中北大学学报（自然科学版）2025，Vol.46Issue(6)：693-702,10.DOI:10.62756/jnuc.issn.1673-3193.2024.07.0013
一种误差补偿的多因耦合定位与轨迹规划方法
A Multifactor Coupled Localization and Trajectory Planning Approach with Error Compensation
摘要
Abstract
With the rapid development of intelligent connected vehicles,precise localization and trajectory planning of vehicles in complex environments have become urgent problems to be solved.Aiming at the issues of low localization accuracy under weak or failed GPS signals and the poor adaptability of traditional trajectory planning methods in complex road conditions,this paper proposed an error-compensated multi-factor coupled localization and trajectory planning approach.By analyzing multi-source fusion data,this paper designed a lateral positioning algorithm and a deflection angle confirmation algorithm,and constructed an intelligent fusion model based on error prediction and compensation.Combined with vehicle behavior,driving mode and perception data,a multi-factor coupled trajectory planner for vehicles in high-speed driving conditions was designed,and an optimization strategy for overtaking trajectories was proposed.Experimental results show that under effective GPS conditions,the lateral positioning accuracy of this method is improved by 3.94 dm and the speed positioning error is reduced by 0.22 m/s compared to GPS.In the case of GPS failure,the localization accuracy is significantly enhanced through error compensation.Furthermore,compared with the Model Predictive Control(MPC)controller,the trajectory planner designed in this paper exhibits higher stability and safety during lane-changing and overtaking,especially on wet roads with a coefficient of friction of 0.4,where the maximum lateral deviation is reduced by 0.93 meters and the maximum yaw rate tracking deviation is decreased by 0.03 radians per second.The proposed method effectively improves the performance of vehicles in terms of high-precision localization and trajectory planning,providing important guarantees for the safe driving of intelligent and connected vehicles.关键词
网联车辆/定位算法/多尺度定位/多源数据融合/轨迹规划Key words
connected vehicles/localization algorithm/multi-scale localization/multi-source data fusion/trajectory planning分类
交通工程引用本文复制引用
吴宏涛,王耀,马宇鹏..一种误差补偿的多因耦合定位与轨迹规划方法[J].中北大学学报（自然科学版）,2025,46(6):693-702,10.基金项目
山西省重点研发计划项目(202102070301019) （202102070301019）
山西省基础研究计划项目(202103021223464) （202103021223464）
山西交通控股集团有限公司科技研发项目(23-JKKJ-20) （23-JKKJ-20）
