Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2510.09539 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918158263123968 |
|---|---|
| author | Ferreira, Patrick Costa, Paula |
| author_facet | Ferreira, Patrick Costa, Paula |
| contents | We present IF-D, a large-scale inertial dataset designed to enable self-supervised and foundational learning for IMU time series. IF-D comprises continuous, long-duration multichannel recordings (accelerometer, gyroscope, magnetometer) sampled at 200Hz using a UM7 IMU mounted inside a 3D-printed spherical enclosure that promotes diverse, free rotations during vehicle traversal. The collection spans approximately 135 minutes of recording, yielding around 1.6 million samples across nine sensor channels. We describe the data acquisition setup, preprocessing, and calibration procedures (six-orientation accelerometer calibration, stationary gyroscope bias estimation, and ellipsoid fitting for magnetometer hard-/soft-iron correction), and provide quantitative calibration results. IF-D is designed to mitigate platform specific motion bias and expose models to both physical dynamics and typical measurement noise, thereby facilitating robust representation learning and downstream tasks such as event detection, motion mode recognition, and inertial navigation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_09539 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | IF-D: A High-Frequency, General-Purpose Inertial Foundation Dataset for Self-Supervised Learning Ferreira, Patrick Costa, Paula Signal Processing 68T05, 94A12 I.2.6; I.5.4; I.4.8 We present IF-D, a large-scale inertial dataset designed to enable self-supervised and foundational learning for IMU time series. IF-D comprises continuous, long-duration multichannel recordings (accelerometer, gyroscope, magnetometer) sampled at 200Hz using a UM7 IMU mounted inside a 3D-printed spherical enclosure that promotes diverse, free rotations during vehicle traversal. The collection spans approximately 135 minutes of recording, yielding around 1.6 million samples across nine sensor channels. We describe the data acquisition setup, preprocessing, and calibration procedures (six-orientation accelerometer calibration, stationary gyroscope bias estimation, and ellipsoid fitting for magnetometer hard-/soft-iron correction), and provide quantitative calibration results. IF-D is designed to mitigate platform specific motion bias and expose models to both physical dynamics and typical measurement noise, thereby facilitating robust representation learning and downstream tasks such as event detection, motion mode recognition, and inertial navigation. |
| title | IF-D: A High-Frequency, General-Purpose Inertial Foundation Dataset for Self-Supervised Learning |
| topic | Signal Processing 68T05, 94A12 I.2.6; I.5.4; I.4.8 |
| url | https://arxiv.org/abs/2510.09539 |