顾及运动约束的自适应UWB/IMU组合定位方法

Adaptive UWB/IMU integrated localization method with motion constraints

针对室内复杂环境下超宽带(UWB)距离测量容易受到信号非视距(NLOS)传播影响以及UWB/惯性测量单元(IMU)组合定位系统不稳定的问题,该文提出了一种基于模糊自适应滤波的改进型UWB/IMU组合定位方法。通过构建运动约束条件并采用卡尔曼滤波,对UWB原始距离测量值进行实时处理,能够有效解决UWB距离测量值缺失、异常情况以及减小NLOS误差。在数据融合阶段,提出基于模糊自适应滤波的数据融合策略,由预测残差向量构造统计量,基于模糊推断系统计算自适应因子,最终实现高精度的UWB/IMU自适应组合定位。典型室内场景的动态定位实验结果表明,提出方法的定位轨迹与真实轨迹吻合度高,90%的径向位置误差在0.25 m以内,且其均值仅为0.128 m,定位精度优于其他三种方法。

To solve the problems that ultra-wideband(UWB)distance measurement in harsh indoor environments is susceptible to the non-line-of-sight(NLOS)propagation of signal and the instability of the UWB/inertial measurement unit(IMU)integrated localization system,the paper proposes an improved UWB/IMU integrated localization method based on fuzzy adaptive filter.By constructing motion constraints and using Kalman filter,the raw UWB distance measurements are processed in real time,which can effectively deal with the missing and abnormal UWB distance measurements and reduce NLOS errors.In the stage of data fusion,a data fusion strategy based on fuzzy adaptive filter is proposed,where the statistic is constructed from the predicted residual vector,and the adaptive factor is computed based on a fuzzy inference system.Finally,the high-precision adaptive UWB/IMU integrated localization is realized.The experimental results of dynamic localization in a typical indoor scenario show that the localization trajectory of this method is in good agreement with the real trajectory,and 90%of the radial position error is within 0.25 m,and its average value is only 0.128 m,which is better than the other three methods in terms of localization accuracy.