基于SAGBA优化粒子滤波的目标跟踪

在目标跟踪领域,粒子滤波技术有处理非线性非高斯问题的优势,但是标准粒子滤波在利用重采样方法解决退化现象时,会产生粒子贫化现象,导致滤波精度不稳定。针对这个问题,利用融合了模拟退火高斯扰动的蝙蝠算法对粒子滤波进行优化改进。该算法将粒子表征为蝙蝠个体,蝙蝠种群通过调节蝙蝠个体的频率、响度和脉冲发射率,伴随当前最优蝙蝠个体在目标图像区域进行搜索,并且可以对全局搜索和局部搜索进行动态决策,从而提高蝙蝠个体整体的质量与合理的分布;融合的模拟退火高斯扰动策略可以增强算法跳出局部最优的能力。为了验证该算法的优化性能,将该算法和标准粒子滤波算法进行性能分析对比。实验结果表明该算法的滤波性能优于标准粒子滤波算法。

Target Tracking Based on SAGBA Optimized Particle Filter

In the field of target tracking,particle filter technology has the advantage of dealing with nonlinear non-Gaussian problems.However,when using standard particle filter to solve the degradation phenomenon,the particle depletion phenomenon will occur,resulting in unstable filtering accuracy.In response to this problem,we adopt a bat algorithm that combines simulated annealing Gaussian perturbation to optimize and improve particle filtering.This algorithm characterizes particles as bat individuals.By adjusting the frequency,loudness and pulse emissivity of bat individuals,the bat population searches with the current optimal bat individuals in the target image area,and can dynamically determine global search and local search,so as to improve the overall quality and reasonable distribution of bat individuals.The fusion of simulated annealing Gaussian perturbation strategy can enhance the ability of the algorithm to jump out of local optimum.In order to verify the optimization performance of the proposed algorithm,it is compared with the standard particle filter algorithm in performance.Experiment shows that the filtering performance of this algorithm is better than the standard particle filtering algorithm.