多传感器融合在无人机室内三维定位中的应用

Application of Multi-sensor Data Fusion in UAV Indoor 3D Positioning

基于超宽带(Ultra Wide Band,UWB)技术,可以解决无人机的室内定位问题,但定位精度还不够高。针对上述情况,提出了基于无迹卡尔曼滤波(Unscented Kalman Filtering,UKF)的UWB定位技术和惯性测量单元(Inertial Measurement Unit,IMU),以及光流传感器相融合的内外环室内定位方法,以基于IMU测得的姿态控制为内环,UWB、光流测得的位置控制为外环,利用UWB在视距情况下得到的精确位置信息来修正IMU的累积误差,同时在非视距情况下,用IMU的短时精确位置信息对UWB的定位误差进行补偿,提高定位精度。最后设计了实物仿真平台。实验结果表明,上述室内定位方法实际定位精度小于10 cm,导航误差为0.7%,且能快速地进入稳定状态,满足无人机室内定位的需求,同时极大地改善了原有设备的定位精度。

Based on ultra-wideband(UWB)technology,the indoor positioning problem of UAVs can be solved,but the positioning accuracy is not high enough.In response to the above situation,the traceless Kalman filter(UKF)-based UWB localization technology and the inertial measurement unit(IMU)and optical flow sensors are proposed to fuse the inner and outer loop indoor localization method,with the attitude control based on the IMU measurement as the inner loop and the position control measured by UWB and optical flow as the outer loop,using the accurate position information obtained by UWB in the line-of-sight case to correct the accumulated error of IMU At the same time,the short-time accurate position information of IMU is used to compensate for the positioning error of UWB in the non-line-of-sight case to improve the positioning accuracy.Finally,a physical simulation platform is designed.The experimental results show that the actual positioning accuracy of the above indoor positioning method is less than 10 cm,and the navigation error is 0.7%,and it can enter the stable state quickly.It meets the demand of indoor positioning of UAVs and greatly improves the positioning accuracy of the original equipment at the same time.