基于UPF的车辆紧耦合定位算法

Tightly-coupled positioning solution for land vehicle based on UPF

针对城市环境下车辆定位精度低、可靠性差且成本高的问题,通过融合GPS/北斗双模卫星导航定位系统和其他低成本的车载传感器信息设计了一种车辆紧耦合定位算法。首先基于低成本的降维惯性导航定位系统和双模卫星导航定位系统建立了非线性系统状态模型和观测模型;然后使用了一种改进无迹粒子滤波算法来进行多传感器信息的融合,在信息融合过程中,运用了一种模糊算法,根据卫星高度角和载波信噪比特征对伪距的观测噪声协方差值进行了自适应调整,从而提高了定位精度;最后,进行了实验评估。实验结果表明:该定位方法相比基于传统粒子滤波的定位方法能有效提高定位精度,具有可行性。

Aiming at achieving reliable,accurate and low-cost vehicle positioning performance in urban environment,a tightly-coupled positioning solution was researched,which integrates the dual-constellation GNSSs(i.e.GPS,BeiDou)and other low-cost complementary in-vehicle sensors.First,the nonlinear filtermodel was established based on a cost-effective reduced inertial sensor system(RISS)and dual-constellation GNSSs.Then,an improved unscented particle filter algorithm was applied to achieve the global fusion.In the implementation of fusion,a fuzzy calibration logic was introduced to adaptively adjust the noise covariance of each pseudo-range measurement based on elevation angle and carrier-to-noise ratio.Finally,the proposed solution was evaluated through experiments.The results indicate that the proposed solution can improve the positioning performance effectively compared with the conventional particle filter,which validates the feasibility of the proposed solution.