论文部分内容阅读
无迹卡尔曼滤波(Unscented Kalman Filter,UKF)被广泛应用到工程实际中,但传统UKF在滤波过程进行无迹变换(Unscented Transform,UT)时的待选参数为固定值,会带来一定误差。为了获取最优的待选参数,提出了基于和声差分进化(Harmony Search Differential Evolution,HSDE)的UKF改进算法,并在目标跟踪中对该算法进行应用。和声差分进化算法对待选参数κ进行最优选择,跳出局部最优的现象还有很强的收敛性,通过改进进一步提高UKF算法滤波精度。Matlab仿真结果表明,基于和声差分进化的UKF改进算法精度更高。
Unscented Kalman Filter (UKF) has been widely used in engineering practice. However, the traditional parameters of UKF with unscented transform (Unscented Transform, UT) are fixed, resulting in certain errors . In order to obtain the optimal candidate parameters, an improved UKF algorithm based on Harmony Search Evolution (HSDE) is proposed and applied in target tracking. The harmonic difference evolutionary algorithm optimizes the choice parameter κ, and jumps out the local optimum and has strong convergence. The filtering accuracy of the UKF algorithm is further improved through improvement. Matlab simulation results show that the UKF improved algorithm based on harmonic differential evolution is more accurate.