numerical stability about ice-ba

[liuhaomin]

Do you mean the precision of S and d, or the precision of solving Sx = d?

in general

local BA, marginalize will generate Sx = d which S is dense matrix.

your solution

give sub-track to construct sparse Sx = d, the method reduce constraints.
so i want to know whether the tips make optimization(LM or GN) non convergence sometimes?

[liuhaomin]

Indeed that is a problem for those points with too small parallax. Marginalizing such points will results in numerical instability and sub-track will make it worse. We do not marginalize such sub-tracks. If the information garthered on a sub-track is too small (less than the threshold epsdST), it will be treated as a fixed point and not be marginalized for Schur complement construction.

ICE-BA/Backend/BundleAdjustment/LocalBundleAdjustorGN.cpp

Line 927 in 7fe5fcf

if (KF.m_AxsST[iST].m_Sadd.m_a > epsdST) {

And for back substitution performed on the original tracks, we also check the information garthered on the original track using another threshold espd, and do not update the ill-conditioned points.

ICE-BA/Backend/BundleAdjustment/LocalBundleAdjustorGN.cpp

Line 914 in 7fe5fcf

if (KF.m_Axs[ix].m_Sadd.m_a > epsd) {

Got it, thank you

[wangjio]

The question has been solved.