Variable resolution particle filter
IJCAI'03 Proceedings of the 18th international joint conference on Artificial intelligence
A tutorial on particle filters for online nonlinear/non-GaussianBayesian tracking
IEEE Transactions on Signal Processing
Hi-index | 0.00 |
A fuzzy adaptive particle filter for fault diagnosis of dead-reckoning sensors of wheeled mobile robots was presented. The key idea was to constrain sampling space to a fuzzy subset of discrete fault space according to domain knowledge. Domain knowledge was employed to describe 5 kinds of planar movement modes of wheeled mobile robots. The uncertainties of domain knowledge (due to imprecise driving system and inaccurate locomotion system) were represented with fuzzy sets. Five subjection functions were defined and aggregated to determine discrete transitional probability. Two typical advantages of this method are: (1) most particles will be drawn from the most hopeful area of the state space; (2) logical inference abilities can be integrated into particle filter by domain constraints. The method is testified in the problem of fault diagnosis for wheeled mobile robots.