rock-ass's profile - activity

Hi,

I was trying to start hector quadrotor demo launch files and keep getting gazebo segmentation faults when the script tries to spawn quadrotor. After some time I have tried to spawn it on my own using these launch files:

roslaunch gazebo_ros empty_world.launch
roslaunch hector_quadrotor_gazebo spawn_quadrotor.launch

But even then the gazebo server crashes with following error:

spawn_model script started
[INFO] [WallTime: 1423382867.059244] [0.000000] Loading model xml from ros parameter
[INFO] [WallTime: 1423382867.061820] [0.000000] Waiting for service /gazebo/spawn_urdf_model
[INFO] [WallTime: 1423382867.064229] [0.000000] Calling service /gazebo/spawn_urdf_model
[INFO] [WallTime: 1423382867.111548] [0.000000] Controller Spawner: Waiting for service controller_manager/load_controller
Segmentation fault (core dumped)
Service call failed: transport error completing service call: unable to receive data from sender, check sender's logs for details
[gazebo-1] process has died [pid 24654, exit code 139, cmd /opt/ros/indigo/lib/gazebo_ros/gzserver worlds/empty.world __name:=gazebo __log:=/home/rokas/.ros/log/25359450-af08-11e4-930f-fcf8ae311424/gazebo-1.log].

The log files after the error is empty, so I can't traceback what happening. But it seems like spawning quadrotor crashes the gzserver.

I have tried using the packages installed via apt-get and then tried with latest version of hector_quadrotor from github and the problem persists. Maybe some of you will have some ideas on this problem?

The 2nd statement also applied to me when I was testing rgbdslam and then I stopped. I think rgbdslam is optimised for kinect sensor only, which gives a very detailed depth image. Stereo only is not so detailed and I think that's why it failes to reconstruct environment

viso2_ros is not a replacement for rgbdslam, sorry I lead to this confusion. Rgbdslam is fully integrated SLAM solution, while viso2_ros only estimates odometry - it integrates camera motion, then it applies motion to the robot frame and calculates how robot has moved only from stereo images.

As far as I understand positive Z would be going down faster than gravity pulls you. But this time +10Z mean hover

Finally figured out, pose estimation package requires gps sensor_msgs/NavSatFix data. Now I can get position (x, y) estimated. Also I noticed some strange behavior - algorithm estimated that quadrotor is falling (estimated altitude was negative), because IMU sensor Z axis was -9,8. I had to invert IMU's Z axis value to get the algorithm stabilize quadcopter's altitude.

Hope this helps if someone encounter similar problem.

It is possible to use stereo only with rgbdslam, I have tried it and it worked on gazebo stereo camera plugin, although results was bad, algorithm was hardly able to find matches between images

EDIT: As Ken_in_JAPAN asked in comments I'll try to tell you how I did it. I take into account that you have already compiled and launch rgbdslam package. I used gazebo_ros_multicamera plugin to generate synchronised image streams. Then I used stereo_image_proc package to generate point clouds and redirected them to the rgbdslam package.

I digged into my workspace and found launch file I used, here is a part of it:

<param name="config/topic_image_mono"              value=""/> <!--could also be color -->
<param name="config/topic_image_depth"             value=""/>
<param name="config/topic_points"                  value=""/> <!--if empty, poincloud will be reconstructed from image and depth -->
<param name="config/camera_info_topic"             value="/front_cam/left/camera_info"/>
<param name="config/wide_topic"                    value="/front_cam/left/image_rect_color"/>
<param name="config/wide_cloud_topic"              value="/front_cam/points2"/>

I left first three params empty and used only left image for colors and cloud topic from stereo_image_proc. I think that was all required to at least start it up.

After fondling with algorithm parameters, I did not achieve any good results with rgbdslam and later used viso2_ros stereo odometry node which was able to match point clouds a lot better. At least this was the case for me.

I hope that helps at least a bit, you can ask me if something is unclear.

Well you are looking in the right file. But that what sounds like a "folder" are ros topics. You just need a better background about ROS core features likes topics, those are used to exchange data between processes (they are called nodes). You read about them in http://wiki.ros.org/Topics

Look at rgbdslam\launch folder, there will be a few launch files, one of them will contain all possible properties which you can use as a template for sturtup, just fill in topics with point clouds and image streams from your camera and you should be good to go.

IMU does include all data from acc and gyroscopes. Double checked data standard in sensor_data, also compared with data provided from gazebo simulator.

Hello,

I am trying to use hector_uav_msgs/Altimeter for barometer sensor data publishing, but I cant figure out what does qnh attribute stand for? Is it possible it's pressure difference?

I understand this is a RTFM-like question, but I can't find any documentation regarding this stack. Any links to documentation would be greatly appreciated.

Thanks in advance

Hello,

I am trying to use hector_pose_estimation package for a real quadcopter. I have published information from IMU, barometer using sensor_msgs/Imu, hector_uav_msgs/Altimeter and pressure height messages using geometry_msgs/PointStamped message (just like hector_quadrotor package does in simulator).

When I visualise pose estimate using rviz I see changes in orientation and height, but it doesn't move sideways, it's stuck in center and only height and orientation changes.

Am I missing something out or hector_pose_estimation package doesn't calculate movement on XY axes? Using this package with simulator works fine, but with my real data from sensors - pose stuck in center.

Hi. That's weird, i have tried that code and i was able to get view, but it was not accurate. Have you tried to view kinect image stream using image_view package?

could you please copy whole rosmake error trace, there should be more information. It was some time ago, but as far as I remember g2o is an external dependancy, so maybe it;s missing. I'll take a look at it later this evening

Insightful answer, didn't thought of timestamps in first place

could you show frames graph ? you can generate pdf file of it with `rosrun tf view_frames`. Also you could try increasing publishing rate from 1000 to 100

I posted an answer with link to repo, mark it as answered if it helps :) Let me know if you get any results. i am also working on a quadrotor simulation, only i'm using stereo camera only instead of kinect

I have pushed the code with resolved issues for hydro distribution to a git repo, here is the address:

https://bitbucket.org/supelis/rgbdslam_freiburg

Try clone it, then try rosdep install to get dependancies (not sure if everything is available for ARM architecture) and then call rosmake from root directory of the package to compile. Let me know if there will be any issues.

There are example launch files inside the package, see them how to start it working.

Yes i'm able to use them on hydro. I can provide updated source code of rgbdslam package to be able to compile on ROS hydro, i will provide the info today. But you want it on ARM? simulator on arm should be a real pain to setup and probably will be of poor performance

I have tried to do the same thing, i got it working (although there is some issues). It required to resolve all dependancy issues and to migrate some old packages into Hydro distro, if you're still interested I could give a few pointers..

Hello, I'm working on a project in which I am trying to use my laptop as roscore and using beaglebone black as ros client. I have run into some issue, I cannot make a ros topic which would be readable on both sides (beaglebone and laptop). I tried running the most simple beginner_tutorial talker.py:

1. I set ROS_MASTER_URI on both machines to my laptops ip address and start roscore.

2. On beaglebone i run rosrun beginner_tutorials talker.py and everything starts without any errors. I find new topic listed under rostopic list and everything looks fine.

3. I try to subscribe to messages using rostopic echo /chatter on laptop and nothing appears.
4. Running rostopic echo /chatter on beaglebone I can see "Hello world" messages.

Both machines are running Ubuntu 12.04 with ROS Groovy installations. Beaglebone has ros-groovy-ros-bare-bones package and laptop has ros-groovy-ros-desktop-full package. It seems communication is fine, but messages is not interpreted on my laptop, maybe someone have some information why can`t I receive messages?