Research And Application Of Particle Filter Resampling Algorithms

The Kalman Filter is widely applied in the Information Fusion at the present, which can get the optimal estimate in the Linear-Gaussian model, but not applied in the nonlinear and non-Gaussian model. In this case, Particle Filter is studied abroad for its wide application.The Particle Filter is a filter method based on Monte-Carlo Simulation and Recursive Beyesian Estimation. As other predictive filters, state space is recursively got from measure space with system model by using the Particle Filter. It uses particles to describe the state space. The discretely random measure composed by particles and associated weights approximates to the true posterior state distribution, and is updated by iteration of the algorithm. The Particle Filter can resolve a problem on nonlinear and non-Gaussian model, and has been applied successfully in many fields.But the Particle Filter has also some drawbacks, such as particle degeneration and suboptimal proposal distribution. Resampling is adopted to resolve the particle degeneration. Therefore, research and improving of resampling algorithms have the significance for increasing efficiency of the Particle Filter.The paper proposes a new resampling algorithm named as Divisional Resampling based on several resampling algorithms. Analysis is made to compare the Divisional Resampling with the Multinomial Resampling, the Residual Resampling and the Stratified Resampling. The results show that the average performance of the Divisional Resampling is superior to other Resampling algorithms's. In the end, the Divisional Resampling is applied to the demo system of trajectory planning and trajectory control, which indicates that it can increase the efficiency of the Particle Filter and be applied in practice.