基于迭代平滑的GNSS/INS/轮速组合定位后处理算法

GNSS/INS/Wheel Speed Integrated Positioning Post-Processing Algorithm Based on Iterative Smoothing

近年来智能驾驶发展迅速,高精度地图的重要性受到越来越多的重视与关注。在城市街道、高架桥等非开阔场景下,全球卫星导航系统(GNSS)信号会出现濒繁丢失的情况,仅依靠传统的GNSS/INS松耦合组合定位精度无法满足高精度地图要求,因此需要通过离线后处理来提高精度,为此本文提出基于迭代平滑的GNSS/INS/轮速组合定位后处理算法。通过推导双向平滑算法,对GNSS/INS/轮速组合导航参数进行后处理。针对惯性测量单元零偏估计收敛慢,导致滤波解算精度较差的问题,采用迭代平滑策略离线补偿零偏。通过在实车试验中屏蔽GNSS信号模拟其丢失情况,验证了所提出算法的有效性。

Intelligent driving has been developing rapidly in recent years,the importance of high-precision maps has received more and more attention and attention.In non-open scenes such as urban streets and viaducts,GNSS signals will be frequently lost,and only relying on the traditional GNSS(Global Navigation Satellite System)/INS(Inertial Navigation System)loosely on the traditional GNSS/INSoosely coupled combined positioning accuracy cannot meet the requirements of high-precision maps,so it is necessary to improve the accuracy through offline post-processing.Wheel speed combined positioning post-processing algorithm.Aiming at the problem of immeasurable measurement information feedback in the GNSS signal lockout scenario,the vehicle speed information obtained by the wheel speed sensor is used for error feedback.A bidirectional smoothing algorithm is derived to post-process the combined GNSS/INS/wheel speed navigation parameters.Aiming at the problem that the zero-bias estimation of inertial measurement unit converges slowly,resulting in poor filtering and calculation accuracy,an iterative smoothing strategy is proposed.By shielding the GNSS signal in the real vehicle experiment to simulate its loss,the efTectiveness of the proposed iterative smoothing algorithm is verified.