基于重力模型阶次自适应的实时重力补偿系统研究

Research on real-time gravity compensation systembased on gravity model order adaptation

重力扰动已成为高精度惯性导航系统的一项主要误差源,对其进行补偿可进一步提升导航精度。通过EIGEN-6C4重力场模型可计算全球范围内的重力扰动,模型阶次越高,计算结果越精确,但相应的计算量和存储量越大,在传统的导航计算机上使用高阶模型很难满足实时性要求,而使用低阶模型会导致较大的截断误差。针对以上问题,根据模型阶次与空间分辨率的关系,以及重力扰动在惯导力学编排中的传播特性,研究了一种易于在嵌入式系统实现的可随载体运动状态降频率、降阶次的实时重力补偿方案,并采用DSP实现了软硬件方案。最后,通过半实物仿真验证了所述方案既能满足实时性要求,也能保证补偿精度:东向及北向位置精度最大提升649 m和686 m,平均提升效果可达2 190阶重力模型补偿后精度提升量级的94%。

Gravity disturbance has become a major error source in high-precision inertial navigation systems,and compensating it can further improve navigation accuracy.The EIGEN-6C4 gravity field model can calculate gravity disturbances on a global scale.The higher the model order is,the more accurate the calculation results are,however,the corresponding computational and storage costs become larger,and it is difficult to meet real-time requirements by using high-order models on traditional navigation computers,while using low-order models can lead to significant truncation errors.In response to the above issues,based on the relationship between model order and spatial resolution,as well as the propagation characteristics of gravity disturbances in inertial navigation mechanics arrangement,a real-time gravity compensation scheme that can be easily implemented in embedded systems and can be reduced in frequency and order with the motion state of the carrier is studied.The software and hardware solutions are implemented using DSP,and finally,the real-time performance and required compensation accuracy of the scheme are verified through semi-physical simulation,the maximum improvement of position accuracy in the east and north directions is 649 m and 686 m respectively,and the average improvement effect is 94%of the magnitude of the accuracy improvement after compensating the 2190-order gravity model.