Particle Fllter Research Articles

An Improved Particle Filter for Phase Measurement of X-ray Pulsar Signal

In order to solve the defect of traditional method based on accumulate pulse proflle process X-ray pulse signal from Pulsar, this paper changes the measurement of the X-ray pulsar signal delay to the matching between standard pulse signal proflle and the actual measured value. At flrst, Obtain pulsar standard signal proflle function by artiflcial neural network, and then the particle fllter algorithm is used to online estimate the pulsar signal delay time of observation. Secondly, To solve the sample impoverishment phenomenon of standard particle fllter, an improved particle fllter was designed and proved, which does not exist the degradation phenomena, and has high adaptive ability. Finally, because of the special features of the system, the function which indicates the output value precision is derived, and it can provide a reference for spacecraft navigation strategy. The comparisons of the new fllter and standard fllter illustrate the new fllter has better adaptability and higher accuracy.

The Tracking Algorithm of Reentry Ballistic Target

By analyzing the force of the ballistic reentry ∞ight, the three-dimensional motion model of ballistic missile has been constructed without ballistic coe‐cient. Because the aerodynamics force is complex in reentry trajectory, the estimation of aerodynamic force can convert to the estimation of ballistic coe‐cient. Extended Kalman Filter, Unscented Kalman Filter and Unscented Particle Filter have been used to design tracking algorithm of re-entry ballistic target. The results of the simulation verify that the motion model of the reentry ∞ight segment is correct. Advantages and disadvantages of difierent algorithms also have been analyzed. On tracking target of reentry ballistic missile, the extended Kalman fllter has poor stability and easily dissipates. The unscented particle fllter algorithm has large computation and slow convergence, but the adaptability, stability and accuracy in the ∞icker noise condition are better than UKF and EKF. The Unscented Kalman Filter is a compromise algorithm which has moderate amount of computation and accuracy.

A MULTI-SCAN MIXTURE PARTICLE FILTER FOR JOINT DETECTION AND TRACKING OF MULTIPLE TARGETS

In this paper, a novel algorithm named multi-scan mixture particle fllter is proposed for joint detection and tracking for a varying number of targets. The posterior distribution of multiple target state in a single-target state space is a multi-mode distribution with each mode corresponding to either a target or clutter. A general global posterior distribution is adopted in this work, which consists of existing components and new components. The new components are generated at each time step to capture the new modes due to newly appeared targets or clutter. In order to distinguish targets from clutter, multiple scan information is incorporated. The history of each component's associate weights is stored in a multi-scan sliding window, which is used to judge whether the component is from a target or clutter. Moreover, a novel sampling method which combines the likelihood sampling and prior sampling is proposed to draw particles from the desired parts of the state space at each time step. From the simulation results, it could be seen that the proposed algorithm can efiectively detect the appearance/disappearance of the targets as well as track the existing target.

A NOVEL INTERACTING MULTIPLE MODEL PARTICLE FILTER FOR MANEUVERING TARGET TRACKING IN CLUTTER

In this paper, a novel interactive multiple model particle fllter (IMMPF) is developed after a Bayesian estimator for maneuvering target tracking in clutter is derived theoretically. In this new algorithm, base state estimation and modal state estimation are completely separated to control the number of particles in each maneuvering mode. Only continuous-valued particles are used to numerically implement the procedure of Bayesian base state estimation, whereas modal state is estimated analytically without dependence on the number of particles. Density mixing is performed by aggregation of the total particles and mixing associated weights. To prevent the exponentially growing number of particles with the time, a resampling step is included following the interaction step. Through MC simulations, the new IMMPF has been tested and shown to provide reliable performance improvements with difierent sample sizes and under various clutter conditions.

H-INFINITY FILTER BASED PARTICLE FILTER FOR MANEUVERING TARGET TRACKING

In this paper, we propose a novel H-inflnity fllter based particle fllter (H1PF), which incorporates the H-inflnity fllter (H1F) algorithm into the particle fllter (PF). The basic idea of the H1PF is that new particles are sampled by the H1F algorithm. Since the H1F algorithm can fully take into account the current measurements, when the new algorithm calculates the proposed probability density distribution, the sampling particles can take advantage of the system current measurements to predict the system state. The particles distribution we obtained approaches nearer to the state posterior probability distribution and the H1PF alleviates the sample degeneracy problem which is common in the PF, especially when the maneuvers of the target tracking are large. Furthermore, the H1F algorithm can adjust gain imbalance factor by adjusting disturbance attenuation factor, from that the new algorithm can get the compromise between the accuracy and robustness and we can obtain satisfled accuracy and robustness. Some simulations and experimental results show that the proposed particle fllter performed better than the PF and the Kalman particle fllter (KPF) in tracking maneuvering target.

