Improved design of proposal distribution for particle filter

One crucial issue in particle filtering is the selection of proposal distribution. Good proposal can effectively alleviate particle degeneracy and thus improve filtering accuracy. In this paper, we propose a new type of proposal distribution for particle filter, called as R-IEKF proposal. By combining iterated extended kalman filter with Rauch-Tung-Striebel optimal smoother, the new proposal integrates the latest observation into system and approximates the true posterior distribution reasonably well, hence generating more precise and stable particles against measurement noise. The simulation results indicate that the improved particle filter with R-IEKF proposal prevails over PF-EKF and UPF both in tracking accuracy and filtering stability. Consequently, PF-RIEKF is a competitive choice in noisy measurement environment.

Improved Particle Filter for Target Tracking

A new improved particle filter algorithm with the simplified UT （unscented transformation） and the modified unscented Kalman filter （UKF） proposal distribution is presented. The scaling factor is added to adaptively estimate on line and to improve the filtering performance. An adaptive algorithm is developed. In the bearings-only tracking experiments, the results confirm the improved particle filter algorithm outperforms others.

An improved particle filtering algorithm based on observation inversion optimal sampling

According to the effective sampling of particles and the particles impoverishment caused by re-sampling in particle filter,an improved particle filtering algorithm based on observation inversion optimal sampling was proposed. Firstly,virtual observations were generated from the latest observation,and two sampling strategies were presented. Then,the previous time particles were sampled by utilizing the function inversion relationship between observation and system state. Finally,the current time particles were generated on the basis of the previous time particles and the system one-step state transition model. By the above method,sampling particles can make full use of the latest observation information and the priori modeling information,so that they further approximate the true state. The theoretical analysis and experimental results show that the new algorithm filtering accuracy and real-time outperform obviously the standard particle filter,the extended Kalman particle filter and the unscented particle filter.

Particle Filter and Its Application in the Integrated Train Speed Measurement

Particle filter(PF) can solve the problem of state estimation under strong non-linear non-Gaussian noise condition with respect to traditional Kalman filter(KF) and those improved KFs such as extended KF(EKF) and unscented KF(UKF). However, problems such as particle depletion and particle degradation affect the performance of PF. Optimizing the particle set to high likelihood region with intelligent optimization algorithm results in a more reasonable distribution of the sampling particles and more accurate state estimation. In this paper, a novel bird swarm algorithm based PF(BSAPF) is presented. Firstly, different behavior models are established by emulating the predation, flight, vigilance and follower behavior of the birds. Then, the observation information is introduced into the optimization process of the proposal distribution with the design of fitness function. In order to prevent particles from getting premature(being stuck into local optimum) and increase the diversity of particles, Lévy flight is designed to increase the randomness of particle's movement. Finally,the proposed algorithm is applied to estimate the speed of the train under the condition that the measurement noise of the wheel sensor is non-Gaussian distribution. Simulation study and experimental results both show that BSAPF is more accurate and has more effective particle number as compared with PF and UKF, demonstrating the promising performance of the method.