The non-recursive estimation using a batch filter based on particle filtering

The distributed SLAM system has a similar estimation performance and requires only one-fifth of the computation time compared with centralized particle filter. However, particle impoverishment is inevitably because of the random particles prediction and resampling applied in generic particle filter, especially in SLAM problem that involves a large number of dimensions. In this paper, particle filter use in distributed SLAM was improved in two aspects. First, we improved the important function of the local filters in particle filter. The adaptive values were used to replace a set of constants in the computational process of importance function, which improved the robustness of the particle filter. Second, an information fusion method was proposed by mixing the innovation method and the number of effective particles method, which combined the advantages of these two methods. And this paper extends the previously known convergence results for particle filter to prove that improved particle filter converges to the optimal filter in mean square as the number of particles goes to infinity. The experiment results show that the proposed algorithm improved the virtue of the DPF-SLAM system in isolate faults and enabled the system to have a better tolerance and robustness.

We present a new unscented particle filter for dynamic systems that outperforms the general particle filter and the unscented particle filter when the variance of the observation noise is small. Our algorithm uses a bank of unscented Kalman filters to refine the prediction in particle filter. The key difference with the traditional unscented particle filter is the introduction of an auxiliary model and a bank of unscented Kalman filter with this auxiliary model to generate the importance distribution in the particle filter. This structure makes efficient use of the latest observation information. Our new algorithm use fewer particles than the general particle filters and its performance outperforms them.

This paper presents a new approach to estimate the true position of an unmanned aerial vehicle (UAV) in the conditions of spoofing attacks on global positioning system (GPS) receivers. This approach consists of two phases, the spoofing detection phase which is accomplished by hypothesis test and the trajectory estimation phase which is carried out by applying the adapted particle filters to the integrated inertial navigation system (INS) and GPS. Due to nonlinearity and unfavorable impacts of spoofing signals on GPS receivers, deviation in position calculation is modeled as a cumulative uniform error. This paper also presents a procedure of applying adapted particle swarm optimization filter (PSOF) to the INS/GPS integration system as an estimator to compensate spoofing attacks. Due to memory based nature of PSOF and benefits of each particle’s experiences, application of PSOF algorithm in the INS/GPS integration system leads to more precise positioning compared with general particle filter (PF) and adaptive unscented particle filer (AUPF) in the GPS spoofing attack scenarios. Simulation results show that the adapted PSOF algorithm is more reliable and accurate in estimating the true position of UAV in the condition of spoofing attacks. The validation of the proposed method is done by root mean square error (RMSE) test.

Particle filter (PF) serves as an effective method applied to the fault diagnosis of nonlinear and non-Gaussian systems. However, the result of state estimation is influenced by the particle impoverishment problem which is common in the typical PF algorithm. Based on the analysis of the PF algorithm, the general particle impoverishment problem is attributed to the deficiency of particle diversity. In this paper a novel intelligent particle filter (NIPF) is designed to deal with the problem of particle impoverishment by means of the genetic and adaptive strategy. The general PF can be regarded as a particular instance of NIPF. The experiment on the vertically falling body model shows that the NIPF can increase the particles diversity and improve the results of state estimation.

Non-linear filter has been widely used in many applications, and particle filter (PF) is a promising approach because it can treat the non-linear and non-Gaussian system. If there are correlations between process noise and measurement noise in the non-linear system, the performance of general PF degenerates. The study addresses that there are correlated measurement and process noise at the same time in the non-linear system. Aiming at this problem, the authors decouple the correlation and rearrange the state transition equation to a new one, and remove the correlation. And then apply the general PF based on the new state transition equation and the original measurement equation. From the derivation, it is easy to know that the proposed method is equal to general PF when there are no correlated noises. Experimental results show that the proposed method outperforms the general PF when there are correlated noises.

The square root cubature particle filter (SRCPF) uses the square root cubature Kalman filter (SRCKF) for generating the proposal distribution. The SRCPF algorithm is easy to be implemented and has numerical stability. Moreover, the SRCKF based proposal distribution approximates the optimal importance distribution by incorporating the current measurement. Simulation results demonstrate that the SRCPF algorithm has the better performance for state estimation than the generic particle filter (GPF), extended particle filter (EPF) and unscented particle filter (UPF), and its calculation cost decreases largely.

Particle filtering (PF) schemes are a set of simulation-based techniques relying on proposal distributions which have a crucial impact on their performance. In this paper, we introduce a novel constrained Extended Kalman particle filter (CEPF), the constrained Extended Kalman filter is used to generate the porposal distribution. The algorithm integrates the nonlinear state constraint information and the latest measurement information simultaneously into the dynamic system transition density. The proposed algorithm selects particles with higher likelihood to propagate into the next time ste by an efficient series of constraint optimization. The method can convergent theoretically and the simulation results show estimate advantage compared with other convention filters such as Extended Kalman Filter (EKF), Unscented Kalman Filer (UKF), Generic Particle Filter(GPF), particle filter with EKF proposal (EPF) and particle filter with UKF proposal (UPF).

