非高斯噪声情况下多平台机动目标被动跟踪新方法

摘要

To overcome the problem of passive sensors bearing-only target tracking, an augmented unscented Kalman filtering (UKF) algorithm is proposed. A variable dimension interacting multiple mode (1MM) algorithm is proposed by improving the classic IMM to solve the problem of the maneuvering target tracking by the interacting fusion of different dimension modes. In consideration of the real problem that the measuring noise may be the non-Gaussian noise is considered, the selfada-ptive robust filtering noise is introduced. Finally, A VDIMM-AAUKF algorithm is proposed. The maneuvering target tracking is implemented by multiple bearing-only passive sensors under Gaussian or non-Gaussian noise. Simulation results show that the algorithm has high tracking accuracy and good stability, and has a practical application value.%针对仅有角度测量信息条件下,被动传感器融合目标跟踪问题,提出了扩维UKF滤波算法；并对经典IMM进行改进提出变维IMM算法,利用不同维数模型之间的交互式融合解决对机动目标的跟踪问题；进一步考虑实际情况中往往存在的测量噪声为非高斯情况,引入自适应滤波方法.最终提出变维交互式多模型自适应抗差扩维无迹滤波方法(VDIMMAAUKF),成功实现了被动多传感器在高斯和非高斯噪声情况下对机动目标跟踪.仿真实验结果表明该算法跟踪精度高、稳定性好,具有较好的实际应用价值.