In this paper, a low-cost small-sized strap-down inertial navigation system (SINS)—Gyrolab GL-VG 109—is studied. When the system is installed on an unmanned vehicle and works in autonomous mode, it is difficult to determine the navigation parameters of the unmanned vehicle. Correcting the SINS information from the Global Navigation Satellite System (GNSS) can significantly increase the determination accuracy of the navigation parameters. However, this is only available when the GNSS signals are stable. A new adaptive estimation algorithm that can automatically detect, evaluate, and process the abnormal measurements is proposed in the present work. The determination of the navigation parameters can reach the third accuracy class using the proposed method. The effectiveness of the algorithm is verified by the mathematical simulation and the experimental tests (with a real SINS GL-VG 109), which are conducted in urban environments with a GNSS signal containing 15% and 40% abnormal measurements. The results show that the proposed method can significantly reduce the impact of abnormal measurements and improve the estimation accuracy.

中文翻译：

---

异常测量条件下无人车辆低成本MEMS-SINS误差的自适应估计算法

本文研究了一种低成本的小型捷联惯性导航系统（SINS）-Gyrolab GL-VG 109。当系统安装在无人驾驶车辆上并以自主模式工作时，很难确定无人驾驶车辆的导航参数。校正来自全球导航卫星系统（GNSS）的SINS信息可以显着提高导航参数的确定精度。但是，仅当GNSS信号稳定时才可用。本文提出了一种新的自适应估计算法，该算法可以自动检测，评估和处理异常测量值。使用所提出的方法，导航参数的确定可以达到第三精度等级。通过在城市环境中使用包含15％和40％异常测量值的GNSS信号进行的数学模拟和实验测试（使用真实的SINS GL-VG 109）验证了该算法的有效性。结果表明，该方法可以显着减少异常测量的影响，提高估计精度。