基于简化模型与自适应滤波的车载SINS静基座快速对准

Fast Alignment of Vehicle-based SINS Based on Simplified Model and Adaptive Filtering

摘要:

为了实现捷联惯性导航系统(strap-down inertial navigation system,SINS)快速初始对准,根据已有可观测性分析结果,通过理论分析和计算得到了扩展观测量时初始对准系统最优可观测状态量组合,在此基础上简化了对准模型,建立了新的系统方程；针对载车发动机启动或其他情况导致系统噪声无法精确统计,提出了运用基于强跟踪滤波原理的自适应卡尔曼滤波算法抑制滤波发散,加快收敛速度；仿真结果表明运用简化模型和自适应滤波在系统噪声不匹配时具有更快的收敛速度和更高的对准精度,车载实验结果也表明运用简化模型和自适应滤波可以实现快速对准。

Abstract:

To realize fast initial alignment of SINS, according to the results of the observability analysis, state combinations with best observability are found by theoretical analysis and calculation, then the alignment model is simplified and new system function is proposed. As for system noise is unknown when engine is starting, an adaptive Kalman filtering (KF)algorithm based on strong tracking filter theory is proposed, which could restrain filtering divergence and speed up the convergence. The simulation results show that the adaptive algorithm has faster convergence speed and higher precision when the system noises mismatches. The vehicle-based experiment result also shows that fast alignment can be realized with the application of simplified model and adaptive Kalman filter.