Optimized Particle Filter with Genetic Algorithm

This paper presents a new approach to optimize the particle filtering with genetic algorithm. In the sate-space approach to model a non-linear dynamic system, the particle filtering has been successfully proven. In the standard particle filter, a re-sampling scheme is used to reduce the degeneracy problem and improve the estimation performance. But this resampling introduces some problem which is called sample impoverishment or particle deprivation. Our proposed Genetic Particle Filter (GPF) can overcome this problem in standard particle filter by using some genetic operators like crossover, mutation in prediction steps and roulette-wheel in sample selection step based on some error threshold. Our experiments results show that this strategic genetic algorithm approach improve particle filter performance and accuracy. M. Z. Islam, C. Oh, C. Lee, Optimized Particle Filter with Genetic Algorithm, TENCON2010, pp.1410-1414, Nov. 2010