long trajectory with big z drift dataset, converge not good in z direction

[DaojunZhu]

hi, thanks for your great work. we adapt hba in our own dataset, when the data has multilayer, the converge reuslt is good in xy direction, but is not good in z direction, even has bad effects. do you have any suggestion about this problem ?

we also confused about the hba code in PGO, which the first layer constraint build from initial pose not the optimized pose as measurement.

expect your reply, thanks!!

[KILLERQUEEN-L]

我也很疑惑，第一层约束从初始姿势构建而不是local BA的姿势构建，这和第一层使用的hess看起来无法对应上 @samsdolphin

[sherlockwzj]

我也很疑惑，第一层约束从初始姿势构建而不是local BA的姿势构建，这和第一层使用的hess看起来无法对应上 @samsdolphin

源代码是不是还只用了第一层和最后一层约束，中间层约束也没用？请问你是否尝试过修改代码以和论文对应？效果如何？