Back to search

University of Minnesota

Tuning an output-only dual Kalman filter for displacement estimation from measured accelerations

Abstract

dc:description.abstract

Direct measurement of the vertical displacement of bridge superstructures is typically not feasible; therefore, methods to estimate displacement from acceleration measurements exist. Joint input-state estimation techniques based on Kalman-type filters have been used for estimating displacements, but the task of tuning the filter is not straightforward. Therefore, this work explored a tuning procedure for the output-only implementation of the dual Kalman filter for displacement estimation via noisy acceleration measurements. The validity of the tuning procedure was demonstrated using simulated systems and experimental tests, and good dynamic displacement estimates were achieved in all cases. The approach indicated that model noise (e.g., due to poor system identification results) did not affect the tuning process as much as measurement noise. The output-only formulation of the problem caused the input estimates to lose physical meaning; however, the input estimator increased state estimation performance under various loading types as compared to a single Kalman filter.

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Brown, Riley

Rights

Language dc:language.iso
en

Identifiers

dc:identifier.*
Dc Identifier Citation
Brown, Riley. (2019). Tuning an output-only dual Kalman filter for displacement estimation from measured accelerations. Retrieved from the University Digital Conservancy, https://hdl.handle.net/11299/206709.

Chain of custody

source
Harvested from
University of Minnesota
Base URL
conservancy.umn.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Brown, Riley. Tuning an output-only dual Kalman filter for displacement estimation from measured accelerations. 2019. http://hdl.handle.net/11299/206709