Package Abstract :: This module provides methods to define autonomous linear state space models (ALSSMs) and to define squared error cost functions based on such ALSSMs. ALSSMs are input-free linear state space models (LSSMs), i.e., the model outputs are fully defined by a single state vector, often denoted as initial state vector x. The output vector of such a deterministic model forms the signal model used. Cost functions based on ALSSM are internally efficiently computed using recursive computation rules.

This module implements the methods published in [Wildhaber2018] PDF, with extensions from [Zalmai2017] and [Wildhaber2019].



This module provides methods to define linear state space models and methods to use them as signal models in recursive least squares problems.


Definition of recursively computed squared error cost functions (such as Cost Segments and Composite Costs), all based on ALSSMs


Selection tool to switch between Python interpreter (default) and JIT (Just-in-Time) compilation execution for time-critical routines in package lmlib.statespace.


Provides useful and specific applications based on lmlib.statespace methods

Introductory Example#