Robust Kalman filter eigen/cppoptlib/c++ implementation based on the python implementation by Milos Stojanovic (github: milsto). Also uses CppNumericalSolver by Parick Wieschollek (github: PatWie). See milsto's nice explanation in his github for detail.

Highly experimental. This header file only implementation gives

template <int N, int M> class RKF

where N is number of states and M is number of outputs.

When compiling, Eigen and CppNumericalSolvers headers have to be seen by your compiler. For example, if Eigen is installed as /usr/include/eigen3/Eigen and CppNumericalSolvers is installed as /usr/local/include/cppoptlib, the optional compiler option -I/usr/include/eigen3 -I/usr/local/include/ will be needed.