Two fusion predictors for continuous-time linear systems with different types of observations

作者：Song Haryong; Lee Kyung Min; Shin Vladimir^*

来源：International Journal of Systems Science, 2012, 43(1): 41-49.

DOI：10.1080/00207721003768159

• continuous-time linear system
• Kalman predictors
• multisensor
• fusion formula

This article proposes new fusion predictors for continuous-time linear systems with different types of observations. The fusion predictors are formed by the summation of the local Kalman estimators (filters and predictors) with matrix weights depending only on time instants. Both fusion predictors represent the optimal linear combination of an arbitrary number of local Kalman estimators and each is fused by the minimum mean square error criterion. As a consequence of the parallel structure of the proposed predictors, parallel computers can be used for their design. This article also establishes the relationship between fusion predictors. High accuracy and computational efficiency of the fusion predictors are demonstrated through several examples, including the damper harmonic oscillator motion with a multisensory environment.

• 出版日期2012