组网冗余MEMS惯性传感器网络优化配置与融合处理方法

针对单个惯性传感器精度与可靠性问题,提出了一种组网冗余微机电系统(micro-electro-mechanical system, MEMS)惯性传感网络优化配置与融合处理方法。首先,研究了多惯性传感器节点的空间配置形式,给出了网络观测模型。利用节点之间的最优化冗余配置策略,提升系统测量可靠性。在此基础上,进一步考虑MEMS陀螺仪和加速度计的安装误差对输出结果的影响,提出了相应的误差模型及校正算法。以随机游走为主要误差源,设计了新型的卡尔曼滤波方法,实现了冗余配置下的高精度导航信息解算。最后,采用自主构建的实验系统进行试验,证明所提出的融合方法能够有效降低陀螺仪和加速度计的随机游走误差。车载试验进一步证明了所提出的方法及系统的有效性。

Networking redundant MEMS inertial sensor network optimal configuration and fusion processing method

To improve the precision and reliability of the single inertial sensor, an optimal configuration and fusion processing method of inertial sensing network for the network redundant micro-electro-mechanical system (MEMS) is proposed. Firstly, the spatial configuration of multi-inertial sensor nodes is investigated, and the network observation model is proposed. By utilizing the optimal redundancy configuration, the reliability of the system is improved. On the basis, the impacts of the installation errors for the MEMS gyroscope and the accelerometer on output are further considered, the error model and the correction algorithm are proposed. By setting the random walk as the major error source, a novel Kalman filter method is designed, and the navigation information with high precision is calculated under the condition of redundancy configuration. The experiments on a self-built system show that the proposed fusion method can substantially reduce the random walk errors of the gyroscope and the accelerometer. The on-vehicle test further verifies the effectiveness of the proposed method and system.