FUZZY-CONTROL-BASED PARTICLE FILTER FOR MANEUVERING TARGET TRACKING

In this paper, we propose a novel fuzzy-control-based particle fllter (FCPF) for maneuvering target tracking, which combines the advantages of standard particle fllter (SPF) and multiple model particle fllter (MMPF). That is, the SPF is adopted during non-maneuvering movement while the MMPF is adopted during maneuvering movement. The key point of the FCPF is to use a fuzzy controller, which could imitate the thoughts of human beings in some degree, to detect the target's maneuver and use a backward correction sub-algorithm to alleviate the performance degradation of MMPF caused by detection delay. Simulation results indicate that the proposed algorithm has a much better tracking accuracy than the SPF while keeps approximately equal computational complexity. Compared with MMPF, both algorithms have no tracking lost, but the tracking accuracy of the proposed FCPF is a little better than the MMPF, and the FCPF consumes about 66% computation time of the MMPF. Thus, the proposed algorithm ofiers a more efiective way for maneuvering target tracking.

Solving, Estimating, and Selecting Nonlinear Dynamic Models Without the Curse of Dimensionality

We present a comprehensive framework for Bayesian estima- tion of structural nonlinear dynamic economic models on sparse grids. The Smolyak operator underlying the sparse grids approach frees global approx- imation from the curse of dimensionality and we apply it to a Chebyshev approximation of the model solution. The operator also eliminates the curse from Gaussian quadrature and we use it for the integrals arising from ratio- nal expectations and in three new nonlinear state space fllters. The fllters substantially decrease the computational burden compared to the sequential importance resampling particle fllter. The posterior of the structural pa- rameters is estimated by a new Metropolis-Hastings algorithm with mixing parallel sequences. The parallel extension improves the global maximization property of the algorithm, simplifles the choice of the innovation variances, allows for unbiased convergence diagnostics and for a simple implementation of the estimation on parallel computers. Finally, we provide all algorithms in the open source software JBendge 4 for the solution and estimation of a

GREY PREDICTION BASED PARTICLE FILTER FOR MANEUVERING TARGET TRACKING

For maneuvering target tracking, we propose a novel grey prediction based particle fllter (GP-PF), which incorporates the grey prediction algorithm into the standard particle fllter (SPF). The basic idea of the GP-PF is that new particles are sampled by both the state transition prior and the grey prediction algorithm. Since the grey prediction algorithm is a kind of model-free method and is able to predict the system state based on historical measurements other than establishing a priori dynamic model, the GP-PF can signiflcantly alleviate the sample degeneracy problem which is common in SPF, especially when it is used for maneuvering target tracking. Simulations are conducted in the context of two typical maneuvering motion scenarios and the results indicate that the overall performance of the proposed GP-PF is better than the SPF and the multiple model particle fllter (MMPF) when the tracking accuracy, computational complexity and tracking lost probability are considered. The performance improvements can be attributed to that the GP-PF has both model-based and model-free features.

DISTRIBUTED PARTICLE FILTER FOR TARGET TRACKING IN SENSOR NETWORKS

In this paper, we present a distributed particle fllter (DPF) for target tracking in a sensor network. The proposed DPF consists of two major steps. First, particle compression based on support vector machine is performed to reduce the cost of transmission among sensors. Second, each sensor fuses the compressed information from its neighboring nodes with use of consensus or gossip algorithm to estimate the target track. Computer simulations are included to verify the efiectiveness of the proposed approach.

TARGET TRACKING WITH LINE-OF-SIGHT IDENTIFICATION IN SENSOR NETWORKS UNDER UNKNOWN MEASUREMENT NOISES

Tracking a target is a fundamental and crucial problem in wireless sensor networks. It is well known that non-line-of-sight (NLOS) propagation will signiflcantly degrade the tracking accuracy if its efiects are ignored. In this paper, a line-of-sight (LOS) identiflcation approach for range-based tracking systems is developed to discard the NLOS measurements. Based on Lp-norm LOS identiflcation strategy, a novel target tracking method is devised with the use of cost- reference particle fllter, which does not require the knowledge of the measurement noise distribution. Computer simulations are included to verify the efiectiveness of the proposed approach under difierent noise distributions.

A Robust and E±cient Algorithm for Mobile Robot Localization

Robot self-localization is realized by using adaptive particle fllter scheme with map-matching algorithm. A two-phase-updating method based on similarity measurements of range and geometry is proposed to improve the performance of general particle fllter localization which uses a laser range flnder. The algorithm not only enhances the robustness of the system, but also increases the computational e-ciency. Simulation results demonstrate that by using the proposed algorithm and nature characteristics in indoor environment, mobile robot can localize itself accurately and