【摘要】 针对基于传统粒子滤波的目标跟踪算法过程中由于重采样造成样本粒子的退化，使得估计精度下降和实时性较差的问题，提出一种基于人工免疫的渐消容积粒子滤波目标跟踪算法。该算法将容积粒子滤波与渐消滤波相融合产生自适应提议分布函数，同时根据粒子权重大小对样本粒子进行优化重组，对重组后的不稳定粒子进行人工免疫重采样。通过优化提议分布与重采样这两方面，使得目标跟踪系统具有更高的自适应性，并且保证粒子的多样化，有效地缓解粒子退化问题。仿真实验表明，该算法与基于传统的容积粒子滤波目标跟踪算法相比，跟踪精度大大提高，同时具有更高的实时性，降低了算法的复杂度。更多还原

【Abstract】 In order to solve the problem of the degradation of sample particles due to resampling in the process of target tracking algorithm based on traditional particle filtering, the reduced estimation accuracy, and the poor real-time performance, a target tracking algorithm based on artificial immunity particle filter with fading volume is proposed. An adaptive proposed distribution function was generated by the fusion of the volume particle filter and the fading filter, and the sample particles were optimized and recombined according to the particle weight. Artificial immune resampling was carried out for reconstituted unstable particles. By optimizing the proposed distribution and resampling, the target tracking system had higher adaptability and ensured the diversity of particles to effectively alleviate particle degradation. The simulation results show that the proposed algorithm can improve the tracking accuracy, improve the real-time performance and reduce the complexity of the algorithm.更多还原