高速高机动目标自适应IMM跟踪算法  

Adaptive interacting multiple model algorithm for high maneuvering target tracking

作　　者：崔彦凯[1] 梁晓庚[1,2] 王志刚[3] 贾晓洪[2] 

机构地区：[1]西北工业大学自动化学院,西安710072 [2]中国空空导弹研究院科技部,河南洛阳471000 [3]西北工业大学航天学院,西安710072

出　　处：《计算机工程与应用》2014年第8期198-201,206,共5页Computer Engineering and Applications

摘　　要：针对目标运动过程中有转弯机动等复杂运动模式的高速高机动目标,设计了自适应两层IMM跟踪算法。该算法内层由改进的机动目标当前统计模型构成,把目标速度方向角作为伪测量值进行滤波,实时获得目标的角速度和角加速度;外层模型由常速模型和曲线模型构成,把内层模型得到的切向加速度和转弯角速度作为曲线模型参数,利用IMM算法进行滤波。仿真结果表明,该算法对高速高机动目标具有较高的跟踪精度,算法实现简单,具有一定的实际应用价值。A two-layer Interactive Multiple Model（IMM）tracking algorithm is put forward for high speed and high maneu-vering target, which includes complex maneuvering modes, especially turn maneuvering. The inner layer is composed of modified current statistical model. Turn angular velocity and angular acceleration can be attained by filtering the heading angle of target. The outer layer is composed by constant model and curvilinear model. Cross-track acceleration and turn angular velocity of curvilinear model parameter could be adjusted adaptively from inner layer model. The outer layer models are filtered by IMM. The simulations show that the algorithm has better tracking precision for high speed and high maneu-vering target. The algorithm is easily implemented, and the algorithm has some practical application value.

关 键 词：转弯机动 曲线模型 当前统计模型 自适应交互式多模型 

分 类 号：V249.1[航空宇航科学与技术—飞行器设计]