package that isn't released anymore (openni_kinect) and a camera whose chipset maker got bought by apple and so will likely stop getting sold at some point (xtion).

If you switch to ROS indigo on ubuntu 14.04 and use openni2_camera to read from the camera, you'll have a much better chance of finding people who know how to answer your questions.

Out of curiosity, what kind of robot are you working on?

Just out of convenience. It's really nice to be able to connect to to the robot over wifi from my laptop and display the same webpage that shows up on the UI. Once we've got the interface all figured out, it may make sense to move to a native android app, or change the comms.

Morgan is right on - Robot Web Tools is a great way to go. Especially for early prototyping, it lets you use existing web frameworks to make a GUI very quickly. (I work at Savioke).

This seems like a good approach. Thanks for posting this!

Do the external projects use cmake? If they provide a cmake config file, you can find_package them just like opencv: http://wiki.ros.org/vision_opencv

ROS runs well on ubuntu linux; so you really want to find an SBC that can run ubuntu. If you want to use the newest ROS (Indigo) then you need an SBC that can run Ubuntu 14.04

There's a couple of things here. The queue_size warning should be fixed, but doesn't actually break anything yet. The real problem looks like xdot now returns floating point numbers instead of ints. I can confirm that I see the same problem on my computer. Can you create an issue here: https://github.com/ros-visualization/executive_smach_visualization/issues

Info? What sensor? What software? What OS? What are you launching?

sounds like the laser is actually using the "laser" coordinate frame.... what do you see in the frame_id field of the laser message when you echo it?

This is a transform problem. The laser measurements are published in the base_laser coordinate frame, but rviz is trying to display them in the "laser" or "map" coordinate frame. Set your fixed frame to "base_laser" and you should see the data displayed. If you want to use another frame you need to make sure TF transforms are published to connect base_laser to your other frames (for instance by using http://wiki.ros.org/robot_state_publi... and a urdf)

Did you add the laser topic in rviz? What frame is the laser scan in? What is your fixed_frame in rviz?

You should open an issue on the bug tracker for it: https://github.com/AutonomyLab/ardrone_autonomy/issues If you're comfortable with debugging tools, then use GDB to get a backtrace of the segfault and include that in the issue.

rosbag has a C++ API that you can use directly in your code: http://wiki.ros.org/rosbag/Code%20API

But I think everything you store will need to be ROS messages. So you'd need to define a ROS message type for your array, e.g. a file called "Foo.msg" which contains "float64[] some_array". Then you need to get catkin to generate code wrappers for that message.

How did you install OpenNi2? The ROS debs are built from a fork of that - there may be modifications. Can you uninstall your source copy of OpenNi2 and `sudo apt-get install libopenni2-dev libopenni2-0`?