Open cmakelabs opened 2 weeks ago
Hello @cmakelabs,
In the YAML files within the config folder, each row of extrinsic_T
corresponds to the translation parameters (x, y, z) for each LIDAR with respect to the IMU. Similarly, each row of extrinsic_R
corresponds to the rotation parameters (qw, qx, qy, qz) for each LIDAR with respect to the IMU.
For example, the given matrices translate as follows:
Ensure the order of these rows corresponds to your LIDAR configuration, which might look like this:
lid_topic: ["/os_cloud_node/points", "/livox/avia/points", "/livox/tele/points"]
lid_type: [3, 1, 1] # 1-Livox, 2-Velodyne, 3-Ouster
N_SCANS: [128, 8, 8]
point_filter_num: [8, 4, 4]
This way, each LIDAR's translation and rotation matrices are properly specified with respect to the IMU.
Thank you very much, does that mean that in the yaml
file for two lidars for example I will have:
extrinsic_T:
extrinsic_R:
# followed directly by
extrinsic_T:
extrinsic_R:
Would that be what you meant, that the parser will understand what parameter in the yaml file is for which LIDAR solely based on the order of the parameter in the file? Is it fine that these parameters are having the same name?
For
mapping/extrinsic_T
, andmapping/extrinsic_R
in theyaml
files available in theconfig
folder, there is only: one translation matrix and one rotation matrix, like this for example:but in the image I can see that it is possible to specify for each LIDAR the matrices
extrinsic_T
andextrinsic_R
with respect toIMU
, this means I would have for two LIDARs two translation matrices and two rotation matrices, Could you please tell how to specify that in theyaml
file?