噪声未知情况下的自适应无迹粒子滤波算法

Adaptive unscented particle filter algorithm under unknown noise

为了有效地解决系统噪声未知情况下的目标跟踪问题,提出了一种自适应无迹粒子滤波算法。该算法采用改进的Sage-Husa估计器对系统未知噪声的统计特性进行实时估计和修正,并与无迹卡尔曼滤波器相结合产生优选的建议分布函数,降低系统估计误差的同时有效提升了系统的抗噪声能力。实验结果表明:本文方法明显地改善了系统噪声未知情况下目标的跟踪精度和稳定性。

In order to solve the target tracking problem when the statistic characteristics of the system are unknown, an adaptive unscented particle filter algorithm is proposed. This algorithm estimates and corrects the statistic characteristics of the unknown system noise in real-time using improved Sage- Husa estimator. Combining with unscented Kalman filter, the algorithm produces the optimal proposal distribution function. This method effectively reduces the estimation error and improves the anti-noise ability of the system. Theoretical analysis and experiments show that the new method can significantly improve the accuracy and stability of target tracking when the statistic characteristics of the system are unknown.