Motion Inference Using Sparse Inertial Sensors, Self-Supervised Learning, and a New Dataset of Unscripted Human Motion
In recent years, wearable sensors have become common, with possible applications in biomechanical monitoring, sports and fitness training, rehabilitation, assistive devices, or human-computer interaction. Our goal was to achieve accurate kinematics estimates using a small number of sensors. To accom...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-11-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/20/21/6330 |
| _version_ | 1797548708016947200 |
|---|---|
| author | Jack H. Geissinger Alan T. Asbeck |
| author_facet | Jack H. Geissinger Alan T. Asbeck |
| author_sort | Jack H. Geissinger |
| collection | DOAJ |
| description | In recent years, wearable sensors have become common, with possible applications in biomechanical monitoring, sports and fitness training, rehabilitation, assistive devices, or human-computer interaction. Our goal was to achieve accurate kinematics estimates using a small number of sensors. To accomplish this, we introduced a new dataset (the Virginia Tech Natural Motion Dataset) of full-body human motion capture using XSens MVN Link that contains more than 40 h of unscripted daily life motion in the open world. Using this dataset, we conducted self-supervised machine learning to do kinematics inference: we predicted the complete kinematics of the upper body or full body using a reduced set of sensors (3 or 4 for the upper body, 5 or 6 for the full body). We used several sequence-to-sequence (Seq2Seq) and Transformer models for motion inference. We compared the results using four different machine learning models and four different configurations of sensor placements. Our models produced mean angular errors of 10–15 degrees for both the upper body and full body, as well as worst-case errors of less than 30 degrees. The dataset and our machine learning code are freely available. |
| first_indexed | 2024-03-10T15:03:29Z |
| format | Article |
| id | doaj.art-d4a82ba5370946f298a680580c0dc2b8 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T15:03:29Z |
| publishDate | 2020-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-d4a82ba5370946f298a680580c0dc2b82023-11-20T19:59:19ZengMDPI AGSensors1424-82202020-11-012021633010.3390/s20216330Motion Inference Using Sparse Inertial Sensors, Self-Supervised Learning, and a New Dataset of Unscripted Human MotionJack H. Geissinger0Alan T. Asbeck1Department of Electrical & Computer Engineering, Virginia Tech, Blacksburg, VA 24061, USADepartment of Mechanical Engineering, Virginia Tech, Blacksburg, VA 24061, USAIn recent years, wearable sensors have become common, with possible applications in biomechanical monitoring, sports and fitness training, rehabilitation, assistive devices, or human-computer interaction. Our goal was to achieve accurate kinematics estimates using a small number of sensors. To accomplish this, we introduced a new dataset (the Virginia Tech Natural Motion Dataset) of full-body human motion capture using XSens MVN Link that contains more than 40 h of unscripted daily life motion in the open world. Using this dataset, we conducted self-supervised machine learning to do kinematics inference: we predicted the complete kinematics of the upper body or full body using a reduced set of sensors (3 or 4 for the upper body, 5 or 6 for the full body). We used several sequence-to-sequence (Seq2Seq) and Transformer models for motion inference. We compared the results using four different machine learning models and four different configurations of sensor placements. Our models produced mean angular errors of 10–15 degrees for both the upper body and full body, as well as worst-case errors of less than 30 degrees. The dataset and our machine learning code are freely available.https://www.mdpi.com/1424-8220/20/21/6330motion datasetkinematicsinertial sensorsself-supervised learningsparse sensors |
| spellingShingle | Jack H. Geissinger Alan T. Asbeck Motion Inference Using Sparse Inertial Sensors, Self-Supervised Learning, and a New Dataset of Unscripted Human Motion Sensors motion dataset kinematics inertial sensors self-supervised learning sparse sensors |
| title | Motion Inference Using Sparse Inertial Sensors, Self-Supervised Learning, and a New Dataset of Unscripted Human Motion |
| title_full | Motion Inference Using Sparse Inertial Sensors, Self-Supervised Learning, and a New Dataset of Unscripted Human Motion |
| title_fullStr | Motion Inference Using Sparse Inertial Sensors, Self-Supervised Learning, and a New Dataset of Unscripted Human Motion |
| title_full_unstemmed | Motion Inference Using Sparse Inertial Sensors, Self-Supervised Learning, and a New Dataset of Unscripted Human Motion |
| title_short | Motion Inference Using Sparse Inertial Sensors, Self-Supervised Learning, and a New Dataset of Unscripted Human Motion |
| title_sort | motion inference using sparse inertial sensors self supervised learning and a new dataset of unscripted human motion |
| topic | motion dataset kinematics inertial sensors self-supervised learning sparse sensors |
| url | https://www.mdpi.com/1424-8220/20/21/6330 |
| work_keys_str_mv | AT jackhgeissinger motioninferenceusingsparseinertialsensorsselfsupervisedlearningandanewdatasetofunscriptedhumanmotion AT alantasbeck motioninferenceusingsparseinertialsensorsselfsupervisedlearningandanewdatasetofunscriptedhumanmotion |