基于交互式联邦多模型的多传感器融合算法

针对多传感器机动目标跟踪过程中的航迹滤波发散问题,提出了一种将联邦滤波器与交互式多模型滤波算法（IMM）相结合的交互式联邦多模型融合算法IFMM.在IFMM算法中各传感器均具有相同的滤波模型集合,各传感器在同一模型下所产生的滤波结果先采用联邦滤波算法进行融合,然后采用IMM算法对各模型融合结果进行综合,产生目标状态的全局估计.仿真结果表明,IFMM有效提高了机动目标运动状态估计的精确度和稳定性.

Multi-Sensor Fused Algorithm Based on Interacting Federated Multi-Model

In order to solve the filtering divergence problem in the maneuvering targets tracking of the multi-sensor, the interacting federated multi-model （IFMM） algorithm which combines the federated filter with interacting multi-model （IMM） is proposed. All sensors in IFMM maintain the same set of filtering models, and the filtering results from the same model of all sensors are fused by using the federated filter, afterwards the overall estimation of target is produced by integrating the local estimations of every model through the IMM algorithm. The simulation results show that the IFMM algorithm efficiently enhances the accuracy and stability of the maneuvering targets state estimation.