基于自适应吸引半径的萤火虫算法的粒子滤波

针对粒子滤波算法对粒子数目的大量需求等弊端，提出一种基于改进的萤火虫算法的粒子滤波。首先，在萤火虫的亮度公式中引入观测值信息以提高算法跟踪的准确性；其次，提出自适应吸引半径参数来控制萤火虫群寻优时的吸引范围，使算法的实时性更好；最终利用萤火虫算法的迭代寻优来进行粒子更新。对比实验表明，该算法在跟踪精度和运行时间上都有所优化，说明该算法即使在粒子数目较少的条件下也能保证目标跟踪的准确性和实时性。

Particle filter based on firefly algorithm with adaptive attraction radius

For the large demand of particles when particle filter tracking, this paper proposed a particle filter algorithm based on the improved firefly algorithm. Firstly, in order to improve the accuracy of tracking, the improved algorithm introduced the observation value into the brightness formula of the firefly. Secondly, it proposed an adaptive attraction radius parameter to control the attraction range when the fireflies swarm optimized, so that the real-time performance would be better. Finally, it used the iterative optimization of the firefly algorithm to update the particles. Comparison experiments show that, the algorithm has been optimized for tracking accuracy and running time. It shows that the algorithm can guarantee the accuracy and real-time performance of the target tracking even if the number of particles is small.