Levenberg-Marquardt and Line-Search Extended Kalman Smoothers

Paper in proceedings, 2020

The aim of this article is to present Levenberg–Marquardt and line-search extensions of the classical iterated extended Kalman smoother (IEKS) which has previously been shown to be equivalent to the Gauss–Newton method. The algo- rithms are derived by rewriting the algorithm’s steps in forms that can be efficiently implemented using modified EKS iter- ations. The resulting algorithms are experimentally shown to have superior convergence properties over the classical IEKS.