In a non-Gaussian environment, the accuracy of a Kalman filter might be reduced. In this paper, a two- dimensional Monte Carlo Filter is proposed to overcome the challenge of the non-Gaussian environment for filtering. The two-dimensional Monte Carlo (TMC) method is first proposed to improve the efficacy of the sampling. Then, the TMC filter (TMCF) algorithm is proposed to solve the non-Gaussian filter problem based on the TMC. In the TMCF, particles are deployed in the confidence interval uniformly in terms of the sampling interval, and their weights are calculated based on Bayesian inference. Then, the posterior distribution is described more accurately with less particles and their weights. Different from the PF, the TMCF completes the transfer of the distribution using a series of calculations of weights and uses particles to occupy the state space in the confidence interval. Numerical simulations demonstrated that, the accuracy of the TMCF approximates the Kalman filter (KF) (the error is about 10−6) in a two-dimensional linear/ Gaussian environment. In a two-dimensional linear/non-Gaussian system, the accuracy of the TMCF is improved by 0.01, and the computation time reduced to 0.067 s from 0.20 s, compared with the particle filter.

We demonstrate experimentally that inference in a complex hybrid dynamic Bayesian network (DBN) is possible using the 2-time slice DBN (2T-DBN) from (D. Koller et al., Sequential Monte Carlo Methods in Practice: p.445-464, Springer-Verlag, NY, 2000) to model fault detection in a watertank system. In this, a generic particle filter (PF) is used for inference. We extend the experiment and perform approximate inference using The extended Kalman filter (EKF) and the unscented Kalman filter (UKF). Furthermore, we combine these techniques in a 'non-strict' Rao-Blackwellisation framework and apply it to the watertank system. We show that UKF and UKF in a PF framework outperform the generic PF, EKF and EKF in a PF framework with respect to accuracy and robustness in terms of estimation RMSE (root-mean-square error). Especially, we demonstrate the superiority of UKF in a PF framework when our beliefs of how data was generated are wrong. We also show that the choice of network structure is very important for the performance of the generic PF and the EKF algorithms, but not for the UKF algorithms. Furthermore, we investigate the influence of data noise in the watertank simulation. Theory and implementation is based on the theory presented in (R. v.d. Merwe et al., Tech. Rep. CUED/F-INFENG/TR-380, Dept. of Eng., Cambridge Uni., 2000).

Application of an interacting particle filter to improve nitrogen nutrition index predictions for winter wheat

Visual tracking is one of the key tasks in computer vision. The particle filter algorithm has been extensively used to tackle this problem due to its flexibility. However the conventional particle filter uses system transition as the proposal distribution, frequently resulting in poor priors for the filtering step. The main reason is that it is difficult, if not impossible, to accurately model the target's motion. Such a proposal distribution does not take into account the current observations. It is not a trivial task to devise a satisfactory proposal distribution for the particle filter. In this paper we advance a general augmented particle filtering framework for designing the optimal proposal distribution. The essential idea is to augment a second filter's estimate into the proposal distribution design. We then show that several existing improved particle filters can be rationalised within this general framework. Based on this framework we further propose variant algorithms for robust and efficient visual tracking. Experiments indicate that the augmented particle filters are more efficient and robust than the conventional particle filter.

Most of the countries have paid great attention to flood water level monitoring and tracking because flood may damages people's life and property. Since flood water level fluctuate highly nonlinear, it is very difficult to predict the flood water level. The particle filter algorithm is well known as a very effective solution for handling nonlinear problems. Thus, in this paper, this algorithm is applied to predict the flood water level. There are many variations of particle filter. This paper proposes Sequential Importance Sampling (SIS) particle filter to solve the above mentioned problem. SIS is the basic particle filter. However, the problems with SIS particle filter are the particle degeneration phenomenon, when after a few iterations only a few particles have nonzero weight. So, Sampling Importance Resampling (SIR) particle filter is also introduced as the improved particle filter. From the simulation results using Matlab, SIR particle filter outperforms SIS particle filter by comparing the Root Mean Square Error (RMSE) value.

A new algorithm is proposed using a sequential Monte Carlo method; it is robust even in nonlinear and non-Gaussian environments. In conjunction with current detection algorithms for tracking the channel matrix, the new algorithm improves the performance of the MIMO system under time-varying non-Gaussian channels due to Doppler shifts and carrier frequency offsets. Simulation results show that the proposed scheme exhibits robustness to time-selectivity with a few training symbols.

To resolve the problems of particle impoverishment and sample size dependency, particle swarm optimization (PSO) is introduced into generic particle filter (PF). This novel method, particle swarm optimized particle filter (PSOPF), incorporates the newest observations into sampling process and also optimizes that process. Through particle swarm optimization, particle samples are moved towards regions where particles have larger values of posterior density function. As a result, the impoverishment of particle filter is overcome and the sample size necessary for accurate state estimation is reduced dramatically. Two experiments show the validation of our method.

This paper presents an hybridization of particle filter and local search algorithms, called local search particle filter (LSPF), and its application to human-computer interaction. The proposed algorithm combines both sequential Monte Carlo (particle filter - PF) and local search methods to achieve an accurate real-time hand tracking. The system allows to control different mouse actions through a reduced set of hand movements and gestures. Hand are segmented using a skin-color model based on explicit RGB region definition. The proposed hybrid tracking method increases the performance of general particle filter. It also improves the quality of the hand tracking task (the standard deviation between hand spatial positions for LSPF is reduced a 75% with respect to the PF algorithm). More precisely, a local search enhances a hand-simulated mouse cursor to smoothly move and thus recognize gestures for performing their associate actions.