应用UKF顺序滤波的组合导航研究

Research on integrated navigation using UKF order filtering

针对组合导航中传统卡尔曼滤波(KF)需要计算雅可比矩阵,当模型过于复杂或者雅可比矩阵无法求取时就很难处理的问题,在UKF滤波方法的基础上,提出了UKF顺序滤波的方法.该方法应用UKF滤波,兼顾GPS信息及惯导信息,逐次利用新信息来更正估计值.将该方法用于船用组合导航进行了仿真,仿真结果证明了算法的有效性.

Kalman filtering requires the calculation of Jacobian matrix in integrated navigation,when the model is too complex or Jacobian matrix cannot be obtained it is difficult to deal with the issue,the UKF sequential filtering method was proposed. To solve these problems,the method used the UKF filter,the GPS information and navigation information,successively using new information to make correction estimates.The method was used in ship navigation,and the simulation results proved the effectiveness of the algorithm.