基于模糊交互式多模型算法跟踪临近空间目标

Tracking Near Space Target Based on Fuzzy Interacting Multiple Model Algorithm

由于临近空间高超声速飞行器采用非弹道式机动飞行方式,飞行速度、高度、加速度不断变化,目标机动具有长周期机动特点,而且临近空间目标飞行速度快,具有较高的升阻比,且在大气层内长时间飞行,其运动轨迹往往呈现出"跳跃"特征,给地面防御系统对其定位和跟踪带来了巨大的困难,传统的滤波算法无法给出精确的目标状态估计,跟踪性能变差。为了更好满足非机动式弹道滤波的需要,通过模糊逻辑算法与交互式多模型算法相结合,形成模糊交互式多模型算法,实现对临近空间目标飞行器跟踪,并保证了定位跟踪精度在允许范围之内。Matlab仿真结果验证了算法的有效性。

As hypersonic vehicles in the near space often adopt the way of nonballistic maneuver flight with characteristics of high speed and lift-to-drag ratio, flying in atmosphere for a long time, whose motion trajectories are sometimes characterized by ＂hopping＂, their flight altitudes, flight velocities and flight accelerations are changing. Targets maneuver are characterized by long period maneuver, which has caused great difficulties for ground defense system to maneuver target localization and tracking. Tradition- al filtering algorithms obviously cannot give accurate target state estimation, tracking performance has de- clined. In order to better meet the needs of mobile trajectory filtering, a way of combining fuzzy logic al- gorithm and interacting multiple model algorithm （namely FIMM algorithm） is proposed for localization and tracking of near space target vehicle. This algorithm guarantees the accuracy of localization and tracking for target within the allowable range and validity of the algorithm is verified with Matlab simula- tion results.