多项式预测GNSS信号矢量跟踪算法

Polynomial prediction GNSS vector tracking algorithm

提出了一种多项式预测全球卫星导航系统(GNSS)矢量跟踪算法,不需要接收机运动参数的先验信息,不同于传统卡尔曼滤波器需要根据目标的运动状态选择合适的动态模型并且调节相应的状态噪声。对观测噪声进行自适应估计,实现对目标状态更为鲁棒的跟踪。仿真结果表明,建立的矢量跟踪模型能够准确跟踪目标状态,适应各种加速度的目标运动情况,在加速度发生突变时,也能保持鲁棒的跟踪,有效避免了传统卡尔曼模型由于参数选择带来的问题。

A polynomial prediction Global Navigation Satellite System（GNSS） vector tracking algorithm is proposed, which is independent of the prior information of the receiver’s kinematic parameters. It is different from traditional Kalman filter which requires a proper dynamic model and adjusting its state noise according to the target’s kinematic status. The adaptive measurement noise estimation is also applied to guarantee a more robust tracking process. Numerical simulations validate that the algorithm proposed is able to track the target’s status accurately, and performs well in situations with different accelerations. It can also keep tracking when there occurs abrupt change in acceleration, thus avoiding the problems caused by traditional Kalman models which require proper parameter selections.