在雷达/红外复合制导机动目标跟踪背景下,针对非线性机动目标融合跟踪存在滤波器易发散问题,提出一种基于交互式多模型无迹卡尔曼滤波(IMM-UKF)的分布式加权融合算法。IMM具有对不同目标机动模式自适应跟踪的能力;UKF对观测数据进行滤波估计,避免了计算雅克比矩阵,克服EKF滤波方法受滤波初值影响大、易发散的缺点;分布式融合算法提高了系统抗干扰能力及对目标跟踪的有效性和跟踪精度。仿真结果表明:该算法在处理非线性系统机动目标跟踪融合结果误差均得到减少,更能提高目标跟踪滤波精度,增强了系统稳定性。

A distributed weighted fusion algorithm was presented based on interacting multiple model and unscented Kalman filter (IMM-UKF) in context of radar / IR combined guidance maneuvering target tracking for the problem that divergence prone to occur to filter in nonlinear maneuvering target integration tracking. IMM is capable of adaptively tracking different target maneuver mode. UKF was introduced to obtain state estimation to avoid calculating Jacobian matrix and overcoming EKF's disadvantage of dependence on initial value and divergence. Distributed fusion algorithm was introduced to improve anti-jamming capability and effectiveness of target tracking and tracking accuracy. Simulation results show that the algorithm can reduce errors of maneuvering target tracking fusion result, improve the accuracy of target tracking filter and enhance the stability of the system in the treatment of non-linear system.