Some questions about loop closure detection

[SakuraMemoryKnight]

extern const float surroundingKeyframeSearchRadius = 50.0; // key frame that is within n meters from current pose will be considerd for scan-to-map optimization (when loop closure disabled) extern const int surroundingKeyframeSearchNum = 50; // submap size (when loop closure enabled) // history key frames (history submap for loop closure) extern const float historyKeyframeSearchRadius = 7.0; // key frame that is within n meters from current pose will be considerd for loop closure extern const int historyKeyframeSearchNum = 25; // 2n+1 number of hostory key frames will be fused into a submap for loop closure

If I want to make the loop closure detection as good as possible, which of the above parameters should be increased? and which should be reduced? Thank you!

[YushengWHU]

for my case, I just reduced the historyKeyframeSearchRadius from 7.0 to 5.0. Sensor tested: Hesai 40 channel Lidar Speed: around 50km/h

[gisbi-kim]

BTW fyi, I integrated the global loop detector (scan context) into the LeGO-LOAM pipeline https://github.com/irapkaist/SC-LeGO-LOAM This global loop detector works even large drift exist (e.g., https://www.youtube.com/watch?v=p-NsVs8GATA&feature=youtu.be&t=436) and could enhance LeGO-LOAM's original radius search-based loop detection.

[SakuraMemoryKnight]

@YushengWHU @kissb2 Thank you for your reply！

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.