A Novel Particle Filter Based on Propagation and Prediction
Bai Xiang-feng, LI Ai-hua, Jialei Li, Liu Taiyang · 2011
The problems existing in standard particle filter include large computation and particle degeneration, and a novel particle filter based on propagation and prediction is proposed to solve the problems. In this method, particles after state transition are propagated according to the distribution of state noise, and then the generated filial particles are used to predict corresponding mother particles referring to measurement, then the latest measurement information is fused into estimation. Therefore, the predicted particles are closer to the true state, and the accuracy of particle filter is improved. The efficiency of the algorithm has been proved by experimental results, and the algorithm occupies great predominance with fewer particles.