This article proposes a learning method for denoising gyroscopes of Inertial Measurement Units (IMUs) using ground truth data, and estimating in real time the orientation (attitude) of a robot in dead reckoning. The obtained algorithm outperforms the state-of-the-art on the (unseen) test sequences. The obtained performances are achieved, thanks to a well-chosen model, a proper loss function for orientation increments, and through the identification of key points when training with high-frequency inertial data. Our approach builds upon a neural network based on dilated convolutions, without requiring any recurrent neural network. We demonstrate how efficient our strategy is for 3D attitude estimation on the EuRoC and TUM-VI datasets. Interestingly, we observe our dead reckoning algorithm manages to beat top-ranked visual-inertial odometry systems in terms of attitude estimation although it does not use vision sensors. We believe this article offers new perspectives for visual-inertial localization and constitutes a step toward more efficient learning methods involving IMUs. Our open-source implementation is available at https://github.com/mbrossar/denoise-imu-gyro.

中文翻译：

---

用于开环姿态估计的深度学习去噪 IMU 陀螺仪

本文提出了一种使用地面实况数据对惯性测量单元 (IMU) 的陀螺仪进行去噪的学习方法，并在航位推算中实时估计机器人的方向（姿态）。获得的算法在（看不见的）测试序列上优于最先进的算法。由于精心选择的模型、适当的方向增量损失函数以及在使用高频惯性数据训练时识别关键点，获得了所获得的性能。我们的方法建立在基于扩张卷积的神经网络之上，不需要任何循环神经网络。我们展示了我们的策略在 EuRoC 和 TUM-VI 数据集上进行 3D 姿态估计的效率。有趣的是，我们观察到我们的航位推算算法在姿态估计方面成功击败了排名靠前的视觉惯性里程计系统，尽管它没有使用视觉传感器。我们相信这篇文章为视觉惯性定位提供了新的视角，并朝着更有效的 IMU 学习方法迈出了一步。我们的开源实现可在 https://github.com/mbrossar/denoise-imu-gyro 获得。