机动目标跟踪的一种变维交互改进算法

为了满足工程需要，结合变维卡尔曼滤波器和α－β算法的优点，提出了一种卡 尔曼和α－β变维交互替代目标跟踪算法。通过加入机动检测器监视机动，在目标发生 机动时，采用高阶维数模型和卡尔曼滤波器；机动消失后，退回到低阶维数模型和α－β 滤波器，从而实现了对机动和非机动目标的自适应跟踪，克服了因转弯机动引起的误差突 跳，并显著地减少了计算量。通过Monte Carlo 仿真进一步验证了改进算法的合理性和实 用性。

An improved variable dimension and interactive substitution algorithm for maneuvering target tracking

To meet engineering requirements,considering the advantages of Kalman filter(KF) and α-β algorithm,the algorithm of variable dimension and interactive substitution for KF and α-β is proposed.Through putting in a maneuvering detecter to monitor target movements,higher dimension model and KF are picked to track when target maneuvers,otherwise,choosing lower dimension model and α-β filter when maneuvering fades away.In this way,the improved algorithm can achieve adaptive tracking between maneuvering target and non-maneuvering target.It overcomes the error caused by turn maneuvering and reduces the calculation remarkably.Monte Carlo simulation demonstrates the availability and practicality of the improved algorithm.