New results in modelling derived from Bayesian filtering

This paper suggests an original heuristic modelling algorithm expressed in terms of homogenous combinations of the classical system dynamics and the Bayesian degree of truth employed in modelling. The main benefits of the proposed approach compared to classical modelling are the increased transparency and alleviated computational time. Two case studies, dealing with a mobile robot and an unforced pendulum system, are included to exemplify and test the theoretical results. One of the case studies makes use of the definition and calculation of several discrete plausibilities.