Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Particles always converge to the same position #280

Open
txshang opened this issue Jan 17, 2020 · 9 comments
Open

Particles always converge to the same position #280

txshang opened this issue Jan 17, 2020 · 9 comments
Labels

Comments

@txshang
Copy link

txshang commented Jan 17, 2020

When I run the code, the particles will always converge to a same position,the map is a pointcloud map with height, so i want to know if some parameters should be set.

@at-wat
Copy link
Owner

at-wat commented Jan 17, 2020

Information to reproduce and/or describe your problem should be provided. No one can help you without information.

  • OS and ROS distribution version
  • mcl_3dl version (from source or apt-get)
    • if from source, how did you build (catkin_make/catkin/catkin_make_isolated and optional arguments if specified)
    • if from source, dependent library versions (PCL, Eigen) if changed from rosdep or apt-get provided version
  • Information to reproduce your problem, for example:
    • configurations, if it differs from default or demonstration
    • launch files, if it differs from default or demonstration
    • data
  • Information to show what exactly happens, for example:
    • screenshots
    • console log/error messages
    • data

@txshang
Copy link
Author

txshang commented Jan 18, 2020

I use ros kinetic on ubuntu16.04
I use catkin_make to build mcl_3dl from source,I use pcl 1.8.1
I don't change default configurations,i changed the launch like this
image
I use rslidar points odom and a imu
When I call /global_localization,the position of particles always converge to the same place like this
image
Can I give my bag and pcd map to you,so you can give me some help

@at-wat
Copy link
Owner

at-wat commented Jan 18, 2020

Did you confirmed that the pointcloud, odometry, IMU data are correctly fed to mcl_3dl node?

  <remap from="rslidar_points" to="rslidar_points" />

This has no effect.
If the pointcloud is published to /rslidar_points, it is not fed to mcl_3dl.

Please check that the topics of pointcloud, odometry, IMU are connected to mcl_3dl node. (for example by using rqt_graph)

@txshang
Copy link
Author

txshang commented Jan 18, 2020

I have changed some topics name,so the /rslidar_points is no problem,imu and odom are all correct,the problem is just happened on this map and bag,if a map with height is not support

@at-wat
Copy link
Owner

at-wat commented Jan 18, 2020

#280 (comment)

the map is a pointcloud map with height

#280 (comment)

if a map with height is not support

I couldn't understand what do you mean these part.
As this package is 3-D/6-DOF, the map pointcloud always must have height (z position) information.
Could you explain bit more detail?

@txshang
Copy link
Author

txshang commented Jan 18, 2020

I mean if some configurations should be set,so to achieve a 6DOF localization.Because when I run localization on a robot which always move on a fixed height,it works.but if robot move on terrain , it not works.

@txshang
Copy link
Author

txshang commented Jan 18, 2020

Can I send the bag and map to you, could you do me a favor to test it when you are free

@at-wat
Copy link
Owner

at-wat commented Jan 18, 2020

Sure, seeing the actual data would be easier.
Could you upload the bag and map to somewhere? or, do you want not to make them public?

@txshang
Copy link
Author

txshang commented Jan 18, 2020

I can upload the data to GitHub and give you a link

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

2 participants