静止条件下的月球车INS/CNS自主初始化方法

New INS/CNS autonomous initialization method for lunar rovers on stationary base

月球车工作前自身初始位置和姿态的确定对提高其导航能力具有重要作用,这一过程也称为初始化,初始化精度将严重影响其后续导航性能。针对月面特殊环境,现有的应用于地面环境的初始化方法不再适用这一问题,提出了一种静止条件下的月球车INS/CNS自主初始化方法。该方法分为粗初始化和精初始化两个阶段,粗初始化主要为精初始化提供初始参数。精初始化则综合利用惯性导航解算得到的水平速度、星光方位矢量和天体高度量测信息,并在考虑加速度计偏置引起的天体高度误差的基础上建立了精确的天体高度量测方程,精确估计月球车的初始位置和姿态。半物理仿真实验表明当陀螺漂移为0.1(°)/h、加速度计偏置为10?g、星敏感器精度为3″时,采用所提方法的初始位置估计精度优于30 m,初始姿态估计精度优于10″,是一种非常有效的月球车自主初始化方法。

The determination of initial position and attitude of lunar rover has great influence on its navigation performance. The initialization accuracy has significant impact on the overall navigation accuracy. Since the traditional initial alignment method is not suitable for the lunar special environment, this paper presents a new autonomous initialization method for lunar rover based on INS/CNS integration. The method is divided into two stages: coarse initialization stage and fine initialization stage. The coarse initialization stage is mainly used to provide the initial parameters for the fine initialization stage. In the fine initialization stage, the horizontal velocity errors of INS, starlight vectors and star altitudes are used as measurements. The star altitude error caused by the biases of accelerometers is taken into account and its corresponding measurement equation is accurately established. Based on this, the position and attitude of the rover is estimated accurately. Semi-physics experiments show that the position precision is better than 30 m and the attitude precision is better than 10″ when the gyroscope drift is 0.1(°)/h, the accelerometer bias is 10 μg and the star sensor accuracy is 3″. These results demonstrate that it is a promising and attractive autonomous initialization method for lunar rovers.