Shaohua Hong

A Low-Power Memory-Efficient Resampling Architecture for Particle Filters

In this paper, we propose a compact threshold-based resampling algorithm and architecture for efficient hardware implementation of particle filters (PFs). By using a simple threshold-based scheme, this resampling algorithm can reduce the complexity of hardware implementation and power consumption. Simulation results indicate that this algorithm has approximately equal performance with the traditional systematic resampling (SR) algorithm when the root-mean-square error (RMSE) and lost track are considered. Experimental comparison of the proposed hardware architecture with those based on the SR and the residual systematic resampling (RSR) algorithms was conducted on a Xilinx Virtex-II Pro field programmable gate array (FPGA) platform in the bearings-only tracking context, and the results establish the superiority of the proposed architecture in terms of high memory efficiency, low power consumption, and low latency.

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.

TRACKING AIRBORNE TARGETS HIDDEN IN BLIND DOPPLER USING CURRENT STATISTICAL MODEL PARTICLE FILTER

This paper aims at finding an algorithm featuring good estimation performance and easy hardware implementation for tracking airborne target hidden in blind Doppler. Incorporating the current statistical model which is effective in dealing with the maneuvering motions that most blind Doppler issues are caused, a current statistical model particle filter (CSM-PF) is presented in this paper for tracking airborne targets hidden in blind Doppler. Simulation results demonstrate that the proposed CSM-PF shows similar performance with the interacting multiple model particle filter (IMM-PF) in terms of tracking accuracy and track continuity, but it avoids the difficulty of model selection for maneuvering targets. In addition, when hardware implementation is considered, the proposed CSM-PF has lower processing latency, fewer resource utilization and lower hardware complexity than the IMM-PF.

Novel Roughening Algorithm and Hardware Architecture for Bearings-Only Tracking Using Particle Filter

In the bearings-only tracking problem, usually the uncertainty of process model is small compared with the uncertainty of measurement and this will lead to severe sample impoverishment when using particle filter (PF). In this paper, we proposed a novel roughening algorithm and its hardware architecture to solve this problem. The reasonable distribution of the roughening jitter is calculated from the innovations related to the survived particles. Simulation results indicate that for the bearings-only tracking problem, the PF with the proposed roughening algorithm outperforms the general PF and the PF with the typical roughening algorithm. Experimental study indicates that this roughening algorithm can be efficiently implemented in hardware and can effectively solve the bearings-only tracking problem with fast processing rate and low complexity.

Easy-hardware-implementation MMPF for Maneuvering Target Tracking: Algorithm and Architecture

In this paper, we present an easy-hardware-implementation multiple model particle filter (MMPF) for maneuvering target tracking. In the proposed filter, the sampling importance resampling (SIR) filter typically used for nonlinear and/or non-Gaussian application is extended to incorporating multiple models that are composed of a constant velocity (CV) model and a “current” statistical (CS) model, and the Independent Metropolis Hasting (IMH) sampler is utilized for the resampling unit in each model. Compared with the bootstrap MMPF, the proposed MMPF requires no knowledge of models and model transition probabilities for different maneuvering motions, and keeps a constant number of particles per model at all times. This allows a regular pipelined hardware structure and can be implemented in hardware easily. Furthermore, using the IMH sampler for the resampling unit avoids the bottleneck introduced by the traditional systematic resampler and reduces the latency of the whole implementation. Simulation results indicate that the proposed filter has approximately equal tracking performance with the bootstrap MMPF. Hardware architecture of the IMH sampler and its corresponding sample unit are presented, and a parallel architecture consisting of CV model processing element (PE), CS model PE and a central unit (CU) is described. The proposed architecture is evaluated on a Xilinx Virtex-II Pro FPGA platform for a maneuvering target tracking application and the results show many advantages of the proposed MMPF over existing approaches in terms of efficiency, lower latency, and easy hardware implementation.

Novel multiple-model probability hypothesis density filter for multiple maneuvering targets tracking

In this paper, we present a novel multiple-model probability hypothesis density (MMPHD) filter for multiple maneuvering targets tracking. In the proposed MMPHD filter, the multiple models are composed of two models, namely a constant velocity (CV) model and a “current” statistical (CS) model, and the PHD is approximated by a set of weighted random samples propagated over time using sequential Monte Carlo (SMC) methods. This resulting filter requires no knowledge of models and model transition probabilities for different maneuvering motions. Simulation results demonstrate that compared with the standard MMPHD filter, the proposed filter shows similar tracking performances but has faster processing rate.

Current statistical model probability hypothesis density filter for multiple maneuvering targets tracking

The probability hypothesis density (PHD) filter, which propagates only the first moment (or PHD) instead of the full target posterior, has been shown to be a computationally efficient solution to multi-target tracking problems. Incorporating the current statistical model that is effective in dealing with the maneuvering motions, this paper proposes a current statistical model PHD (CSMPHD) filter for multiple maneuvering targets tracking. This proposed filter approximates the PHD by a set of weighted random samples propagated over time based on the current statistical model using sequential Monte Carlo (SMC) methods. Simulation results demonstrate that the proposed filter shows similar performances with the multiple-model PHD (MMPHD) filter, but it avoids the difficulty of model selection for maneuvering targets and has faster processing rate.

Compact resampling algorithm and hardware architecture for paticle filters

In this paper, we propose a compact threshold-based resampling algorithm and architecture for efficient hardware implementation of particle filters. By using a simple threshold-based scheme and assigning each particle a weight independent of its previous value, this resampling algorithm can reduce the complexity of hardware implementation. Simulation results from Matlab indicate that this algorithm has approximately equal performance with the traditional systematic resampling (SR) algorithm when the RMSE is considered. Compact hardware architecture for resampling is presented and the bearings-only tracking problem is used for illustration and evaluation. Experimental study on a Xilinx Virtex 2 pro FPGA platform shows that this hardware architecture is efficient in terms of resource usage and latency.

Easy-hardware-implementation MMPF for maneuvering target tracking

In this paper, we present an easy-hardware-implementation multiple model particle filter (MMPF) for maneuvering target tracking. In this proposed filter, the sampling importance resampling (SIR) filter is extended to multiple models that consist of two models, namely a constant velocity (CV) model and a “current” statistical (CS) model, and the Independent Metropolis Hasting (IMH) sampler is utilized for the resampling step in each model. Compared with the standard MMPF, the proposed MMPF requires no knowledge of models and model transition probabilities for different maneuvering motions, and keeps a constant number of particles per model at all times. This allows a regular pipelined hardware structure and can be implemented in hardware easily. Furthermore, using the IMH sampler for the resampling step avoids the bottleneck introduced by the traditional systematic resampler and reduces the latency of the whole implementations. Simulation results indicate that the proposed filter shows approximately equal tracking performance with the standard MMPF.