论文部分内容阅读
为解决粒子滤波算法中存在的权值退化和实时性差的问题,提出了一种改进的权值优化组合粒子滤波算法(impWOPF),该算法通过对粒子权值设定门限Thershold,剔除权重小于Thershold的粒子,减少不必要的粒子运算,然后对小于粒子群权值均值的粒子进行权值优化组合,以增大小权值粒子的权值,保持了粒子多样性,提高了算法的实时性。仿真结果表明,该算法能够在保证估计精度的同时,有效降低重采样过程中的计算量,有利于实时信号的处理。
In order to solve the problem of weight degradation and poor real-time performance in particle filter, an improved weighted optimal combination particle filter (impWOPF) is proposed. By setting the threshold of the weight of the object, Thershold, the weight of rejection is less than Thershold , And reduce unnecessary particle operations. Then the weights of particles with weights less than the average particle group are combined optimally to increase the weight of the small-valued particles, maintain the particle diversity and improve the real-time performance of the algorithm. The simulation results show that the proposed algorithm can reduce the amount of computation during resampling while maintaining the accuracy of estimation and is beneficial to the real-time signal processing.