Code Monkey home page Code Monkey logo

Comments (2)

carlosmccosta avatar carlosmccosta commented on July 30, 2024

Hello,

Thank you for reporting the problem :)

By mistake, I committed the usage of lookup tables for 6 DoF tracking with the Meijster algorithm for initialization of the lut.
I haven't had time to finish the implementation of the Meijster algorithm in my pcl fork, so if you want to use it, you can use kdtrees instead of the distance transform algorithm.
https://github.com/carlosmccosta/dynamic_robot_localization/blob/noetic-devel/yaml/schema/drl_configs.yaml#L531
With lookup tables the icp registration algorithm will be faster when performing point cloud registration (when compared with the default kdtree search approach, for finding correspondences).

For having similar results as the video I uploaded to youtube, I made this change:
aeae8be

For testing the 6 DoF tracking, please revert the modifications you made.
roslaunch dynamic_robot_localization_tests ethzasl_kinect_dataset_high_complexity_slow_fly_movement.launch

The FullIntegration is an operation mode for slam, in which the full sensor point cloud is integrated / accumulated.
drl can perform 6 DoF tracking in real time.
But for performing slam in real time you need a very powerful cpu or you need to play the rosbag at a much lower rate, which was what I did to create the reference point cloud using drl.
https://github.com/carlosmccosta/dynamic_robot_localization_tests/blob/noetic-devel/launch/environments/asl/ethzasl_kinect_dataset.launch#L10
https://github.com/carlosmccosta/dynamic_robot_localization_tests/blob/noetic-devel/maps/asl/ethzasl_kinect_dataset/high-complexity-environment_icp-point-point_0.0025-voxel-grid_mls_0.005-rosbag-speed.ply

During tests for other use cases (3D perception), I changed the loading of the reference point cloud to happen before the setup of the ros publishers.

That was why you could not see the reference point cloud in rviz.
With the commit below you can now see the reference point cloud in rviz after its loaded from the .ply file by drl.
a79c09f

For testing, please update all repositories to the latest commit of the most recent branch.
Namely, noetic-devel for:
https://github.com/carlosmccosta/dynamic_robot_localization
https://github.com/carlosmccosta/dynamic_robot_localization_tests
https://github.com/carlosmccosta/robot_localization_tools

And melodic-devel for:
https://github.com/carlosmccosta/pose_to_tf_publisher

Also, make sure that drl has loaded the reference point cloud before pressing the space key to play the rosbag (by looking at topic /dynamic_robot_localization/reference_pointcloud in rviz or command line).

Have a nice day,

from dynamic_robot_localization.

divya-m-s avatar divya-m-s commented on July 30, 2024

Hi Carlos,

Thank you very much for the quick fix and detailed feedback. The reported issue is solved now.

~ Divya

from dynamic_robot_localization.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.