文献类型：期刊文献

中文题名：基于自适应NHC量测噪声的车载测向算法

英文题名：Vehicle-mounted orientation determination algorithm based on the adaptive non-holonomic constraints measurement noise

作者：陈熙源 颜亚强 孙可欣 高宁

第一作者：陈熙源

机构：[1]东南大学仪器科学与工程学院,南京210046;[2]贵州理工学院航空航天工程学院,贵阳550000

第一机构：东南大学仪器科学与工程学院,南京210046

年份：2026

卷号：47

期号：3

起止页码：223-232

中文期刊名：仪器仪表学报

外文期刊名：Chinese Journal of Scientific Instrument

收录：;北大核心:【北大核心2023】；

基金：贵州省科技计划(XKBF(2025)032)项目资助。

语种：中文

中文关键词：组合导航;非完整性约束;自适应噪声;航向估计

外文关键词：integrated navigation;nonholonomic constraints;adaptive noise;heading estimation

摘要：为了提高车载捷联航姿系统(AHRS)的定姿精度,提出利用非完整性运动约束(NHC)对航姿系统误差进行在线自主补偿,然而传统组合导航方案存在非完整性运动约束约束(NHC)失效和横向速度噪声建模不准确等问题,从而导致航向角误差累积,严重影响车载定位定向的可靠性与稳定性。针对车辆转弯过程中因轮胎侧偏效应引起的横向速度分量,提出一种基于自适应非完整性运动约束(NHC)量测噪声的车载测向算法。首先,基于二自由度车辆动力学模型,构建出前向速度与横向加速度的非线性耦合函数,实现对车辆运动过程中横向速度噪声特性的精准表征,完成非完整性运动约束(NHC)量测噪声的动态修正;同时,引入方差受限调节机制,根据车辆运动状态实时调整约束强度。由全球卫星导航系统/捷联惯性导航系统(GNSS/SINS)组合导航的实车实验表明,在连续弯道和环形绕行两种典型转弯场景下,在维持俯仰角、横滚角等姿态测量精度基本不变的前提下,显著改善航向角估计精度,与传统方法相比,航向角测量精度提升15%以上,且在全球卫星导航系统(GNSS)信号短时中断、遮挡等异常工况下表现出良好的鲁棒性与收敛性。算法从噪声建模与约束自适应两方面同步优化,有效解决了车辆转弯运动过程中组合导航系统航向角精度退化、滤波易发散等问题,可为地面无人载具、智能车载导航等系统提供更高精度、更高可靠性的航向姿态参考。
To improve the attitude determination accuracy of vehicle-mounted strapdown attitude and heading reference system(AHRS),the non-holonomic constraint(NHC)is adopted to compensate the errors of attitude and heading system online.However,traditional integrated navigation schemes suffer from problems such as the failure of non-holonomic constraint(NHC)and inaccurate modeling of lateral velocity noise,which lead to the accumulation of heading angle errors and seriously affect the reliability and stability of on-board positioning and orientation.Considering the lateral velocity component caused by the tire camber effect during vehicle turning,this study proposes a vehicle orientation algorithm based on the adaptive non-holonomic constraint(NHC)measurement noise.Firstly,based on the vehicle dynamics model with the two-degree-of-freedom,a nonlinear coupling function of forward velocity and lateral acceleration is constructed to accurately characterize the characteristics of lateral velocity noise during vehicle motion and complete the dynamic correction of non-holonomic constraint(NHC)measurement noise;Meanwhile a variance-limited adjustment mechanism is introduced to adjust the constraint intensity in real time according to the vehicle motion state.The real vehicle experiments of integrated navigation based on the global navigation satellite system/strapdown inertial navigation system(GNSS/SINS)demonstrate that,the proposed algorithm in both typical turning scenarios(continuous curves and circular driving)significantly improves the heading angle estimation accuracy while maintaining almost unchanged measurement accuracy of pitch and roll angles.The heading angle measurement accuracy is improved by more than 15%compared to the traditional method,which shows the good robustness and convergence in abnormal conditions such as short-term interruption and occlusion of global navigation satellite system(GNSs)signal.The algorithn simultaneously optimizes the noise modeling and constraint adaptation,which effectively solves the problems of heading angle accuracy degradation and filter divergence in the integrated navigation system during the vehicle turning motion.It can provide the higher-precision and higher-reliability heading and attitude references for the ground unmanned vehicles,intelligent vehicle navigation and other systems.