基于Kalman滤波器的室内激光/INS融合定位方法

Kalman filter based laser/INS fusion approach for indoor location

讨论未知多边形室内环境下微型飞行器(MAV)的定位问题。利用惯性测量单元(IMU)测量飞行器运动参数,并计算飞行器的位置、航向。二维激光测距仪测量飞行器相对于障碍的位置,用迭代搜索/最小二乘算法匹配扫描数据,获得2次扫描的相对位移和转动。然后分析激光扫描匹配和惯性导航系统(INS)的误差,建立相应误差模型,并以此设计间接反馈校正Kalman滤波器。仿真结果显示,该激光/INS融合定位方法相比纯INS定位和无融合激光扫描匹配定位的精确度更高,并能将误差限制在一个较小的范围内。

This paper discusses the location problem of Micro Aerial Vehicles（MAV） in a unknown indoor environment. The motion parameters of the MAV can be measured by an Inertial Measurement Unit（IMU）, from which the position and heading angle can be derived. A 2D Laser Range Finder（LRF） is adopted to measure the position of obstacles. The transform of two laser scans can be computed by matching the data of two scans using a search/least squares algorithm. The error analysis is performed for the laser scan matching and Inertial Navigation System（INS）, and error models are built. An indirect feedback Kalman filter is designed based on these works. The simulation results indicate that the precision of fusion approach is higher than that of INS only location and that of laser scan matching only location. The error is limited to a small range.