Simultaneous Identification of Time-Varying Parameters and Estimation of System States Using

Authors:	Chen Wen, Nanyang Technological University, Singapore Er Meng Joo, Nanyang Technological University, Singapore
Topic:	2.5 Robust Control
Session:	Robust Adaptive and Switched Control
Keywords:	Iterative learning observer, identification of time-varying parameters, adaptive observers, learning mechanism, simulation.

Abstract

This paper presents the design of an Iterative Learning Observer(ILO) for the purpose of estimating system states whilesimultaneously identifying time-varying parameters. The proposedILO uses a novel updating mechanism to identify time-varyingparameters instead of using integrators which are commonly used inclassical adaptive observers to identify constant parameters whileestimating system states. The main idea behind the design of theILO is the use of learning, i.e. previous information iscombined into the ILO for identifying online time-varyingparameters. Stability of estimation error dynamics and convergenceof parameter estimation error are established and proven. Anillustrative example exhibits the effectiveness of the ILO.