城市遮挡环境下天空图辅助的GNSS/INS组合导航

GNSS/INS integrated navigation aided by sky images in urban occlusion environment

针对城市遮挡环境下,全球卫星导航系统信号受高层建筑物等遮挡的影响,导致非视距信号被接收,定位存在偏差的问题。该文提出一种基于天空图的NLOS信号识别与抑制方法,利用鱼眼相机获取天空图,对图像的天空区域进行分割,获取环境遮挡信息;根据投影模型,获取卫星投影到天空图像的位置,从而检测NLOS信号卫星;基于NLOS检测结果建立综合加权模型,减小NLOS信号观测值对定位的影响,从而提高了定位精度。为验证论文算法的有效性,我们采集了城市环境多源观测数据集进行实验,利用标准单点定位和GNSS SPP/INS松组合定位对提出的算法进行了验证,结果表明,在城市遮挡环境的动态实验中,SPP定位精度在E、N、U方向分别提高22.66%,13.96%和21.50%,GNSS SPP/INS松组合定位精度在E、N、U方向分别提高了23.03%、7.22%和30.49%。

According to the fact that global navigation satellite system signals are affected by occlusion such as high-rise buildings in urban occlusion environment,resulting in non-line-of-sight signals being received,leading to bias in positioning.In this paper,a method of NLOS signal recognition and suppression based on sky images is proposed,which first uses a fisheye camera to obtain sky images,segment the sky region of the image to obtain environmental occlusion information,then obtains the position of the satellite projection onto the sky image based on a projection model to detect NLOS signal satellites,and finally builds a comprehensive weighting model based on the NLOS detection results to reduce the NLOS signal observation on the positioning,thus improving the positioning accuracy.In order to verify the effectiveness of the algorithm of the paper,multi-source observation datasets of urban environment were collected for experiments.The proposed algorithm is verified using standard point positioning and GNSS SPP/INS integrated positioning.The results show that in the dynamic experiment of urban occlusion,the positioning accuracy of SPP is improved by 22.66%,13.96%and 21.50%in the east,north and up directions,respectively,and the GNSS SPP/INS integrated positioning accuracy is improved by 23.03%,7.22%and 30.49%in the east,north and up directions,respectively.