基于似然分布自适应调整的SMC-PHDF算法

SMC-PHDF(Sequential Monte Carlo-Probability Hypothesis Density Filter)算法由于不受高斯和线性的限制,在目标跟踪领域有着广泛的应用;然而当系统量测噪声较大,很多样本的归一化权重很小而成为无效样本,最终导致SMC-PHDF算法滤波精度较低;针对这一问题提出似然分布自适应调整的SMC-PHDF算法,通过在更新步骤中自适应调整粒子权值,增加先验密度和似然的重叠区,从而达到提高滤波性能的目的;仿真结果表明:在系统量测噪声较大时该算法比传统SMC-PHDF算法的滤波效果有所提升.

SMC-PHDF Algorithm Based on Likelihood Distribution Adaptive Adjustment

The SMC-PHDF (Sequential Monte Carlo-Probability Hypothesis Density Filter) has been widely applied in the field of target tracking, because it is not restricted by Gaussian and linear model.However, when the value of system measurement noise becomes large, it brings a lot of samples normalized weights invalid and eventually leads to the problem of low precision filtering of SMC-PHDF algorithm.To solve this problem, the paper proposed an likelihood distribution adaptive SMC-PHDF algorithm, and it can adaptively adjust the value of particles in the update step and increase the overlap region between the prior density and the likelihood, thus to achieve the goal of performance to improve result of the filtering.The simulation results show that this algorithm has a better filtering effect than traditional SMC-PHDF algorithm when the measurement noise in the system is larger.