We are happy to announce that our conference paper "A Factor Graph Approach to Parameter Identification for Affine LPV Systems" has been accepted for presentation and publication in the proceedings of the American Control Conference 2017. The conference will be held from May 24-26, 2017 in Seattle, WA, USA.
Read the abstract below:
Factor graphs are a versatile graphical representation of factorizable functions. As a probabilistic graphical
model, they allow to visualize structured conditional independence, which can be exploited for solving inference problems by means of message passing along the nodes of the graph. In this article we present a novel factor graph formulation of the expectation maximization (EM)-based estimation technique for affine linear parameter-varying system identification. By extending the factor graph representation of the Kalman
Filter/Smoother and the EM algorithm to parameter-varying matrices, a flexible tool for nonlinear system identification in the so-called linear parameter-varying (LPV) representation is obtained. Furthermore, a recursive reformulation of the algorithm suitable for tracking time-varying changes both accounted and
unaccounted for by a pre-defined LPV system description is immediate from its factor graph-based formulation.
