search

vectr-ucla / direct_lidar_inertial_odometry

[IEEE ICRA'23] A new lightweight LiDAR-inertial odometry algorithm with a novel coarse-to-fine approach in constructing continuous-time trajectories for precise motion correction.
MIT License
521 stars 101 forks source link

[ROS2] Fix for params reading #29

Closed Bugday001 closed 7 months ago

Bugday001 commented 7 months ago

Hey! Firstly, thanks for great paper and code! I was trying to test DLIO (ROS2 branch) and found that there seem to be some issues. This may also be the cause of #26 . I am using Ubuntu 22.04 and ROS2 Humble.

Firstly, there seems to be some issues with the writing of the parameter file . yaml, which prevents the algorithm node from successfully reading the parameters in yaml. The location and mapping results are poor when using the default parameters in the . cpp file. I mainly modified the format of yaml, and the parameter names in the *. cpp file.

Another small issue is that when map/sense/filtered is set to false, odomnode will crash. The reason I found is `this ->deskewedThe height and width in the attributes of scan are both 0, resulting in subsequentpcl:: transformPointCloud ( published_cloud, deskewed_scant, Tcloud);Collapse. So I choose to create thedeskewedscanwith height and width, instead of usingresize`.

Finally, thanks again for this extremely good work!

kennyjchen commented 7 months ago

Awesome! Thanks for the contribution :^)