제어로봇시스템학회:학술대회논문집
- 2003.10a
- /
- Pages.559-564
- /
- 2003
Evolution Strategies Based Particle Filters for Nonlinear State Estimation
- Uosaki, Katsuji (Department of Information and Physical Sciences, Graduate School of Information Science and Technology, Osaka University) ;
- Kimura, Yuuya (Department of Information and Physical Sciences, Graduate School of Information Science and Technology, Osaka University) ;
- Hatanaka, Toshiharu (Department of Information and Physical Sciences, Graduate School of Information Science and Technology, Osaka University)
- Published : 2003.10.22
Abstract
Recently, particle filters have attracted attentions for nonlinear state estimation. They evaluate a posterior probability distribution of the state variable based on observations in simulation using so-called importance sampling. However, degeneracy phenomena in the importance weights deteriorate the filter performance. A new filter, Evolution Strategies Based Particle Filter, is proposed to circumvent this difficulty and to improve the performance. Numerical simulation results illustrate the applicability of the proposed idea.
Keywords
- Nonlinear filtering;
- particle filters;
- evolution strategies;
- evolutionary computation;
- importance sampling;
- resampling;
- selection