Title :
Consistency checks for particle filters
Author :
van der Heijden, F.
Author_Institution :
Fac. of EEMCS, Twente Univ., Enschede, Netherlands
Abstract :
An "inconsistent" particle filter produces - in a statistical sense - larger estimation errors than predicted by the model on which the filter is based. Two test variables are introduced that allow the detection of inconsistent behavior. The statistical properties of the variables are analyzed. Experiments confirm their suitability for inconsistency detection.
Keywords :
Monte Carlo methods; particle filtering (numerical methods); state estimation; statistical analysis; Kalman state estimation; Monte Carlo approach; consistency checks; particle filters; statistical analysis; statistical sense-larger estimation errors; Estimation error; Fault detection; Filtering; Hidden Markov models; Mathematical model; Noise measurement; Particle filters; Predictive models; State estimation; Testing; Index Terms- Particle filtering; consistency checks; fault detection; model validation.; modeling errors; Algorithms; Artificial Intelligence; Computer Simulation; Models, Statistical; Pattern Recognition, Automated; Signal Processing, Computer-Assisted; Stochastic Processes; Systems Theory;
Journal_Title :
Pattern Analysis and Machine Intelligence, IEEE Transactions on
DOI :
10.1109/TPAMI.2006.5
