Adaptively robust filter with multi adaptive factors
- 19 July 2008
- journal article
- research article
- Published by Informa UK Limited in Survey Review
- Vol. 40 (309), 260-270
- https://doi.org/10.1179/003962608x325330
Abstract
An adaptively robust filter with multi adaptive factors is proposed, based on the principles of adaptive Kalman filter and bifactor robust estimation for correlated observations. The estimator of the adaptive filter with multi adaptive factors is derived. The corresponding adaptive factors for the components of the position vector and velocity vector are set up based on the discrepancy of the predicted state from the kinematic model and the estimated state from the measurements. The adaptively robust filter with multi adaptive factors is more flexible in controlling the disturbing effects of the state components compared to the adaptively robust filters with unified adaptive factor and classified adaptive factors. The existing problems of the adaptive filter with multi adaptive factors are analyzed. An actual GPS dada set of an aircraft is calculated and analyzed with four schemes. It is shown by the actual calculation that the results of the adaptive filter with multi adaptive factors are slightly superior to those of the adaptive filter with unified adaptive factor. However, the new adaptively robust filter cannot be applied when the number of measurements at some epochs is smaller than the number of state parameters. It is proposed that the adaptively robust filter with multi adaptive factors and that with unified adaptive factor could be integrated in practical applications.Keywords
This publication has 10 references indexed in Scilit:
- Adaptively robust filtering with classified adaptive factorsProgress in Natural Science: Materials International, 2006
- An Optimal Adaptive Kalman FilterJournal of Geodesy, 2006
- Comparison of Adaptive Factors in Kalman Filters on Navigation ResultsJournal of Navigation, 2005
- Application of adaptive filtering by selecting the parameter weight factor in precise kinematic GPS positioningProgress in Natural Science: Materials International, 2005
- Robust estimator for correlated observations based on bifactor equivalent weightsJournal of Geodesy, 2002
- Adaptively robust filtering for kinematic geodetic positioningJournal of Geodesy, 2001
- Adaptive Kalman Filtering for INS/GPSJournal of Geodesy, 1999
- Robust Kalman filter for rank deficient observation modelsJournal of Geodesy, 1998
- Robust bayesian estimationJournal of Geodesy, 1991
- Robust StatisticsWiley Series in Probability and Statistics, 1981