search

hku-mars / FAST-LIVO

A Fast and Tightly-coupled Sparse-Direct LiDAR-Inertial-Visual Odometry (LIVO).
GNU General Public License v2.0
1.26k stars 202 forks source link

question about ImuProcess::UndistortPcl() #43

Closed zzwu29 closed 1 year ago

zzwu29 commented 1 year ago

Dear Authors: Thank you for contributing elegant programs to open-source SLAM, which helped me a lot in my learning. During this process, I have a question about ImuProcess::UndistortPcl() https://github.com/hku-mars/FAST-LIVO/blob/a8918c23dc4dba7d98fa723a9fa0fc2cbdf4d977/src/IMU_Processing.cpp#L808-L811 From my point of view, the fast-livo code corresponds to the following formula image Does it use the extrinsic rotation of Lidar and IMU? Please help check whether the following formula is correct image R is the direction cosine matrix, p is the translation, L is lidar frame, G is global frame, and b is IMU frame. Wish you all the best!

xuankuzcr commented 1 year ago

In the T_ei, e is in the IMU frame, and i is in the LiDAR frame. Therefore, the extrinsic LiDAR frame w.r.t. IMU frame is used only once.

zzwu29 commented 1 year ago

Please forgive my trouble, the translation of lidar w.r.t. IMU is Lid offset to IMU, while the rotation is expressed as Lid rot to IMU, right? When will the extrinsic rotation parameter of lidar w.r.t. IMU be used?