基于混合特征匹配的微惯性/激光雷达组合导航方法

MEMS IMU/LADAR Integrated Navigation Method Based on Mixed Feature Match

微惯性/激光雷达(MEMS IMU/LADAR)组合导航系统在室内应用时,由于室内结构化环境下环境特征(如点和线段)分布稀疏,传统的单一特征匹配算法存在观测盲区,易造成导航定位参数估计误差大的问题。基于此,研究了激光雷达自适应数据分割方法的点和线段的特征提取算法,提出了基于混合特征匹配观测模型的MEMS IMU/LADR扩展卡尔曼滤波(EKF)算法。同时,设计了MEMS IMU/LADR组合导航试验样机,在室内环境下通过试验对滤波算法进行了验证。结果表明:提出的算法在室内结构化环境下相比传统单一点或线特征匹配组合定位算法的定位精度可提高60%,对于小型旋翼无人飞行器在室内结构化环境中的高精度定位具有较高的参考意义。

Usually,there are observation blind areas existing in that the distribution of land marks like angular point and line segment is parse in indoor environment.Then the locating accuracy of vehicle and the estimating precision of land marks is affected by the lack of observed quantities when using the traditional single feature matching algorithm.To solve these problems,the adaptive segmentation method of LADAR is studied,and the extraction algorithm of different types of land marks is researched,then the extended Kalman filter is presented based on the mixed feature matching observation model.At the same time,a test prototype of MEMS IMU/LADAR integrated navigation system is designed.To validate the filtering fusion algorithm,the analysis and tests are carried out in the indoor environment.The test results show that the accuracy of proposed algorithm relative to the traditional algorithm（a single point or line feature match integrated location algorithm）in the structured indoor environment increass about 60%.This indicates that the algorithm is efficient when used to improve accuracy of the MEMS IMU/LADAR integrated navigation system.