A new recursive formulation for fast estimation of time-varying unknown input. Application for estimating a particule acceleration
Abstract
This paper proposes an extension of the Recursive Input Estimation (RIE) for estimating time-varying unknown input. This extension is based on modifications of the Least Squares (LS) algorithm of the RIE, which increase its capability to track timevarying unknown input. These modifications consist in inserting a forgetting factor into the LS algorithm and adding an integral effect. Moreover, alternative formulations are proposed, allowing the reduction of computation time in the case where the number of inputs to estimate is lower than the number of measurements. The different tools obtained are tested and compared on a three-dimensional tracking application.