A new likelihood approach to autonomous multiple model estimation. (April 2020)

• Record Type: Journal Article
• Authors: Soken, Halil Ersin; Sakai, Shin-ichiro
• Is Part Of: ISA transactions. Volume 99(2020)
• Page Start: 50
• Page End: 58
• Publication Date: 2020-04
• Subjects: Hybrid systems -- Parameter estimation -- Multiple model estimation -- Kalman filter; Engineering instruments -- Periodicals; Engineering instruments; Periodicals; Electronic journals; 629.805
• DOI: 10.1016/j.isatra.2019.09.005
• Languages: English
• ISSNs: 0019-0578

Abstract

Abstract: This paper presents an autonomous multiple model (AMM) estimation algorithm for hybrid systems with sudden changes in their parameters. Estimates of Kalman filters (KFs) that are tuned and employed for different system modes are merged based on a newly defined likelihood function without any necessity for filter interaction. The proposed likelihood function is composed of two measures, the filter agility measure and the steady-state error measure. These measures are derived based on filter adaptation rules. The numerical results show that the proposed algorithm, so called Competing AMM (CAMM), guarantees both steady-state estimation accuracy and quick parameter tracking. Highlights: A new likelihood function for autonomous multiple model estimation is derived. The designed estimator is for systems with parameter change. The likelihood function makes different models compete with each other. The new function ensures both estimation accuracy and quick tracking.