NAN observed after solving AX=B, leading to failure when optimizing extrinsics

Question

NAN observed after solving AX=B, leading to failure when optimizing extrinsics

Nicky0325 opened this issue a year ago · comments

Hello, I ran into a problem with my own dataset that matX contains NAN elements after solving linear equation AX=B using cv::solve() function in extrinsic_refine.hpp, causing failure in optimizing extrinsic between base LiDAR and ref LiDAR.

I've checked and made sure that my input point cloud does not contain NAN elements, what are possible solution to this problem to keep the optimization process running, thank you!

Xiyuan Liu · Answer 1 · Thu Feb 23 2023 20:04:30 GMT+0800 (China Standard Time)

Hi @Nicky0325 , have you tried using different voxel sizes and eigen_ratio?
Please let me know if this issue remains. I plan to update the project code but I am pretty busy this week.

Xiyuan Liu · Answer 2 · Wed Mar 01 2023 01:03:33 GMT+0800 (China Standard Time)

Hi @Nicky0325, I have updated the code. Please let me know if your problem is solved.

Nicky Hu · Answer 3 · Tue Apr 04 2023 14:16:35 GMT+0800 (China Standard Time)

Thank you, this is exactly the case. I found that the feature map was not correct due to faulty configuration of these two parameters. Thank you for the hints and the update, much appreciation!