基于运动模型估计的分布式实时时间配准算法

在目标运动模型未知和参数不完备等复杂情况下,目前尚无较好的分布式实时时间配准方法。假设在较短时间内目标在某一方向上的运动可以用匀速、匀加速或匀变加速直线运动模型表示,文中提出了一种复杂情况下基于运动模型估计的分布式实时时间配准算法。首先,对目标航迹序列进行不同阶次的滑窗多项式拟合,并根据拟合误差在线估计目标的运动模型;然后,在状态参数不完备的情况下构造完整的运动状态估计向量,并利用Kalman预测方法进行实时时间配准。仿真结果表明,与传统方法相比,提出的方法具有更好的性能。

A Distributed Real-time Time Registration Algorithm Based on Motion Model Estimation

There is no excellent distributed real-time time registration algorithm under complex conditions that the motion model is not known and some parameters are absent. Supposing that the motion of a target in one direction can be represented by a uniform,uniformly accelerated, or ramp acceleration linear motion model in a short period of time, a distributed real-time time registration algorithm based on motion model estimation under complex conditions is proposed. First, the track sequence of a target is fitted by polynomials of various ranks in a slide window and the on-line motion model is estimated according to the fitting error. Then, the unabridged state vector is constructed when some state parameters are absent and the real-time time registration is carried out using Kalman prediction method. The simulation results show that the proposed method has better performance compared with traditional methods.