Comments (3)
Hello,
If you do not have odometry, then you need to set the odom_frame_id to be the frame as your base_link_frame_id.
Have a nice day :)
from dynamic_robot_localization.
Hey, thank you, that solved the first error :)
now I get these pcl errors. Is there some configuration I have to do?
[pcl::SampleConsensusModelRegistration::computeSampleDistanceThreshold] Covariance matrix has NaN values! Is the input cloud finite?
from dynamic_robot_localization.
Hello,
That error seems to be from sac_model_registration.h#L248, in which the covariance matrix is computed with centroid.hpp#L508.
That class is used in drl in cloud_matcher.hpp#L65 if the max_number_of_ransac_iterations > 0
, which is the default for recovery_3d.yaml#L12 (you can set it to 0 to disable the usage of ransac within ICP) and also in sample_consensus_prerejective.hpp#L542
drl removes NaN points from the point clouds when it receives sensor data (localization.hpp#L1953) and after normal estimation (localization.hpp#L2875).
As such, it is odd that the pcl function computeMeanAndCovarianceMatrix is generating a covariance matrix with NaN values.
Keep in mind that the dynamic_robot_localization_system_6dof.launch does not have the initial pose estimation module loaded by default and as such, it expects the end user to specify the initial pose either in that launch file dynamic_robot_localization_system_6dof.launch#L23 or given in a topic dynamic_robot_localization_system.launch#L4
Alternatively, you can enable the initial pose module by commenting the line with the empty.yaml (dynamic_robot_localization_system_6dof.launch#L44) and uncommenting the line below (dynamic_robot_localization_system_6dof.launch#L45)
The 3d reference point cloud can be provided in reference_pointcloud_filename or in reference_pointcloud_topic.
I tested the 6 DoF tracking with the ETHZ ASL dataset using their high complexity environments rosbags. One of the tests can be started with ethzasl_kinect_dataset_high_complexity_slow_fly_movement.launch which should have results similar to this video
If your sensor data has a scale that is very different (objects much smaller or larger than the ones present in the video above), you may need to adjust the search radius / k / correspondence distance values of the files: initial_pose_estimation_3d.yaml, cluttered_environments_dynamic_large_map_3d.yaml and recovery_3d.yaml
Have a nice day :)
from dynamic_robot_localization.
Related Issues (20)
- How do you get the correct TF? HOT 1
- the angle unit HOT 1
- 2-D global (3 DoF) localization questions HOT 21
- Using sparse and noisy PointCloud2 HOT 1
- build with ros-noetic HOT 2
- ROS2 Support HOT 2
- How to use 6 DoF robot localization HOT 11
- Selecting RoI HOT 1
- HSVto RGB HOT 1
- Lidar in pointcloud localization HOT 9
- Node doesn't start due to invalid time? HOT 1
- Mismatch between 6DoF tracking achieved locally and the reference video with same data set HOT 2
- Initial pose and odometry reuse HOT 6
- Turn on/off the tracking HOT 1
- compile errors HOT 9
- **[URGENT]** What is the result of the 3dof localization? HOT 11
- localization result unstable HOT 1
- 2D localization mode have roll pitch and Zaxis drit HOT 14
- error:registration.registerVisualizationCallback(this->update_visualizer_); HOT 5
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
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.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from dynamic_robot_localization.