利用地磁场测量制导火箭弹的滚转角是常用的一种方法。由于利用磁场测量时存在无法避免的误差,从而不能准确获取火箭弹滚转姿态。本文提出了两步自适应快速无迹卡尔曼滤波(TAUKF)算法,区别于现有算法,TAUKF的状态向量根据滚转角指标参数与测量电路的校正参数而提出,并且经过两步处理,在保证滤波结果精度的前提下,提高了滤波的收敛速度。通过仿真结果表明,所提出的TAUKF算法在收敛特性和计算成本方面优于现有的UKF算法。

The geomagnetic field is often used to measure the roll angle of guided rocket. Due to the unavoidable error in magnetic field measurement, the roll angle cannot be measured accurately. In this paper, a two-step Kalman filter (TAUKF) algorithm is proposed. Being different from the existing algorithm, the state vector of TAUKF is based on the roll angle and the correction parameter of the measurement circuit. And after two steps of processing, the convergence speed of the filter is improved while the accuracy of the filter result is ensured. The simulation results show that the proposed TAUKF algorithm is superior to the existing UKF algorithm in terms of convergence characteristics and computational cost.