基于平方根无迹卡尔曼滤波平滑算法的水下纯方位目标跟踪（英文）
发布时间：2019-05-13 16:35
【摘要】：为了避免被动跟踪中非线性带来的计算复杂化及跟踪精度的下降,提出将平方根无迹卡尔曼滤波平滑算法(SR-UKFS)应用到水下纯方位目标跟踪。SR-UKFS利用Rauch-Tung-Striebel(RTS)平滑算法将平方根无迹卡尔曼滤波(SR-UKF)作为前向滤波算法得到的目标状态估计向后平滑,得到前一时刻目标状态估计,再利用该状态估计值进行再次滤波得到当前时刻目标状态估计。该算法得到的前一时刻的目标状态估计更加精确,从而进一步提高了目标跟踪的精度。最后,通过对SR-UKFS算法和SR-UKF算法的跟踪性能进行了对比分析和验证,仿真结果表明在相同条件下,SR-UKFS算法能减少59%的位置误差和54%的速度误差,SR-UKFS算法应用于水下纯方位目标跟踪系统是有效的,为水下纯方位目标跟踪系统的工程实现提供了非常有价值的参考。
[Abstract]:In order to avoid the computational complexity and the decrease of tracking accuracy caused by inlinearity in passive tracking, In this paper, the square root unscented Kalman filter smoothing algorithm (SR-UKFS) is applied to underwater azimuth target tracking. Sr-UKFS uses the Rauch-Tung-Striebel (RTS) smoothing algorithm to use the square root unscented Kalman filter (SR-UKF) as the square root unscented Kalman filter (SR-UKF). The target state estimated by the forward filtering algorithm is smoothed backward. The target state estimation at the previous time is obtained, and then the target state estimation at the current time is obtained by using the state estimation value to filter again. The target state estimation obtained by the algorithm is more accurate at the previous time, which further improves the accuracy of target tracking. Finally, the tracking performance of SR-UKFS algorithm and SR-UKF algorithm is compared and verified. The simulation results show that the SR-UKFS algorithm can reduce the position error by 59% and the speed error by 54% under the same conditions. The application of SR-UKFS algorithm to underwater azimuth-only target tracking system is effective, which provides a very valuable reference for the engineering implementation of underwater azimuth-only target tracking system.
【作者单位】： 中国船舶重工集团第七一六研究所;南京理工大学自动化学院;
【基金】：国家自然科学基金(61473153,61301217)
【分类号】：TB56
本文编号：2476035
[Abstract]:In order to avoid the computational complexity and the decrease of tracking accuracy caused by inlinearity in passive tracking, In this paper, the square root unscented Kalman filter smoothing algorithm (SR-UKFS) is applied to underwater azimuth target tracking. Sr-UKFS uses the Rauch-Tung-Striebel (RTS) smoothing algorithm to use the square root unscented Kalman filter (SR-UKF) as the square root unscented Kalman filter (SR-UKF). The target state estimated by the forward filtering algorithm is smoothed backward. The target state estimation at the previous time is obtained, and then the target state estimation at the current time is obtained by using the state estimation value to filter again. The target state estimation obtained by the algorithm is more accurate at the previous time, which further improves the accuracy of target tracking. Finally, the tracking performance of SR-UKFS algorithm and SR-UKF algorithm is compared and verified. The simulation results show that the SR-UKFS algorithm can reduce the position error by 59% and the speed error by 54% under the same conditions. The application of SR-UKFS algorithm to underwater azimuth-only target tracking system is effective, which provides a very valuable reference for the engineering implementation of underwater azimuth-only target tracking system.
【作者单位】： 中国船舶重工集团第七一六研究所;南京理工大学自动化学院;
【基金】：国家自然科学基金(61473153,61301217)
【分类号】：TB56
【相似文献】
相关期刊论文 前1条
1 孙开敏;李德仁;眭海刚;;基于多尺度分割的对象级影像平滑算法[J];武汉大学学报(信息科学版);2009年04期
，本文编号：2476035
本文链接：https://www.wllwen.com/guanlilunwen/gongchengguanli/2476035.html
