RIOT: Recursive Inertial Odometry Transformer for Localisation from Low-Cost IMU Measurements
Inertial localisation is an important technique as it enables ego-motion estimation in conditions where external observers are unavailable. However, low-cost inertial sensors are inherently corrupted by bias and noise, which lead to unbound errors, making straight integration for position intractabl...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-03-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/23/6/3217 |
| _version_ | 1827747635289653248 |
|---|---|
| author | James Brotchie Wenchao Li Andrew D. Greentree Allison Kealy |
| author_facet | James Brotchie Wenchao Li Andrew D. Greentree Allison Kealy |
| author_sort | James Brotchie |
| collection | DOAJ |
| description | Inertial localisation is an important technique as it enables ego-motion estimation in conditions where external observers are unavailable. However, low-cost inertial sensors are inherently corrupted by bias and noise, which lead to unbound errors, making straight integration for position intractable. Traditional mathematical approaches are reliant on prior system knowledge, geometric theories and are constrained by predefined dynamics. Recent advances in deep learning, which benefit from ever-increasing volumes of data and computational power, allow for data-driven solutions that offer more comprehensive understanding. Existing deep inertial odometry solutions rely on estimating the latent states, such as velocity, or are dependent on fixed-sensor positions and periodic motion patterns. In this work, we propose taking the traditional state estimation recursive methodology and applying it in the deep learning domain. Our approach, which incorporates the true position priors in the training process, is trained on inertial measurements and ground truth displacement data, allowing recursion and learning both motion characteristics and systemic error bias and drift. We present two end-to-end frameworks for pose invariant deep inertial odometry that utilises self-attention to capture both spatial features and long-range dependencies in inertial data. We evaluate our approaches against a custom 2-layer Gated Recurrent Unit, trained in the same manner on the same data, and tested each approach on a number of different users, devices and activities. Each network had a sequence length weighted relative trajectory error mean <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mo>≤</mo><mn>0.4594</mn></mrow></semantics></math></inline-formula> m, highlighting the effectiveness of our learning process used in the development of the models. |
| first_indexed | 2024-03-11T05:55:33Z |
| format | Article |
| id | doaj.art-a2e69290bcbd4a7bb180339fb2cb83b3 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-11T05:55:33Z |
| publishDate | 2023-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-a2e69290bcbd4a7bb180339fb2cb83b32023-11-17T13:47:35ZengMDPI AGSensors1424-82202023-03-01236321710.3390/s23063217RIOT: Recursive Inertial Odometry Transformer for Localisation from Low-Cost IMU MeasurementsJames Brotchie0Wenchao Li1Andrew D. Greentree2Allison Kealy3School of Science, RMIT University, Melbourne, VIC 3001, AustraliaSchool of Science, RMIT University, Melbourne, VIC 3001, AustraliaARC Centre of Excellence for Nanoscale BioPhotonics, School of Science, RMIT University, Melbourne, VIC 3001, AustraliaSchool of Science, RMIT University, Melbourne, VIC 3001, AustraliaInertial localisation is an important technique as it enables ego-motion estimation in conditions where external observers are unavailable. However, low-cost inertial sensors are inherently corrupted by bias and noise, which lead to unbound errors, making straight integration for position intractable. Traditional mathematical approaches are reliant on prior system knowledge, geometric theories and are constrained by predefined dynamics. Recent advances in deep learning, which benefit from ever-increasing volumes of data and computational power, allow for data-driven solutions that offer more comprehensive understanding. Existing deep inertial odometry solutions rely on estimating the latent states, such as velocity, or are dependent on fixed-sensor positions and periodic motion patterns. In this work, we propose taking the traditional state estimation recursive methodology and applying it in the deep learning domain. Our approach, which incorporates the true position priors in the training process, is trained on inertial measurements and ground truth displacement data, allowing recursion and learning both motion characteristics and systemic error bias and drift. We present two end-to-end frameworks for pose invariant deep inertial odometry that utilises self-attention to capture both spatial features and long-range dependencies in inertial data. We evaluate our approaches against a custom 2-layer Gated Recurrent Unit, trained in the same manner on the same data, and tested each approach on a number of different users, devices and activities. Each network had a sequence length weighted relative trajectory error mean <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mo>≤</mo><mn>0.4594</mn></mrow></semantics></math></inline-formula> m, highlighting the effectiveness of our learning process used in the development of the models.https://www.mdpi.com/1424-8220/23/6/3217inertial navigationdeep learningsensor fusionodometrypose estimationtrajectory estimation |
| spellingShingle | James Brotchie Wenchao Li Andrew D. Greentree Allison Kealy RIOT: Recursive Inertial Odometry Transformer for Localisation from Low-Cost IMU Measurements Sensors inertial navigation deep learning sensor fusion odometry pose estimation trajectory estimation |
| title | RIOT: Recursive Inertial Odometry Transformer for Localisation from Low-Cost IMU Measurements |
| title_full | RIOT: Recursive Inertial Odometry Transformer for Localisation from Low-Cost IMU Measurements |
| title_fullStr | RIOT: Recursive Inertial Odometry Transformer for Localisation from Low-Cost IMU Measurements |
| title_full_unstemmed | RIOT: Recursive Inertial Odometry Transformer for Localisation from Low-Cost IMU Measurements |
| title_short | RIOT: Recursive Inertial Odometry Transformer for Localisation from Low-Cost IMU Measurements |
| title_sort | riot recursive inertial odometry transformer for localisation from low cost imu measurements |
| topic | inertial navigation deep learning sensor fusion odometry pose estimation trajectory estimation |
| url | https://www.mdpi.com/1424-8220/23/6/3217 |
| work_keys_str_mv | AT jamesbrotchie riotrecursiveinertialodometrytransformerforlocalisationfromlowcostimumeasurements AT wenchaoli riotrecursiveinertialodometrytransformerforlocalisationfromlowcostimumeasurements AT andrewdgreentree riotrecursiveinertialodometrytransformerforlocalisationfromlowcostimumeasurements AT allisonkealy riotrecursiveinertialodometrytransformerforlocalisationfromlowcostimumeasurements |